remap every 100000 values

existing written view

python/obitools3/obidms/_obidms.cfiles

@@ -1,5 +1,7 @@
 ../../../src/obidms.h
 ../../../src/obidms.c
+../../../src/obiview.h
+../../../src/obiview.c
 ../../../src/obidmscolumn.h
 ../../../src/obidmscolumn.c
 ../../../src/obidmscolumndir.h
@@ -12,5 +14,11 @@
 ../../../src/obitypes.c
 ../../../src/private_at_functions.h
 ../../../src/private_at_functions.c
-../../../src/obiarray.h
-../../../src/obiarray.c
+../../../src/obiavl.h
+../../../src/obiavl.c
+../../../src/encode.h
+../../../src/encode.c
+../../../src/obidmscolumn_idx.h
+../../../src/obidmscolumn_idx.c
+../../../src/obidms_taxonomy.c
+../../../src/obidms_taxonomy.h

python/obitools3/obidms/_obidms.pxd

@@ -2,10 +2,82 @@
 
 from .capi.obidms       cimport OBIDMS_p
 from .capi.obidmscolumn cimport OBIDMS_column_p
+from .capi.obiview      cimport Obiview_p
 from .capi.obitypes     cimport obiversion_t, OBIType_t, index_t
+from ._obitaxo          cimport OBI_Taxonomy
+
+cdef class OBIDMS_column:
+
+    cdef  OBIDMS_column_p*   pointer
+    cdef  OBIDMS             dms
+    cdef  Obiview_p          view
+    cdef  str                data_type
+    cdef  str                dms_name
+    cdef  str                column_name
+    cdef  index_t            nb_elements_per_line
+    cdef  list               elements_names
+        
+    cpdef         update_pointer(self)
+    cpdef list    get_elements_names(self)
+    cpdef str     get_data_type(self)
+    cpdef index_t get_nb_lines_used(self)
+    cpdef str     get_creation_date(self)
+    cpdef str     get_comments(self)
+    cpdef         close(self)
+    @staticmethod
+    cdef  object  get_subclass_type(OBIDMS_column_p column_p)
 
 
-cdef class OBIDMS_column
+cdef class OBIDMS_column_multi_elts(OBIDMS_column):
+    
+    cpdef set_line(self, index_t line_nb, dict values)
+
+
+cdef class OBIDMS_column_line:
+
+    cdef OBIDMS_column column
+    cdef index_t index
+
+
+cdef class OBIView:
+    
+    cdef Obiview_p pointer
+    cdef str       name
+    cdef str       comments
+    cdef dict      columns
+    cdef dict      columns_pp    # TODO this dict might be unnecessary
+    cdef OBIDMS    dms
+    
+    cpdef delete_column(self, str column_name)
+    cpdef add_column(self, 
+                     str column_name, 
+                     obiversion_t version_number=*, 
+                     str type=*,
+                     index_t nb_lines=*, 
+                     index_t nb_elements_per_line=*,
+                     list elements_names=*,
+                     str avl_name=*,
+                     str comments=*,
+                     bint create=*
+                    )
+    cpdef select_line(self, index_t line_nb)
+    cpdef select_lines(self, list line_selection)
+    cpdef save_and_close(self)
+
+
+cdef class OBIView_NUC_SEQS(OBIView):
+
+    cdef OBIDMS_column ids
+    cdef OBIDMS_column sequences
+    cdef OBIDMS_column descriptions
+    
+    cpdef delete_column(self, str column_name)
+    
+
+cdef class OBIView_line :
+
+    cdef index_t index
+    cdef OBIView view
 
 
 cdef class OBIDMS:
@@ -13,35 +85,10 @@ cdef class OBIDMS:
     cdef OBIDMS_p pointer
     cdef str dms_name
     
-    cpdef dict list(self)
     cpdef close(self)
-    cpdef OBIDMS_column open_column(self,
-                                    str column_name,
-                                    bint create=*, 
-                                    bint clone=*, bint clone_data=*,
-                                    obiversion_t version_number=*, 
-                                    OBIType_t data_type=*,
-                                    index_t nb_lines=*, 
-                                    index_t nb_elements_per_line=*, 
-                                    list elements_names=*,
-                                    str array_name=*)
-
-
-cdef class OBIDMS_column:
-
-    cdef OBIDMS_column_p    pointer
-    cdef OBIDMS             dms
-    cdef str                data_type   # TODO keep as OBIType_t? both?
-    cdef str                dms_name
-    cdef str                column_name
-    cdef index_t             nb_elements_per_line
-    cdef list               elements_names
-
-#    cpdef object get_item(self, index_t line_nb, str element_name)                  TODO can't declare because not the same in all subclasses
-#    cpdef        set_item(self, index_t line_nb, str element_name, object value)    TODO can't declare because object value
-    cpdef list    get_elements_names(self)
-    cpdef str     get_data_type(self)
-    cpdef index_t get_nb_lines_used(self)
-    cpdef str     get_creation_date(self)
-    cpdef         close(self)
+    cpdef OBI_Taxonomy open_taxonomy(self, str taxo_name)
+    cpdef OBIView open_view(self, str view_name)
+    cpdef OBIView new_view(self, str view_name, object view_to_clone=*, list line_selection=*, str view_type=*, str comments=*)
+    cpdef dict read_view_infos(self, str view_name)
+    cpdef dict read_views(self)
 

python/obitools3/obidms/_obidmscolumn_bool.cfiles

@@ -1,5 +1,7 @@
 ../../../src/obidmscolumn_bool.c
 ../../../src/obidmscolumn_bool.h
+../../../src/obiview.h
+../../../src/obiview.c
 ../../../src/obidmscolumn.h
 ../../../src/obidmscolumn.c
 ../../../src/obidmscolumndir.h
@@ -14,5 +16,5 @@
 ../../../src/obitypes.c
 ../../../src/private_at_functions.h
 ../../../src/private_at_functions.c
-../../../src/obiarray.h
-../../../src/obiarray.c
+../../../src/obiavl.h
+../../../src/obiavl.c

python/obitools3/obidms/_obidmscolumn_bool.pxd

@@ -1,25 +1,14 @@
 #cython: language_level=3
 
-from .capi.obitypes cimport obibool_t, index_t
-from ._obidms       cimport OBIDMS_column
+from .capi.obitypes cimport index_t
+from ._obidms       cimport OBIDMS_column, OBIDMS_column_multi_elts
 
 
 cdef class OBIDMS_column_bool(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb)
     cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
 
-cdef class OBIDMS_column_bool_writable(OBIDMS_column_bool):     
-    cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
-
-cdef class OBIDMS_column_bool_multi_elts(OBIDMS_column_bool):
+cdef class OBIDMS_column_multi_elts_bool(OBIDMS_column_multi_elts):
     cpdef object get_item(self, index_t line_nb, str element_name)
     cpdef object get_line(self, index_t line_nb)
-    cpdef set_item(self, index_t line_nb, str element_name, obibool_t value)
-    cpdef set_line(self, index_t line_nb, object values)
-
-cdef class OBIDMS_column_bool_multi_elts_writable(OBIDMS_column_bool_multi_elts):
-    cpdef set_item(self, index_t line_nb, str element_name, obibool_t value)
-    cpdef set_line(self, index_t line_nb, object values)
-    cpdef close(self)
+    cpdef set_item(self, index_t line_nb, str element_name, object value)

python/obitools3/obidms/_obidmscolumn_bool.pyx

@@ -1,13 +1,12 @@
 #cython: language_level=3
  
-from .capi.obidmscolumn cimport obi_close_column,\
-                                obi_truncate_and_close_column, \
-                                obi_column_get_obibool_with_elt_name, \
-                                obi_column_get_obibool_with_elt_idx, \
-                                obi_column_set_obibool_with_elt_name, \
-                                obi_column_set_obibool_with_elt_idx
+from .capi.obidmscolumn cimport obi_truncate_and_close_column, \
+                                obi_column_get_obibool_with_elt_name_in_view, \
+                                obi_column_get_obibool_with_elt_idx_in_view, \
+                                obi_column_set_obibool_with_elt_name_in_view, \
+                                obi_column_set_obibool_with_elt_idx_in_view
 from .capi.obierrno     cimport obi_errno
-from .capi.obitypes     cimport OBIBool_NA
+from .capi.obitypes     cimport OBIBool_NA, obibool_t
 
 from obitools3.utils cimport str2bytes
 
@@ -19,7 +18,7 @@ cdef class OBIDMS_column_bool(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb):
         cdef obibool_t value
         cdef object result
-        value = obi_column_get_obibool_with_elt_idx(self.pointer, line_nb, 0)
+        value = obi_column_get_obibool_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0)
         if obi_errno > 0 :
             raise IndexError(line_nb)
         if value == OBIBool_NA :
@@ -29,30 +28,18 @@ cdef class OBIDMS_column_bool(OBIDMS_column):
         return result
     
     cpdef set_line(self, index_t line_nb, object value):
-        raise Exception("Column is read-only")
-    
-    cpdef close(self):
-        if obi_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-
-
-cdef class OBIDMS_column_bool_writable(OBIDMS_column_bool):     
-
-    cpdef set_line(self, index_t line_nb, object value):
-        if obi_column_set_obibool_with_elt_idx(self.pointer, line_nb, 0, <obibool_t> value) < 0:
+        if value is None :
+            value = OBIBool_NA
+        if obi_column_set_obibool_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0, <obibool_t> value) < 0:
             raise Exception("Problem setting a value in a column")
 
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
 
-
-cdef class OBIDMS_column_bool_multi_elts(OBIDMS_column_bool):
+cdef class OBIDMS_column_multi_elts_bool(OBIDMS_column_multi_elts):
 
     cpdef object get_item(self, index_t line_nb, str element_name):
         cdef obibool_t value
         cdef object result
-        value = obi_column_get_obibool_with_elt_name(self.pointer, line_nb, str2bytes(element_name))
+        value = obi_column_get_obibool_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name))
         if obi_errno > 0 :
             raise IndexError(line_nb, element_name)
         if value == OBIBool_NA :
@@ -63,42 +50,29 @@ cdef class OBIDMS_column_bool_multi_elts(OBIDMS_column_bool):
     
     cpdef object get_line(self, index_t line_nb) :
         cdef obibool_t value
-        cdef object result
+        cdef object value_in_result
+        cdef dict result
         cdef index_t i
         cdef bint all_NA
         result = {}
         all_NA = True
         for i in range(self.nb_elements_per_line) :
-            value = obi_column_get_obibool_with_elt_idx(self.pointer, line_nb, i)
+            value = obi_column_get_obibool_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, i)
             if obi_errno > 0 :
                 raise IndexError(line_nb)
-            result[self.elements_names[i]] = PyBool_FromLong(value)
-            if all_NA and (value != OBIBool_NA) :
+            if value == OBIBool_NA :
+                value_in_result = None
+            else :
+                value_in_result = PyBool_FromLong(value)   
+            result[self.elements_names[i]] = value_in_result
+            if all_NA and (value_in_result is not None) :
                 all_NA = False
         if all_NA :
             result = None
         return result
 
-    cpdef set_item(self, index_t line_nb, str element_name, obibool_t value):
-        raise Exception("Column is read-only")
-
-    cpdef set_line(self, index_t line_nb, object values):
-        raise Exception("Column is read-only")
-
-
-cdef class OBIDMS_column_bool_multi_elts_writable(OBIDMS_column_bool_multi_elts):     
-
-    cpdef set_item(self, index_t line_nb, str element_name, obibool_t value):
-        if obi_column_set_obibool_with_elt_name(self.pointer, line_nb, str2bytes(element_name), value) < 0:
+    cpdef set_item(self, index_t line_nb, str element_name, object value):
+        if value is None :
+            value = OBIBool_NA
+        if obi_column_set_obibool_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name), <obibool_t> value) < 0:
             raise Exception("Problem setting a value in a column")
-    
-    cpdef set_line(self, index_t line_nb, object values):
-        cdef obibool_t value
-        for element_name in values :
-            value = <obibool_t> values[element_name]
-            self.set_item(line_nb, element_name, value)
-
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-

python/obitools3/obidms/_obidmscolumn_char.cfiles

@@ -1,5 +1,7 @@
 ../../../src/obidmscolumn_char.c
 ../../../src/obidmscolumn_char.h
+../../../src/obiview.h
+../../../src/obiview.c
 ../../../src/obidmscolumn.h
 ../../../src/obidmscolumn.c
 ../../../src/obidmscolumndir.h
@@ -14,5 +16,5 @@
 ../../../src/obitypes.c
 ../../../src/private_at_functions.h
 ../../../src/private_at_functions.c
-../../../src/obiarray.h
-../../../src/obiarray.c
+../../../src/obiavl.h
+../../../src/obiavl.c

python/obitools3/obidms/_obidmscolumn_char.pxd

@@ -1,25 +1,14 @@
 #cython: language_level=3
 
-from .capi.obitypes cimport obichar_t, index_t
-from ._obidms       cimport OBIDMS_column
+from .capi.obitypes cimport index_t
+from ._obidms       cimport OBIDMS_column, OBIDMS_column_multi_elts
 
 
 cdef class OBIDMS_column_char(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb)
     cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
 
-cdef class OBIDMS_column_char_writable(OBIDMS_column_char):     
-    cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
-
-cdef class OBIDMS_column_char_multi_elts(OBIDMS_column_char):
+cdef class OBIDMS_column_multi_elts_char(OBIDMS_column_multi_elts):
     cpdef object get_item(self, index_t line_nb, str element_name)
     cpdef object get_line(self, index_t line_nb)
-    cpdef set_item(self, index_t line_nb, str element_name, bytes value)
-    cpdef set_line(self, index_t line_nb, object values)
-
-cdef class OBIDMS_column_char_multi_elts_writable(OBIDMS_column_char_multi_elts):
-    cpdef set_item(self, index_t line_nb, str element_name, bytes value)
-    cpdef set_line(self, index_t line_nb, object values)
-    cpdef close(self)
+    cpdef set_item(self, index_t line_nb, str element_name, object value)

python/obitools3/obidms/_obidmscolumn_char.pyx

@@ -1,15 +1,14 @@
 #cython: language_level=3
  
-from .capi.obidmscolumn cimport obi_close_column,\
-                                obi_truncate_and_close_column, \
-                                obi_column_get_obichar_with_elt_name, \
-                                obi_column_get_obichar_with_elt_idx, \
-                                obi_column_set_obichar_with_elt_name, \
-                                obi_column_set_obichar_with_elt_idx
+from .capi.obidmscolumn cimport obi_truncate_and_close_column, \
+                                obi_column_get_obichar_with_elt_name_in_view, \
+                                obi_column_get_obichar_with_elt_idx_in_view, \
+                                obi_column_set_obichar_with_elt_name_in_view, \
+                                obi_column_set_obichar_with_elt_idx_in_view
 from .capi.obierrno     cimport obi_errno
-from .capi.obitypes     cimport OBIChar_NA
+from .capi.obitypes     cimport OBIChar_NA, obichar_t
 
-from obitools3.utils cimport str2bytes
+from obitools3.utils cimport str2bytes, bytes2str
 
 
 cdef class OBIDMS_column_char(OBIDMS_column):
@@ -17,122 +16,62 @@ cdef class OBIDMS_column_char(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb):
         cdef obichar_t value
         cdef object result
-        value = obi_column_get_obichar_with_elt_idx(self.pointer, line_nb, 0)
+        value = obi_column_get_obichar_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0)
         if obi_errno > 0 :
             raise IndexError(line_nb)
         if value == OBIChar_NA :
             result = None
         else :
-            result = <bytes> value
+            result = bytes2str(value)
         return result
 
     cpdef set_line(self, index_t line_nb, object value):
-        raise Exception("Column is read-only")
-    
-    cpdef close(self):
-        if obi_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-
-
-cdef class OBIDMS_column_char_writable(OBIDMS_column_char):     
-
-    cpdef set_line(self, index_t line_nb, object value):
-        if obi_column_set_obichar_with_elt_idx(self.pointer, line_nb, 0, <bytes> value[0]) < 0:
+        if value is None :
+            value = OBIChar_NA
+        if obi_column_set_obichar_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0, str2bytes(value)[0]) < 0:
             raise Exception("Problem setting a value in a column")
 
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
 
-
-cdef class OBIDMS_column_char_multi_elts(OBIDMS_column_char):
+cdef class OBIDMS_column_multi_elts_char(OBIDMS_column_multi_elts):
 
     cpdef object get_item(self, index_t line_nb, str element_name):
         cdef obichar_t value
         cdef object result
-        value = obi_column_get_obichar_with_elt_name(self.pointer, line_nb, str2bytes(element_name))
+        value = obi_column_get_obichar_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name))
         if obi_errno > 0 :
             raise IndexError(line_nb, element_name)
         if value == OBIChar_NA :
             result = None
         else :
-            result = <bytes> value
+            result = bytes2str(value)
         return result
 
     cpdef object get_line(self, index_t line_nb) :
         cdef obichar_t value
-        cdef object result
+        cdef object value_in_result
+        cdef dict result
         cdef index_t i
         cdef bint all_NA
         result = {}
         all_NA = True
         for i in range(self.nb_elements_per_line) :
-            value = obi_column_get_obichar_with_elt_idx(self.pointer, line_nb, i)
+            value = obi_column_get_obichar_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, i)
             if obi_errno > 0 :
                 raise IndexError(line_nb)
-            result[self.elements_names[i]] = <bytes> value
-            if all_NA and (value != OBIChar_NA) :
+            if value == OBIChar_NA :
+                value_in_result = None
+            else :
+                value_in_result = bytes2str(value)   
+            result[self.elements_names[i]] = value_in_result
+            if all_NA and (value_in_result is not None) :
                 all_NA = False
         if all_NA :
             result = None
         return result
 
-    cpdef set_item(self, index_t line_nb, str element_name, bytes value):
-        raise Exception("Column is read-only")
-
-    cpdef set_line(self, index_t line_nb, object values):
-        raise Exception("Column is read-only")
-
-
-cdef class OBIDMS_column_char_multi_elts_writable(OBIDMS_column_char_multi_elts):     
-
-    cpdef set_item(self, index_t line_nb, str element_name, bytes value):
-        if obi_column_set_obichar_with_elt_name(self.pointer, line_nb, str2bytes(element_name), value[0]) < 0:
+    cpdef set_item(self, index_t line_nb, str element_name, object value):
+        if value is None :
+            value = OBIChar_NA
+        if obi_column_set_obichar_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name), str2bytes(value)[0]) < 0:
             raise Exception("Problem setting a value in a column")
 
-    cpdef set_line(self, index_t line_nb, object values):
-        cdef bytes value
-        for element_name in values :
-            value = <bytes> values[element_name]
-            self.set_item(line_nb, element_name, value)
-
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-
-
-
-
-
-# cdef class OBIDMS_column_char(OBIDMS_column) :
-#  
-#     cpdef object get_item(self, index_t line_nb, str element_name):
-#         cdef char value
-#         cdef object result
-#         value = obi_column_get_obichar_with_elt_name(self.pointer, line_nb, str2bytes(element_name))
-#         if obi_errno > 0 :
-#             raise IndexError(line_nb, element_name)
-#         if value == OBIChar_NA :
-#             result = None
-#         else :
-#             result = <bytes> value
-#         return result
-# 
-#     cpdef set_item(self, index_t line_nb, str element_name, bytes value):
-#         raise Exception("Column is read-only")
-#  
-#     cpdef close(self):
-#         if obi_close_column(self.pointer) < 0 :
-#             raise Exception("Problem closing a column")
-#  
-#          
-# cdef class OBIDMS_column_char_writable(OBIDMS_column_char) :     
-#  
-#     cpdef set_item(self, index_t line_nb, str element_name, bytes value):
-#         if obi_column_set_obichar_with_elt_name(self.pointer, line_nb, str2bytes(element_name), value[0]) < 0:
-#             raise Exception("Problem setting a value in a column")
-#  
-#     cpdef close(self):
-#         if obi_truncate_and_close_column(self.pointer) < 0 :
-#             raise Exception("Problem closing a column")
-#       

python/obitools3/obidms/_obidmscolumn_float.cfiles

@@ -1,5 +1,7 @@
 ../../../src/obidmscolumn_float.c
 ../../../src/obidmscolumn_float.h
+../../../src/obiview.h
+../../../src/obiview.c
 ../../../src/obidmscolumn.h
 ../../../src/obidmscolumn.c
 ../../../src/obidmscolumndir.h
@@ -14,5 +16,5 @@
 ../../../src/obitypes.c
 ../../../src/private_at_functions.h
 ../../../src/private_at_functions.c
-../../../src/obiarray.h
-../../../src/obiarray.c
+../../../src/obiavl.h
+../../../src/obiavl.c

python/obitools3/obidms/_obidmscolumn_float.pxd

@@ -1,25 +1,14 @@
 #cython: language_level=3
 
-from .capi.obitypes cimport obifloat_t, index_t
-from ._obidms       cimport OBIDMS_column
+from .capi.obitypes cimport index_t
+from ._obidms       cimport OBIDMS_column, OBIDMS_column_multi_elts
 
 
 cdef class OBIDMS_column_float(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb)
     cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
 
-cdef class OBIDMS_column_float_writable(OBIDMS_column_float):     
-    cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
-
-cdef class OBIDMS_column_float_multi_elts(OBIDMS_column_float):
+cdef class OBIDMS_column_multi_elts_float(OBIDMS_column_multi_elts):
     cpdef object get_item(self, index_t line_nb, str element_name)
     cpdef object get_line(self, index_t line_nb)
-    cpdef set_item(self, index_t line_nb, str element_name, obifloat_t value)
-    cpdef set_line(self, index_t line_nb, object values)
-
-cdef class OBIDMS_column_float_multi_elts_writable(OBIDMS_column_float_multi_elts):
-    cpdef set_item(self, index_t line_nb, str element_name, obifloat_t value)
-    cpdef set_line(self, index_t line_nb, object values)
-    cpdef close(self)
+    cpdef set_item(self, index_t line_nb, str element_name, object value)

python/obitools3/obidms/_obidmscolumn_float.pyx

@@ -1,13 +1,12 @@
 #cython: language_level=3
 
-from .capi.obidmscolumn cimport obi_close_column,\
-                                obi_truncate_and_close_column, \
-                                obi_column_get_obifloat_with_elt_name, \
-                                obi_column_get_obifloat_with_elt_idx, \
-                                obi_column_set_obifloat_with_elt_name, \
-                                obi_column_set_obifloat_with_elt_idx
+from .capi.obidmscolumn cimport obi_truncate_and_close_column,\
+                                obi_column_get_obifloat_with_elt_name_in_view, \
+                                obi_column_get_obifloat_with_elt_idx_in_view, \
+                                obi_column_set_obifloat_with_elt_name_in_view, \
+                                obi_column_set_obifloat_with_elt_idx_in_view
 from .capi.obierrno     cimport obi_errno
-from .capi.obitypes     cimport OBIFloat_NA
+from .capi.obitypes     cimport OBIFloat_NA, obifloat_t
 
 from obitools3.utils cimport str2bytes
 
@@ -17,7 +16,7 @@ cdef class OBIDMS_column_float(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb):
         cdef obifloat_t value
         cdef object result
-        value = obi_column_get_obifloat_with_elt_idx(self.pointer, line_nb, 0)
+        value = obi_column_get_obifloat_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0)
         if obi_errno > 0 :
             raise IndexError(line_nb)
         if value == OBIFloat_NA :
@@ -27,30 +26,18 @@ cdef class OBIDMS_column_float(OBIDMS_column):
         return result
 
     cpdef set_line(self, index_t line_nb, object value):
-        raise Exception("Column is read-only")
-    
-    cpdef close(self):
-        if obi_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-
-
-cdef class OBIDMS_column_float_writable(OBIDMS_column_float):     
-
-    cpdef set_line(self, index_t line_nb, object value):
-        if obi_column_set_obifloat_with_elt_idx(self.pointer, line_nb, 0, <obifloat_t> value) < 0:
+        if value is None :
+            value = OBIFloat_NA
+        if obi_column_set_obifloat_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0, <obifloat_t> value) < 0:
             raise Exception("Problem setting a value in a column")
 
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
 
-
-cdef class OBIDMS_column_float_multi_elts(OBIDMS_column_float):
+cdef class OBIDMS_column_multi_elts_float(OBIDMS_column_multi_elts):
 
     cpdef object get_item(self, index_t line_nb, str element_name):
         cdef obifloat_t value
         cdef object result
-        value = obi_column_get_obifloat_with_elt_name(self.pointer, line_nb, str2bytes(element_name))
+        value = obi_column_get_obifloat_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name))
         if obi_errno > 0 :
             raise IndexError(line_nb, element_name)
         if value == OBIFloat_NA :
@@ -61,43 +48,30 @@ cdef class OBIDMS_column_float_multi_elts(OBIDMS_column_float):
 
     cpdef object get_line(self, index_t line_nb) :
         cdef obifloat_t value
-        cdef object result
+        cdef object value_in_result
+        cdef dict result
         cdef index_t i
         cdef bint all_NA
         result = {}
         all_NA = True
         for i in range(self.nb_elements_per_line) :
-            value = obi_column_get_obifloat_with_elt_idx(self.pointer, line_nb, i)
+            value = obi_column_get_obifloat_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, i)
             if obi_errno > 0 :
                 raise IndexError(line_nb)
-            result[self.elements_names[i]] = <double> value
-            if all_NA and (value != OBIFloat_NA) :
+            if value == OBIFloat_NA :
+                value_in_result = None
+            else :
+                value_in_result = <double> value
+            result[self.elements_names[i]] = value_in_result
+            if all_NA and (value_in_result is not None) :
                 all_NA = False
         if all_NA :
             result = None
         return result
    
-    cpdef set_item(self, index_t line_nb, str element_name, obifloat_t value):
-        raise Exception("Column is read-only")
-
-    cpdef set_line(self, index_t line_nb, object values):
-        raise Exception("Column is read-only")
-
-
-cdef class OBIDMS_column_float_multi_elts_writable(OBIDMS_column_float_multi_elts):     
-
-    cpdef set_item(self, index_t line_nb, str element_name, obifloat_t value):
-        if obi_column_set_obifloat_with_elt_name(self.pointer, line_nb, str2bytes(element_name), value) < 0:
+    cpdef set_item(self, index_t line_nb, str element_name, object value):
+        if value is None :
+            value = OBIFloat_NA
+        if obi_column_set_obifloat_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name), <obifloat_t> value) < 0:
             raise Exception("Problem setting a value in a column")
  
-    cpdef set_line(self, index_t line_nb, object values):
-        cdef obifloat_t value
-        for element_name in values :
-            value = <obifloat_t> values[element_name]
-            self.set_item(line_nb, element_name, value)
-
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-
- 

python/obitools3/obidms/_obidmscolumn_int.cfiles

@@ -1,5 +1,7 @@
 ../../../src/obidmscolumn_int.c
 ../../../src/obidmscolumn_int.h
+../../../src/obiview.h
+../../../src/obiview.c
 ../../../src/obidmscolumn.h
 ../../../src/obidmscolumn.c
 ../../../src/obidmscolumndir.h
@@ -14,5 +16,5 @@
 ../../../src/obitypes.c
 ../../../src/private_at_functions.h
 ../../../src/private_at_functions.c
-../../../src/obiarray.h
-../../../src/obiarray.c
+../../../src/obiavl.h
+../../../src/obiavl.c

python/obitools3/obidms/_obidmscolumn_int.pxd

@@ -1,25 +1,14 @@
 #cython: language_level=3
 
-from .capi.obitypes cimport obiint_t, index_t
-from ._obidms       cimport OBIDMS_column
+from .capi.obitypes cimport index_t
+from ._obidms       cimport OBIDMS_column, OBIDMS_column_multi_elts
 
 
 cdef class OBIDMS_column_int(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb)
     cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
 
-cdef class OBIDMS_column_int_writable(OBIDMS_column_int):     
-    cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
-
-cdef class OBIDMS_column_int_multi_elts(OBIDMS_column_int):
+cdef class OBIDMS_column_multi_elts_int(OBIDMS_column_multi_elts):
     cpdef object get_item(self, index_t line_nb, str element_name)
     cpdef object get_line(self, index_t line_nb)
-    cpdef set_item(self, index_t line_nb, str element_name, obiint_t value)
-    cpdef set_line(self, index_t line_nb, object values)
-
-cdef class OBIDMS_column_int_multi_elts_writable(OBIDMS_column_int_multi_elts):
-    cpdef set_item(self, index_t line_nb, str element_name, obiint_t value)
-    cpdef set_line(self, index_t line_nb, object values)
-    cpdef close(self)
+    cpdef set_item(self, index_t line_nb, str element_name, object value)

python/obitools3/obidms/_obidmscolumn_int.pyx

@@ -1,13 +1,12 @@
 #cython: language_level=3
 
-from .capi.obidmscolumn cimport obi_close_column,\
-                                obi_truncate_and_close_column, \
-                                obi_column_get_obiint_with_elt_name, \
-                                obi_column_get_obiint_with_elt_idx, \
-                                obi_column_set_obiint_with_elt_name, \
-                                obi_column_set_obiint_with_elt_idx
+from .capi.obidmscolumn cimport obi_truncate_and_close_column, \
+                                obi_column_get_obiint_with_elt_name_in_view, \
+                                obi_column_get_obiint_with_elt_idx_in_view, \
+                                obi_column_set_obiint_with_elt_name_in_view, \
+                                obi_column_set_obiint_with_elt_idx_in_view
 from .capi.obierrno     cimport obi_errno
-from .capi.obitypes     cimport OBIInt_NA
+from .capi.obitypes     cimport OBIInt_NA, obiint_t
 
 from obitools3.utils cimport str2bytes
 
@@ -19,7 +18,7 @@ cdef class OBIDMS_column_int(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb):
         cdef obiint_t value
         cdef object result
-        value = obi_column_get_obiint_with_elt_idx(self.pointer, line_nb, 0)
+        value = obi_column_get_obiint_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0)
         if obi_errno > 0 :
             raise IndexError(line_nb)
         if value == OBIInt_NA :
@@ -29,30 +28,18 @@ cdef class OBIDMS_column_int(OBIDMS_column):
         return result
             
     cpdef set_line(self, index_t line_nb, object value):
-        raise Exception("Column is read-only")
-    
-    cpdef close(self):
-        if obi_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-
-
-cdef class OBIDMS_column_int_writable(OBIDMS_column_int):     
-
-    cpdef set_line(self, index_t line_nb, object value):
-        if obi_column_set_obiint_with_elt_idx(self.pointer, line_nb, 0, <obiint_t> value) < 0:
+        if value is None :
+            value = OBIInt_NA
+        if obi_column_set_obiint_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0, <obiint_t> value) < 0:
             raise Exception("Problem setting a value in a column")
 
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
 
-
-cdef class OBIDMS_column_int_multi_elts(OBIDMS_column_int):
+cdef class OBIDMS_column_multi_elts_int(OBIDMS_column_multi_elts):
 
     cpdef object get_item(self, index_t line_nb, str element_name):
         cdef obiint_t value
         cdef object result
-        value = obi_column_get_obiint_with_elt_name(self.pointer, line_nb, str2bytes(element_name))
+        value = obi_column_get_obiint_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name))
         if obi_errno > 0 :
             raise IndexError(line_nb, element_name)
         if value == OBIInt_NA :
@@ -63,42 +50,30 @@ cdef class OBIDMS_column_int_multi_elts(OBIDMS_column_int):
     
     cpdef object get_line(self, index_t line_nb) :
         cdef obiint_t value
-        cdef object result
+        cdef object value_in_result
+        cdef dict result
         cdef index_t i
         cdef bint all_NA
         result = {}
         all_NA = True
         for i in range(self.nb_elements_per_line) :
-            value = obi_column_get_obiint_with_elt_idx(self.pointer, line_nb, i)
+            value = obi_column_get_obiint_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, i)
             if obi_errno > 0 :
                 raise IndexError(line_nb)
-            result[self.elements_names[i]] = PyInt_FromLong(value)
-            if all_NA and (value != OBIInt_NA) :
+            if value == OBIInt_NA :
+                value_in_result = None
+            else :
+                value_in_result = PyInt_FromLong(value)
+            result[self.elements_names[i]] = value_in_result
+            if all_NA and (value_in_result is not None) :
                 all_NA = False
         if all_NA :
-            result = None
+            result = None   # TODO discuss
         return result
 
-    cpdef set_item(self, index_t line_nb, str element_name, obiint_t value):
-        raise Exception("Column is read-only")
-
-    cpdef set_line(self, index_t line_nb, object values):
-        raise Exception("Column is read-only")
-
-
-cdef class OBIDMS_column_int_multi_elts_writable(OBIDMS_column_int_multi_elts):     
-
-    cpdef set_item(self, index_t line_nb, str element_name, obiint_t value):
-        if obi_column_set_obiint_with_elt_name(self.pointer, line_nb, str2bytes(element_name), value) < 0:
+    cpdef set_item(self, index_t line_nb, str element_name, object value):
+        if value is None :
+            value = OBIInt_NA
+        if obi_column_set_obiint_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name), <obiint_t> value) < 0:
             raise Exception("Problem setting a value in a column")
 
-    cpdef set_line(self, index_t line_nb, object values):
-        cdef obiint_t value
-        for element_name in values :
-            value = <obiint_t> values[element_name]
-            self.set_item(line_nb, element_name, value)
-
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-

python/obitools3/obidms/_obidmscolumn_seq.cfiles

@@ -0,0 +1,20 @@
+../../../src/obidmscolumn_seq.c
+../../../src/obidmscolumn_seq.h
+../../../src/obiview.h
+../../../src/obiview.c
+../../../src/obidmscolumn.h
+../../../src/obidmscolumn.c
+../../../src/obidmscolumndir.h
+../../../src/obidmscolumndir.c
+../../../src/obidms.h
+../../../src/obidms.c
+../../../src/obierrno.h
+../../../src/obierrno.c
+../../../src/obilittlebigman.h
+../../../src/obilittlebigman.c
+../../../src/obitypes.h
+../../../src/obitypes.c
+../../../src/private_at_functions.h
+../../../src/private_at_functions.c
+../../../src/obiavl.h
+../../../src/obiavl.c

python/obitools3/obidms/_obidmscolumn_seq.pxd

@@ -0,0 +1,14 @@
+#cython: language_level=3
+
+from .capi.obitypes cimport index_t
+from ._obidms       cimport OBIDMS_column, OBIDMS_column_multi_elts
+
+
+cdef class OBIDMS_column_seq(OBIDMS_column):
+    cpdef object get_line(self, index_t line_nb)
+    cpdef set_line(self, index_t line_nb, object value)
+
+cdef class OBIDMS_column_multi_elts_seq(OBIDMS_column_multi_elts):
+    cpdef object get_item(self, index_t line_nb, str element_name)
+    cpdef object get_line(self, index_t line_nb)
+    cpdef set_item(self, index_t line_nb, str element_name, object value)

python/obitools3/obidms/_obidmscolumn_seq.pyx

@@ -0,0 +1,83 @@
+#cython: language_level=3
+ 
+from .capi.obidmscolumn cimport obi_truncate_and_close_column, \
+                                obi_column_get_obiseq_with_elt_name_in_view, \
+                                obi_column_get_obiseq_with_elt_idx_in_view, \
+                                obi_column_set_obiseq_with_elt_name_in_view, \
+                                obi_column_set_obiseq_with_elt_idx_in_view
+from .capi.obierrno     cimport obi_errno
+from .capi.obitypes     cimport OBISeq_NA, const_char_p
+
+from obitools3.utils cimport str2bytes, bytes2str
+
+
+cdef class OBIDMS_column_seq(OBIDMS_column):
+
+    cpdef object get_line(self, index_t line_nb):
+        cdef bytes value
+        cdef object result
+        value = <bytes> obi_column_get_obiseq_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0)
+        if obi_errno > 0 :
+            raise IndexError(line_nb)
+        if value == OBISeq_NA :     # TODO
+            result = None
+        else :
+            result = bytes2str(value)
+        return result
+
+    cpdef set_line(self, index_t line_nb, object value):
+        cdef bytes value_b
+        if value is None :
+            value_b = OBISeq_NA
+        else :
+            value_b = str2bytes(value)
+        if obi_column_set_obiseq_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0, value_b) < 0:
+            raise Exception("Problem setting a value in a column")
+
+
+cdef class OBIDMS_column_multi_elts_seq(OBIDMS_column_multi_elts):
+
+    cpdef object get_item(self, index_t line_nb, str element_name):
+        cdef bytes value
+        cdef object result
+        value = <bytes> obi_column_get_obiseq_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name))
+        if obi_errno > 0 :
+            raise IndexError(line_nb, element_name)
+        if value == OBISeq_NA :
+            result = None
+        else :
+            result = bytes2str(value)
+        return result
+
+    cpdef object get_line(self, index_t line_nb) :
+        cdef bytes value
+        cdef object value_in_result
+        cdef dict result
+        cdef index_t i
+        cdef bint all_NA
+        result = {}
+        all_NA = True
+        for i in range(self.nb_elements_per_line) :
+            value = <bytes> obi_column_get_obiseq_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, i)
+            if obi_errno > 0 :
+                raise IndexError(line_nb)
+            if value == OBISeq_NA :
+                value_in_result = None
+            else :
+                value_in_result = bytes2str(value)    
+            result[self.elements_names[i]] = value_in_result
+            if all_NA and (value_in_result is not None) :
+                all_NA = False
+        if all_NA :
+            result = None
+        return result
+
+    cpdef set_item(self, index_t line_nb, str element_name, object value):
+        cdef bytes value_b
+        if value is None :
+            value_b = OBISeq_NA
+        else :
+            value_b = str2bytes(value)
+        if obi_column_set_obiseq_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name), value_b) < 0:
+            raise Exception("Problem setting a value in a column")
+    

python/obitools3/obidms/_obidmscolumn_str.cfiles

@@ -1,5 +1,7 @@
 ../../../src/obidmscolumn_str.c
 ../../../src/obidmscolumn_str.h
+../../../src/obiview.h
+../../../src/obiview.c
 ../../../src/obidmscolumn.h
 ../../../src/obidmscolumn.c
 ../../../src/obidmscolumndir.h
@@ -14,5 +16,5 @@
 ../../../src/obitypes.c
 ../../../src/private_at_functions.h
 ../../../src/private_at_functions.c
-../../../src/obiarray.h
-../../../src/obiarray.c
+../../../src/obiavl.h
+../../../src/obiavl.c

python/obitools3/obidms/_obidmscolumn_str.pxd

@@ -1,25 +1,14 @@
 #cython: language_level=3
 
 from .capi.obitypes cimport index_t
-from ._obidms       cimport OBIDMS_column
+from ._obidms       cimport OBIDMS_column, OBIDMS_column_multi_elts
 
 
 cdef class OBIDMS_column_str(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb)
     cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
 
-cdef class OBIDMS_column_str_writable(OBIDMS_column_str):     
-    cpdef set_line(self, index_t line_nb, object value)
-    cpdef close(self)
-
-cdef class OBIDMS_column_str_multi_elts(OBIDMS_column_str):
+cdef class OBIDMS_column_multi_elts_str(OBIDMS_column_multi_elts):
     cpdef object get_item(self, index_t line_nb, str element_name)
     cpdef object get_line(self, index_t line_nb)
-    cpdef set_item(self, index_t line_nb, str element_name, str value)
-    cpdef set_line(self, index_t line_nb, object values)
-
-cdef class OBIDMS_column_str_multi_elts_writable(OBIDMS_column_str_multi_elts):
-    cpdef set_item(self, index_t line_nb, str element_name, str value)
-    cpdef set_line(self, index_t line_nb, object values)
-    cpdef close(self)
+    cpdef set_item(self, index_t line_nb, str element_name, object value)

python/obitools3/obidms/_obidmscolumn_str.pyx

@@ -1,13 +1,12 @@
 #cython: language_level=3
  
-from .capi.obidmscolumn cimport obi_close_column,\
-                                obi_truncate_and_close_column, \
-                                obi_column_get_obistr_with_elt_name, \
-                                obi_column_get_obistr_with_elt_idx, \
-                                obi_column_set_obistr_with_elt_name, \
-                                obi_column_set_obistr_with_elt_idx
+from .capi.obidmscolumn cimport obi_truncate_and_close_column, \
+                                obi_column_get_obistr_with_elt_name_in_view, \
+                                obi_column_get_obistr_with_elt_idx_in_view, \
+                                obi_column_set_obistr_with_elt_name_in_view, \
+                                obi_column_set_obistr_with_elt_idx_in_view
 from .capi.obierrno     cimport obi_errno
-from .capi.obitypes     cimport OBIIdx_NA, const_char_p
+from .capi.obitypes     cimport OBIStr_NA, const_char_p
 
 from obitools3.utils cimport str2bytes, bytes2str
 
@@ -17,43 +16,34 @@ cdef class OBIDMS_column_str(OBIDMS_column):
     cpdef object get_line(self, index_t line_nb):
         cdef bytes value
         cdef object result
-        value = <bytes> obi_column_get_obistr_with_elt_idx(self.pointer, line_nb, 0)
+        value = <bytes> obi_column_get_obistr_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0)
         if obi_errno > 0 :
             raise IndexError(line_nb)
-        if value == OBIIdx_NA :
+        if value == OBIStr_NA :     # TODO
             result = None
         else :
             result = bytes2str(value)
         return result
 
     cpdef set_line(self, index_t line_nb, object value):
-        raise Exception("Column is read-only")
-    
-    cpdef close(self):
-        if obi_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-
-
-cdef class OBIDMS_column_str_writable(OBIDMS_column_str):     
-
-    cpdef set_line(self, index_t line_nb, object value):
-        if obi_column_set_obistr_with_elt_idx(self.pointer, line_nb, 0, str2bytes(value)) < 0:
+        cdef bytes value_b
+        if value is None :
+            value_b = OBIStr_NA
+        else :
+            value_b = str2bytes(value)
+        if obi_column_set_obistr_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, 0, value_b) < 0:
             raise Exception("Problem setting a value in a column")
 
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
 
-
-cdef class OBIDMS_column_str_multi_elts(OBIDMS_column_str):
+cdef class OBIDMS_column_multi_elts_str(OBIDMS_column_multi_elts):
 
     cpdef object get_item(self, index_t line_nb, str element_name):
         cdef bytes value
         cdef object result
-        value = <bytes> obi_column_get_obistr_with_elt_name(self.pointer, line_nb, str2bytes(element_name))
+        value = <bytes> obi_column_get_obistr_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name))
         if obi_errno > 0 :
             raise IndexError(line_nb, element_name)
-        if value == OBIIdx_NA :
+        if value == OBIStr_NA :
             result = None
         else :
             result = bytes2str(value)
@@ -61,43 +51,33 @@ cdef class OBIDMS_column_str_multi_elts(OBIDMS_column_str):
 
     cpdef object get_line(self, index_t line_nb) :
         cdef bytes value
-        cdef object result
+        cdef object value_in_result
+        cdef dict result
         cdef index_t i
         cdef bint all_NA
         result = {}
         all_NA = True
         for i in range(self.nb_elements_per_line) :
-            value = <bytes> obi_column_get_obistr_with_elt_idx(self.pointer, line_nb, i)
+            value = <bytes> obi_column_get_obistr_with_elt_idx_in_view(self.view, (self.pointer)[0], line_nb, i)
             if obi_errno > 0 :
                 raise IndexError(line_nb)
-            result[self.elements_names[i]] = bytes2str(value)
-            if all_NA and (value != OBIIdx_NA) :
+            if value == OBIStr_NA :
+                value_in_result = None
+            else :
+                value_in_result = bytes2str(value)    
+            result[self.elements_names[i]] = value_in_result
+            if all_NA and (value_in_result is not None) :
                 all_NA = False
         if all_NA :
             result = None
         return result
 
-    cpdef set_item(self, index_t line_nb, str element_name, str value):
-        raise Exception("Column is read-only")
-
-    cpdef set_line(self, index_t line_nb, object values):
-        raise Exception("Column is read-only")
-
-
-cdef class OBIDMS_column_str_multi_elts_writable(OBIDMS_column_str_multi_elts):     
-
-    cpdef set_item(self, index_t line_nb, str element_name, str value):
-        if obi_column_set_obistr_with_elt_name(self.pointer, line_nb, str2bytes(element_name), str2bytes(value)) < 0:
+    cpdef set_item(self, index_t line_nb, str element_name, object value):
+        cdef bytes value_b
+        if value is None :
+            value_b = OBIStr_NA
+        else :
+            value_b = str2bytes(value)
+        if obi_column_set_obistr_with_elt_name_in_view(self.view, (self.pointer)[0], line_nb, str2bytes(element_name), value_b) < 0:
             raise Exception("Problem setting a value in a column")
     
-    cpdef set_line(self, index_t line_nb, object values):
-        cdef str value
-        for element_name in values :
-            value = values[element_name]
-            self.set_item(line_nb, element_name, value)
-
-    cpdef close(self):
-        if obi_truncate_and_close_column(self.pointer) < 0 :
-            raise Exception("Problem closing a column")
-
- 

python/obitools3/obidms/_obiseq.cfiles

@@ -0,0 +1,22 @@
+../../../src/obidms.h
+../../../src/obidms.c
+../../../src/obiview.h
+../../../src/obiview.c
+../../../src/obidmscolumn.h
+../../../src/obidmscolumn.c
+../../../src/obidmscolumndir.h
+../../../src/obidmscolumndir.c
+../../../src/obierrno.h
+../../../src/obierrno.c
+../../../src/obilittlebigman.h
+../../../src/obilittlebigman.c
+../../../src/obitypes.h
+../../../src/obitypes.c
+../../../src/private_at_functions.h
+../../../src/private_at_functions.c
+../../../src/obiavl.h
+../../../src/obiavl.c
+../../../src/encode.h
+../../../src/encode.c
+../../../src/obidmscolumn_idx.h
+../../../src/obidmscolumn_idx.c

python/obitools3/obidms/_obiseq.pxd

@@ -0,0 +1,34 @@
+#cython: language_level=3
+
+from ._obidms cimport OBIView_line
+
+
+cdef class OBI_Seq(dict) :
+    cdef str id
+    cdef str description
+    cdef str sequence
+
+    cpdef set_id(self, str id)
+    cpdef get_id(self)
+    cpdef set_description(self, str description)
+    cpdef get_description(self)
+    cpdef get_sequence(self)
+
+
+cdef class OBI_Nuc_Seq(OBI_Seq) :
+    #cpdef str reverse_complement(self)    
+    cpdef set_sequence(self, str sequence)
+
+
+cdef class OBI_Nuc_Seq_Stored(OBIView_line) :
+    cdef str id
+    cdef str description
+    cdef str sequence
+
+    cpdef set_id(self, str id)
+    cpdef get_id(self)
+    cpdef set_description(self, str description)
+    cpdef get_description(self)
+    cpdef set_sequence(self, str sequence)
+    cpdef get_sequence(self)
+   # cpdef str reverse_complement(self)

python/obitools3/obidms/_obiseq.pyx

@@ -0,0 +1,78 @@
+#cython: language_level=3
+
+from obitools3.utils cimport bytes2str, str2bytes
+
+from .capi.obiview cimport NUC_SEQUENCE_COLUMN, \
+                           ID_COLUMN, \
+                           DESCRIPTION_COLUMN
+
+cdef class OBI_Seq(dict) :
+
+    def __init__(self, str id, str seq, str description=None) :
+        self.set_id(id)
+        self.set_sequence(seq)
+        if description is not None :
+            self.set_description(description)
+        
+    cpdef set_id(self, str id) :
+        self.id = id
+        self[bytes2str(ID_COLUMN)] = id
+        
+    cpdef get_id(self) :
+        return self.id
+            
+    cpdef set_description(self, str description) :
+        self.description = description
+        self[bytes2str(DESCRIPTION_COLUMN)] = description
+        
+    cpdef get_description(self) :
+        return self.description
+        
+    cpdef get_sequence(self) :
+        return self.sequence
+
+    def __str__(self) :
+        return self.sequence    # or not
+
+
+cdef class OBI_Nuc_Seq(OBI_Seq) :
+    
+    cpdef set_sequence(self, str sequence) :
+        self.sequence = sequence
+        self[bytes2str(NUC_SEQUENCE_COLUMN)] = sequence
+
+#    cpdef str reverse_complement(self) :    TODO in C ?
+#        pass
+
+
+cdef class OBI_Nuc_Seq_Stored(OBIView_line) :
+
+    cpdef set_id(self, str id) :
+        self.id = id
+        self[bytes2str(ID_COLUMN)] = id
+        
+    cpdef get_id(self) :
+        return self.id
+            
+    cpdef set_description(self, str description) :
+        self.description = description
+        self[bytes2str(DESCRIPTION_COLUMN)] = description
+
+    cpdef get_description(self) :
+        return self.description
+
+    cpdef set_sequence(self, str sequence) :
+        self.sequence = sequence
+        self[bytes2str(NUC_SEQUENCE_COLUMN)] = sequence
+
+    cpdef get_sequence(self) :
+        return self.sequence
+    
+    def __str__(self) :
+        return self.sequence    # or not
+    
+#    cpdef str reverse_complement(self) :    TODO in C ?
+#        pass
+
+    # TODO static method to import?
+    

python/obitools3/obidms/_obitaxo.cfiles

@@ -0,0 +1,24 @@
+../../../src/obidms.h
+../../../src/obidms.c
+../../../src/obiview.h
+../../../src/obiview.c
+../../../src/obidmscolumn.h
+../../../src/obidmscolumn.c
+../../../src/obidmscolumndir.h
+../../../src/obidmscolumndir.c
+../../../src/obierrno.h
+../../../src/obierrno.c
+../../../src/obilittlebigman.h
+../../../src/obilittlebigman.c
+../../../src/obitypes.h
+../../../src/obitypes.c
+../../../src/private_at_functions.h
+../../../src/private_at_functions.c
+../../../src/obiavl.h
+../../../src/obiavl.c
+../../../src/encode.h
+../../../src/encode.c
+../../../src/obidmscolumn_idx.h
+../../../src/obidmscolumn_idx.c
+../../../src/obidms_taxonomy.c
+../../../src/obidms_taxonomy.h

python/obitools3/obidms/_obitaxo.pxd

@@ -0,0 +1,31 @@
+#cython: language_level=3
+
+from .capi.obitaxonomy cimport ecotx_t, OBIDMS_taxonomy_p
+
+from libc.stdint   cimport int32_t
+
+
+cdef class OBI_Taxonomy :
+
+    cdef str name
+    cdef OBIDMS_taxonomy_p pointer
+    
+    cpdef close(self)
+
+ 
+cdef class OBI_Taxon :
+
+    cdef ecotx_t* pointer
+    cdef int32_t  taxid
+    cdef int32_t  rank
+    cdef int32_t  farest
+    cdef ecotx_t* parent
+    cdef str name
+        
+    cpdef int32_t taxid(self)
+    
+    cpdef int32_t rank(self)
+    
+    cpdef int32_t farest(self)
+    
+    cpdef OBI_Taxon parent(self)

python/obitools3/obidms/_obitaxo.pyx

@@ -0,0 +1,65 @@
+#cython: language_level=3
+
+from obitools3.utils   cimport bytes2str, str2bytes
+
+from .capi.obitaxonomy cimport obi_read_taxonomy, \
+                               obi_close_taxonomy, \
+                               obi_taxo_get_taxon_with_taxid
+
+from ._obidms cimport OBIDMS
+
+from cpython.pycapsule cimport PyCapsule_New, PyCapsule_GetPointer
+
+
+cdef class OBI_Taxonomy :
+
+    def __init__(self, OBIDMS dms, str name) :
+        
+        self.name = name
+        self.pointer = obi_read_taxonomy(dms.pointer, str2bytes(name), True)    # TODO discuss
+    
+    
+    def __getitem__(self, object ref):
+        
+        cdef ecotx_t* taxon_p
+        cdef object taxon_capsule
+        
+        if type(ref) == int :
+            taxon_p = obi_taxo_get_taxon_with_taxid(self.pointer, ref)
+            taxon_capsule = PyCapsule_New(taxon_p, NULL, NULL)
+            return OBI_Taxon(taxon_capsule)
+    
+         
+    cpdef close(self) :
+        if (obi_close_taxonomy(self.pointer) < 0) :
+            raise Exception("Error closing the taxonomy")
+        
+
+cdef class OBI_Taxon :    # dict subclass?
+
+    def __init__(self, object taxon_capsule) :
+        
+        cdef ecotx_t* taxon
+        
+        taxon = <ecotx_t*> PyCapsule_GetPointer(taxon_capsule, NULL)   
+        self.pointer = taxon
+        self.taxid = taxon.taxid
+        self.rank = taxon.rank
+        self.farest = taxon.farest
+        self.parent = taxon.parent
+        self.name = bytes2str(taxon.name)
+        
+    cpdef int32_t taxid(self):
+        return self.taxid
+    
+    cpdef int32_t rank(self):
+        return self.rank
+    
+    cpdef int32_t farest(self):
+        return self.farest
+    
+    cpdef OBI_Taxon parent(self):
+        cdef object parent_capsule
+        parent_capsule = PyCapsule_New(self.parent, NULL, NULL)
+        return OBI_Taxon(parent_capsule)
+    

python/obitools3/obidms/capi/obidmscolumn.pxd

@@ -15,24 +15,27 @@ from ..capi.obitypes cimport const_char_p, \
 cdef extern from "obidmscolumn.h" nogil:
     
     struct OBIDMS_column_header_t:
-        bint            little_endian
-        int             header_size
+        size_t          header_size
+        size_t          data_size
         index_t         line_count
         index_t         lines_used
         index_t         nb_elements_per_line
         const_char_p    elements_names
-        OBIType_t       data_type
+        OBIType_t       returned_data_type
+        OBIType_t       stored_data_type
         time_t          creation_date
         obiversion_t    version
         obiversion_t    cloned_from
         const_char_p    name
-        const_char_p    array_name
+        const_char_p    avl_name
         const_char_p    comments
     
     ctypedef OBIDMS_column_header_t* OBIDMS_column_header_p
 
     struct OBIDMS_column_t:
+        OBIDMS_p               dms
         OBIDMS_column_header_p header
+        bint                   writable
         
     ctypedef OBIDMS_column_t* OBIDMS_column_p
 
@@ -42,7 +45,8 @@ cdef extern from "obidmscolumn.h" nogil:
                                       index_t nb_lines, 
                                       index_t nb_elements_per_line, 
                                       const_char_p elements_names,
-                                      const_char_p array_name)
+                                      const_char_p avl_name,
+                                      const_char_p comments)
                 
     OBIDMS_column_p obi_open_column(OBIDMS_p dms, 
                                     const_char_p column_name, 
@@ -51,6 +55,7 @@ cdef extern from "obidmscolumn.h" nogil:
     int obi_close_column(OBIDMS_column_p column)
         
     OBIDMS_column_p obi_clone_column(OBIDMS_p dms, 
+                                     OBIDMS_column_p line_selection,
                                      const_char_p column_name, 
                                      obiversion_t version_number, 
                                      bint clone_data)
@@ -61,14 +66,41 @@ cdef extern from "obidmscolumn.h" nogil:
                                                          const_char_p column_name)
         
     OBIDMS_column_header_p obi_column_get_header_from_name(OBIDMS_p dms, 
-                                                           const_char_p column_name)
+                                                           const_char_p column_name,
+                                                           obiversion_t version_number)
     
     int obi_unmap_header(OBIDMS_column_header_p header)
         
     char* obi_column_format_date(time_t date)
 
+    int obi_select(OBIDMS_column_p line_selection_column, index_t line_to_grep)
+
+
+from ..capi.obiview  cimport Obiview_p  # TODO
 
 cdef extern from "obidmscolumn_int.h" nogil:
+    int obi_column_set_obiint_with_elt_name_in_view(Obiview_p view,
+                                                    OBIDMS_column_p column, 
+                                                    index_t line_nb, 
+                                                    const_char_p element_name, 
+                                                    obiint_t value)
+
+    int obi_column_set_obiint_with_elt_idx_in_view(Obiview_p view,
+                                                   OBIDMS_column_p column, 
+                                                   index_t line_nb, 
+                                                   index_t element_idx, 
+                                                   obiint_t value)
+    
+    obiint_t obi_column_get_obiint_with_elt_name_in_view(Obiview_p view,
+                                                         OBIDMS_column_p column, 
+                                                         index_t line_nb, 
+                                                         const_char_p element_name)
+    
+    obiint_t obi_column_get_obiint_with_elt_idx_in_view(Obiview_p view,
+                                                        OBIDMS_column_p column, 
+                                                        index_t line_nb, 
+                                                        index_t element_idx)
+
     int obi_column_set_obiint_with_elt_name(OBIDMS_column_p column, 
                                             index_t line_nb, 
                                             const_char_p element_name, 
@@ -77,7 +109,7 @@ cdef extern from "obidmscolumn_int.h" nogil:
     int obi_column_set_obiint_with_elt_idx(OBIDMS_column_p column, 
                                            index_t line_nb, 
                                            index_t element_idx, 
-                                        obiint_t value)
+                                           obiint_t value)
     
     obiint_t obi_column_get_obiint_with_elt_name(OBIDMS_column_p column, 
                                                  index_t line_nb, 
@@ -87,9 +119,30 @@ cdef extern from "obidmscolumn_int.h" nogil:
                                                 index_t line_nb, 
                                                 index_t element_idx)
 
-
 cdef extern from "obidmscolumn_bool.h" nogil:
 
+    int obi_column_set_obibool_with_elt_name_in_view(Obiview_p view,
+                                                     OBIDMS_column_p column, 
+                                                     index_t line_nb, 
+                                                     const_char_p element_name, 
+                                                     obibool_t value)
+
+    int obi_column_set_obibool_with_elt_idx_in_view(Obiview_p view,
+                                                    OBIDMS_column_p column, 
+                                                    index_t line_nb, 
+                                                    index_t element_idx, 
+                                                    obibool_t value)
+    
+    obibool_t obi_column_get_obibool_with_elt_name_in_view(Obiview_p view,
+                                                          OBIDMS_column_p column, 
+                                                          index_t line_nb, 
+                                                          const_char_p element_name)
+    
+    obibool_t obi_column_get_obibool_with_elt_idx_in_view(Obiview_p view,
+                                                         OBIDMS_column_p column, 
+                                                         index_t line_nb, 
+                                                         index_t element_idx)
+
     int obi_column_set_obibool_with_elt_name(OBIDMS_column_p column, 
                                              index_t line_nb, 
                                              const_char_p element_name, 
@@ -110,6 +163,28 @@ cdef extern from "obidmscolumn_bool.h" nogil:
 
 cdef extern from "obidmscolumn_char.h" nogil:
 
+    int obi_column_set_obichar_with_elt_name_in_view(Obiview_p view,
+                                                     OBIDMS_column_p column, 
+                                                     index_t line_nb, 
+                                                     const_char_p element_name, 
+                                                     obichar_t value)
+
+    int obi_column_set_obichar_with_elt_idx_in_view(Obiview_p view,
+                                                    OBIDMS_column_p column, 
+                                                    index_t line_nb, 
+                                                    index_t element_idx, 
+                                                    obichar_t value)
+    
+    obichar_t obi_column_get_obichar_with_elt_name_in_view(Obiview_p view,
+                                                          OBIDMS_column_p column, 
+                                                          index_t line_nb, 
+                                                          const_char_p element_name)
+    
+    obichar_t obi_column_get_obichar_with_elt_idx_in_view(Obiview_p view,
+                                                         OBIDMS_column_p column, 
+                                                         index_t line_nb, 
+                                                         index_t element_idx)
+
     int obi_column_set_obichar_with_elt_name(OBIDMS_column_p column, 
                                              index_t line_nb, 
                                              const_char_p element_name, 
@@ -130,26 +205,70 @@ cdef extern from "obidmscolumn_char.h" nogil:
 
 cdef extern from "obidmscolumn_float.h" nogil:
 
-    int obi_column_set_obifloat_with_elt_name(OBIDMS_column_p column, 
-                                              index_t line_nb, 
-                                              const_char_p element_name, 
-                                              obifloat_t value)
+    int obi_column_set_obifloat_with_elt_name_in_view(Obiview_p view,
+                                                     OBIDMS_column_p column, 
+                                                     index_t line_nb, 
+                                                     const_char_p element_name, 
+                                                     obifloat_t value)
 
-    int obi_column_set_obifloat_with_elt_idx(OBIDMS_column_p column, 
+    int obi_column_set_obifloat_with_elt_idx_in_view(Obiview_p view,
+                                                    OBIDMS_column_p column, 
+                                                    index_t line_nb, 
+                                                    index_t element_idx, 
+                                                    obifloat_t value)
+    
+    obifloat_t obi_column_get_obifloat_with_elt_name_in_view(Obiview_p view,
+                                                          OBIDMS_column_p column, 
+                                                          index_t line_nb, 
+                                                          const_char_p element_name)
+    
+    obifloat_t obi_column_get_obifloat_with_elt_idx_in_view(Obiview_p view,
+                                                         OBIDMS_column_p column, 
+                                                         index_t line_nb, 
+                                                         index_t element_idx)
+
+    int obi_column_set_obifloat_with_elt_name(OBIDMS_column_p column, 
                                              index_t line_nb, 
-                                             index_t element_idx, 
+                                             const_char_p element_name, 
                                              obifloat_t value)
 
+    int obi_column_set_obifloat_with_elt_idx(OBIDMS_column_p column, 
+                                            index_t line_nb, 
+                                            index_t element_idx, 
+                                            obifloat_t value)
+    
     obifloat_t obi_column_get_obifloat_with_elt_name(OBIDMS_column_p column, 
-                                                     index_t line_nb, 
-                                                     const_char_p element_name)
+                                                   index_t line_nb, 
+                                                   const_char_p element_name)
     
     obifloat_t obi_column_get_obifloat_with_elt_idx(OBIDMS_column_p column, 
-                                                    index_t line_nb, 
-                                                    index_t element_idx)
+                                                  index_t line_nb, 
+                                                  index_t element_idx)
 
 cdef extern from "obidmscolumn_str.h" nogil:
 
+    int obi_column_set_obistr_with_elt_name_in_view(Obiview_p view,
+                                                    OBIDMS_column_p column, 
+                                                    index_t line_nb, 
+                                                    const_char_p element_name, 
+                                                    char* value)
+
+    int obi_column_set_obistr_with_elt_idx_in_view(Obiview_p view,
+                                                   OBIDMS_column_p column, 
+                                                   index_t line_nb, 
+                                                   index_t element_idx, 
+                                                   char* value)
+    
+    const_char_p obi_column_get_obistr_with_elt_name_in_view(Obiview_p view,
+                                                           OBIDMS_column_p column, 
+                                                           index_t line_nb, 
+                                                           const_char_p element_name)
+    
+    const_char_p obi_column_get_obistr_with_elt_idx_in_view(Obiview_p view,
+                                                            OBIDMS_column_p column, 
+                                                            index_t line_nb, 
+                                                            index_t element_idx)
+
     int obi_column_set_obistr_with_elt_name(OBIDMS_column_p column, 
                                             index_t line_nb, 
                                             const_char_p element_name, 
@@ -161,10 +280,52 @@ cdef extern from "obidmscolumn_str.h" nogil:
                                            char* value)
     
     const_char_p obi_column_get_obistr_with_elt_name(OBIDMS_column_p column, 
-                                              index_t line_nb, 
-                                              const_char_p element_name)
+                                                    index_t line_nb, 
+                                                    char* element_name)
     
     const_char_p obi_column_get_obistr_with_elt_idx(OBIDMS_column_p column, 
-                                             index_t line_nb, 
-                                             index_t element_idx)
+                                                    index_t line_nb, 
+                                                    index_t element_idx)
+
+cdef extern from "obidmscolumn_seq.h" nogil:
+
+    int obi_column_set_obiseq_with_elt_name_in_view(Obiview_p view,
+                                                    OBIDMS_column_p column, 
+                                                    index_t line_nb, 
+                                                    const_char_p element_name, 
+                                                    char* value)
+
+    int obi_column_set_obiseq_with_elt_idx_in_view(Obiview_p view,
+                                                   OBIDMS_column_p column, 
+                                                   index_t line_nb, 
+                                                   index_t element_idx, 
+                                                   char* value)
+    
+    const_char_p obi_column_get_obiseq_with_elt_name_in_view(Obiview_p view,
+                                                           OBIDMS_column_p column, 
+                                                           index_t line_nb, 
+                                                           const_char_p element_name)
+    
+    const_char_p obi_column_get_obiseq_with_elt_idx_in_view(Obiview_p view,
+                                                            OBIDMS_column_p column, 
+                                                            index_t line_nb, 
+                                                            index_t element_idx)
+
+    int obi_column_set_obiseq_with_elt_name(OBIDMS_column_p column, 
+                                            index_t line_nb, 
+                                            const_char_p element_name, 
+                                            char* value)
+
+    int obi_column_set_obiseq_with_elt_idx(OBIDMS_column_p column, 
+                                           index_t line_nb, 
+                                           index_t element_idx, 
+                                           char* value)
+    
+    const_char_p obi_column_get_obiseq_with_elt_name(OBIDMS_column_p column, 
+                                                    index_t line_nb, 
+                                                    char* element_name)
+    
+    const_char_p obi_column_get_obiseq_with_elt_idx(OBIDMS_column_p column, 
+                                                    index_t line_nb, 
+                                                    index_t element_idx)
 

python/obitools3/obidms/capi/obitaxonomy.pxd

@@ -0,0 +1,42 @@
+#cython: language_level=3
+
+from .obitypes     cimport const_char_p
+from .obidms       cimport OBIDMS_p
+from libc.stdint   cimport int32_t
+
+
+cdef extern from "obidms_taxonomy.h" nogil:
+
+    struct OBIDMS_taxonomy_t
+    ctypedef OBIDMS_taxonomy_t* OBIDMS_taxonomy_p
+    
+    struct ecotxnode :
+        int32_t           taxid
+        int32_t           rank
+        int32_t           farest
+        ecotxnode*        parent
+        char*             name
+    
+    ctypedef ecotxnode ecotx_t
+
+
+    OBIDMS_taxonomy_p obi_read_taxonomy(OBIDMS_p dms, const_char_p taxonomy_name, bint read_alternative_names)
+    
+    int obi_close_taxonomy(OBIDMS_taxonomy_p taxonomy)
+    
+    ecotx_t* obi_taxo_get_parent_at_rank(ecotx_t* taxon, int32_t rankidx)
+    
+    ecotx_t* obi_taxo_get_taxon_with_taxid(OBIDMS_taxonomy_p taxonomy, int32_t taxid)
+    
+    bint obi_taxo_is_taxon_under_taxid(ecotx_t* taxon, int32_t other_taxid)
+    
+    ecotx_t* obi_taxo_get_species(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+    
+    ecotx_t* obi_taxo_get_genus(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+    
+    ecotx_t* obi_taxo_get_family(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+    
+    ecotx_t* obi_taxo_get_kingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+    
+    ecotx_t* obi_taxo_get_superkingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+    

python/obitools3/obidms/capi/obitypes.pxd

@@ -8,6 +8,9 @@ from posix.types cimport time_t
 cdef extern from *:
     ctypedef char* const_char_p "const char*"
 
+cdef extern from "encode.h" nogil:
+    bint only_ATGC(const_char_p seq)
+
 
 cdef extern from "obidmscolumn.h" nogil:
 
@@ -16,14 +19,17 @@ cdef extern from "obidmscolumn.h" nogil:
 
 cdef extern from "obitypes.h" nogil:
 
-    enum OBIType:       # TODO je sais pas si ça sert de declarer le contenu de l'enum
-        OBI_VOID = 0,
+    enum OBIType:
+        OBI_VOID,
         OBI_INT,
         OBI_FLOAT,
         OBI_BOOL,
         OBI_CHAR,
+        OBI_STR,
+        OBI_SEQ,
         OBI_IDX
 
+    
     ctypedef OBIType OBIType_t
     
     enum OBIBool:
@@ -35,10 +41,12 @@ cdef extern from "obitypes.h" nogil:
     ctypedef char obichar_t  
     ctypedef int64_t index_t
 
-    extern obiint_t   OBIInt_NA
-    extern index_t    OBIIdx_NA
-    extern obifloat_t OBIFloat_NA
-    extern obichar_t  OBIChar_NA
-    extern obibool_t  OBIBool_NA
+    extern obiint_t      OBIInt_NA
+    extern index_t       OBIIdx_NA
+    extern obifloat_t    OBIFloat_NA
+    extern obichar_t     OBIChar_NA
+    extern obibool_t     OBIBool_NA
+    extern const_char_p  OBISeq_NA
+    extern const_char_p  OBIStr_NA
 
-    char* name_data_type(int data_type)
+    const_char_p name_data_type(int data_type)

python/obitools3/obidms/capi/obiview.pxd

@@ -0,0 +1,114 @@
+#cython: language_level=3
+
+from .obitypes           cimport const_char_p, \
+                                 OBIType_t, \
+                                 obiversion_t, \
+                                 index_t, \
+                                 time_t
+from ..capi.obidms       cimport OBIDMS_p
+from ..capi.obidmscolumn cimport OBIDMS_column_p
+
+
+cdef extern from "obiview.h" nogil:
+
+    extern const_char_p VIEW_TYPE_NUC_SEQS
+    extern const_char_p NUC_SEQUENCE_COLUMN
+    extern const_char_p ID_COLUMN
+    extern const_char_p DESCRIPTION_COLUMN
+
+    struct Column_reference_t :
+        const_char_p  column_name
+        obiversion_t  version
+    
+    ctypedef Column_reference_t* Column_reference_p
+
+
+    struct Obiview_t :
+        OBIDMS_p            dms
+        const_char_p        name
+        OBIDMS_column_p     line_selection
+        OBIDMS_column_p     new_line_selection
+        OBIDMS_column_p     columns
+        bint                read_only
+        Column_reference_t  line_selection_reference
+        index_t             line_count
+        int                 column_count
+        const_char_p        comments
+
+    ctypedef Obiview_t* Obiview_p
+
+
+    struct Obiview_infos_t :
+        int                 view_number
+        int                 column_count
+        index_t             line_count
+        const_char_p        name
+        const_char_p        created_from
+        time_t              creation_date
+        bint                all_lines
+        Column_reference_t  line_selection
+        Column_reference_p  column_references
+        const_char_p        view_type
+        const_char_p        comments
+
+    ctypedef Obiview_infos_t* Obiview_infos_p
+
+
+    struct Obiviews_header_t :
+        size_t header_size
+        size_t views_size
+        int    view_count
+    
+    ctypedef Obiviews_header_t* Obiviews_header_p
+
+
+    struct Obiviews_infos_all_t :
+        Obiviews_header_p    header
+        Obiview_infos_p      view_infos
+
+    ctypedef Obiviews_infos_all_t* Obiviews_infos_all_p
+
+
+    Obiview_p obi_new_view_nuc_seqs(OBIDMS_p dms, const_char_p view_name, Obiview_p view_to_clone, index_t* line_selection, const_char_p comments)
+
+    Obiview_p obi_new_view(OBIDMS_p dms, const_char_p view_name, Obiview_p view_to_clone, index_t* line_selection, const_char_p comments)
+    
+    Obiview_p obi_new_view_cloned_from_name(OBIDMS_p dms, const_char_p view_name, const_char_p view_to_clone_name, index_t* line_selection, const_char_p comments)
+
+    Obiview_p obi_new_view_nuc_seqs_cloned_from_name(OBIDMS_p dms, const_char_p view_name, const_char_p view_to_clone_name, index_t* line_selection, const_char_p comments)
+              
+    Obiview_p obi_open_view(OBIDMS_p dms, const_char_p view_name)
+    
+    Obiviews_infos_all_p obi_read_views(OBIDMS_p dms)
+
+    int obi_unmap_read_views(Obiviews_infos_all_p views)
+
+    int obi_view_add_column(Obiview_p view,
+                            const_char_p column_name,
+                            obiversion_t version_number,
+                            OBIType_t data_type,
+                            index_t nb_lines,
+                            index_t nb_elements_per_line,
+                            const_char_p elements_names,
+                            const_char_p avl_name,
+                            const_char_p comments,
+                            bint create)
+
+    int obi_view_delete_column(Obiview_p view, const_char_p column_name)
+    
+    int obi_select_line(Obiview_p view, index_t line_nb)
+
+    int obi_select_lines(Obiview_p view, index_t* line_nbs)
+
+    OBIDMS_column_p obi_view_clone_column(Obiview_p view, const_char_p column_name)
+        
+    OBIDMS_column_p obi_view_get_column(Obiview_p view, const_char_p column_name)
+
+    OBIDMS_column_p* obi_view_get_pointer_on_column_in_view(Obiview_p view, const_char_p column_name)
+
+    int obi_save_view(Obiview_p view)
+    
+    int obi_close_view(Obiview_p view)
+    
+    int obi_save_and_close_view(Obiview_p view)
+

python/obitools3/unit_tests.py

@@ -10,9 +10,8 @@ from obitools3.obidms._obidms import OBIDMS
 
 LINE_COUNT_FOR_TEST_COLUMN = 10000         # TODO randomize?
 SMALLER_LINE_COUNT_FOR_TEST_COLUMN = 1000  # TODO randomize?
-NB_ELEMENTS_PER_LINE = 20                  # TODO randomize?
+NB_ELEMENTS_PER_LINE = 10                  # TODO randomize?
 DMS_NAME = "unit_test_dms"
-DATA_TYPES = ['OBI_INT', 'OBI_FLOAT', 'OBI_BOOL', 'OBI_CHAR', 'OBI_IDX']
 
 
 def create_test_obidms():
@@ -22,26 +21,24 @@ def create_test_obidms():
     return (dms, dms_name, dms_dir_name)
 
 
-def create_test_column(dms, data_type_code, multiple_elements_per_line=False):
-    data_types = DATA_TYPES
-    data_type_code = data_type_code
-    data_type_str = data_types[data_type_code-1]
-    col_name = "unit_test_"+data_type_str
+def create_test_column(dms, data_type, multiple_elements_per_line=False):
+    
+    col_name = "unit_test_"+data_type
 
     if multiple_elements_per_line :
         elts_names = elements_names()
         col = dms.open_column(col_name, 
                               create=True, 
-                              data_type=data_type_code, 
+                              type=data_type, 
                               nb_elements_per_line=NB_ELEMENTS_PER_LINE,
                               elements_names=elts_names)
-        return (col, col_name, elts_names, data_type_str)
+        return (col, col_name, elts_names)
 
     else :
         col = dms.open_column(col_name, 
                               create=True, 
-                              data_type=data_type_code)
-        return (col, col_name, data_type_str)
+                              type=data_type)
+        return (col, col_name)
 
 
 def elements_names():
@@ -58,12 +55,15 @@ def random_obivalue(data_type):
     elif data_type == "OBI_BOOL" :
         return randint(0,1)
     elif data_type == "OBI_CHAR" :
-        nucs = 'atgc'
-        return bytes(nucs[randint(0,3)], 'utf-8')
-    elif data_type == "OBI_IDX" :
-        length = randint(1,500)
+        return choice(string.ascii_lowercase)
+    elif data_type == "OBI_STR" :
+        length = randint(1,200)
         randoms = ''.join(choice(string.ascii_lowercase) for i in range(length))
         return randoms
+    elif data_type == "OBI_SEQ" :
+        length = randint(1,200)
+        randoms = ''.join(choice("atgcryswkmdbhvn") for i in range(length))
+        return randoms
 
 class OBIDMS_Column_TestCase(unittest.TestCase):
     def tearDown(self):
@@ -71,10 +71,10 @@ class OBIDMS_Column_TestCase(unittest.TestCase):
         self.dms.close()
         shutil.rmtree(self.dms_dir_name, ignore_errors=True)
     def test_OBIDMS_column_type(self):
-        assert self.col.get_data_type() == self.data_type_str, 'Wrong data type associated with column'
+        assert self.col.get_data_type() == self.data_type, 'Wrong data type associated with column'
     def test_OBIDMS_column_cloning(self):
         for i in range(LINE_COUNT_FOR_TEST_COLUMN):
-            self.col[i]= random_obivalue(self.data_type_str)
+            self.col[i]= random_obivalue(self.data_type)
         self.col.close()
         clone = self.dms.open_column(self.col_name, clone=True)
         self.col = self.dms.open_column(self.col_name)
@@ -86,10 +86,21 @@ class OBIDMS_Column_TestCase(unittest.TestCase):
         clone.close()
     def test_OBIDMS_column_set_and_get(self):
         for i in range(LINE_COUNT_FOR_TEST_COLUMN):
-            v = random_obivalue(self.data_type_str)
+            v = random_obivalue(self.data_type)
             self.col[i] = v
             assert self.col[i] == v, "Different value than the set value"
             assert self.col[i] is not None, "None value"
+    def test_OBIDMS_referring_column(self):
+        for i in range(LINE_COUNT_FOR_TEST_COLUMN):
+            self.col[i] = random_obivalue(self.data_type)
+        ref_col = self.dms.open_column(self.col_name, referring=True)
+        j = 0
+        for i in range(LINE_COUNT_FOR_TEST_COLUMN):
+            if i%2 :    # TODO randomize
+                ref_col.grep_line(i)
+                assert ref_col[j] == self.col[i], "Different value in original column and returned by referring column"
+                assert ref_col[j] is not None, "None value"
+                j+=1
 
 
 class OBIDMS_Column_multiple_elements_TestCase(OBIDMS_Column_TestCase):
@@ -98,7 +109,7 @@ class OBIDMS_Column_multiple_elements_TestCase(OBIDMS_Column_TestCase):
         for i in range(SMALLER_LINE_COUNT_FOR_TEST_COLUMN):
             v = {}
             for e in self.elts_names :
-                v[e] = random_obivalue(self.data_type_str)
+                v[e] = random_obivalue(self.data_type)
             self.col[i] = v
         self.col.close()
         clone = self.dms.open_column(self.col_name, clone=True)
@@ -112,7 +123,7 @@ class OBIDMS_Column_multiple_elements_TestCase(OBIDMS_Column_TestCase):
     def test_OBIDMS_column_set_and_get_with_elements_names(self):
         for i in range(SMALLER_LINE_COUNT_FOR_TEST_COLUMN):
             for e in range(NB_ELEMENTS_PER_LINE) :
-                v = random_obivalue(self.data_type_str)
+                v = random_obivalue(self.data_type)
                 self.col.set_item(i, self.elts_names[e], v)
                 assert self.col.get_item(i, self.elts_names[e]) == v, "Different value than the set value"
                 assert self.col.get_item(i, self.elts_names[e]) is not None, "None value"
@@ -120,141 +131,168 @@ class OBIDMS_Column_multiple_elements_TestCase(OBIDMS_Column_TestCase):
         for i in range(SMALLER_LINE_COUNT_FOR_TEST_COLUMN):
             v = {}
             for e in self.elts_names :
-                v[e] = random_obivalue(self.data_type_str)
+                v[e] = random_obivalue(self.data_type)
             self.col[i] = v
             assert self.col[i] == v, "Different value than the set value"
             assert self.col[i] is not None, "None value"
+    def test_OBIDMS_referring_column(self):
+        for i in range(SMALLER_LINE_COUNT_FOR_TEST_COLUMN):
+            v = {}
+            for e in self.elts_names :
+                v[e] = random_obivalue(self.data_type)
+            self.col[i] = v
+        ref_col = self.dms.open_column(self.col_name, referring=True)
+        j = 0
+        for i in range(SMALLER_LINE_COUNT_FOR_TEST_COLUMN):
+            if i%2 :    # TODO randomize
+                ref_col.grep_line(i)
+                assert ref_col[j] == self.col[i], "Different value in original column and returned by referring column"
+                assert ref_col[j] is not None, "None value"
+                j+=1
+        ref_col.close()
 
 
 class OBIDMS_Column_OBI_INT_TestCase(OBIDMS_Column_TestCase):
     def setUp(self):
-        self.data_type_code = 1
+        self.data_type = 'OBI_INT'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
-        self.col_name, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code)
+        self.col_name = create_test_column(self.dms, 
+                                           self.data_type)
        
 
 class OBIDMS_Column_OBI_INT_multiple_elements_TestCase(OBIDMS_Column_multiple_elements_TestCase):
     def setUp(self):
-        self.data_type_code = 1
+        self.data_type = 'OBI_INT'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
         self.col_name, \
-        self.elts_names, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code,
-                                                multiple_elements_per_line=True)
+        self.elts_names = create_test_column(self.dms, 
+                                             self.data_type,
+                                             multiple_elements_per_line=True)
 
 
 class OBIDMS_Column_OBI_FLOAT_TestCase(OBIDMS_Column_TestCase):
     def setUp(self):
-        self.data_type_code = 2
+        self.data_type = 'OBI_FLOAT'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
-        self.col_name, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code)
+        self.col_name = create_test_column(self.dms, 
+                                           self.data_type)
        
 
 class OBIDMS_Column_OBI_FLOAT_multiple_elements_TestCase(OBIDMS_Column_multiple_elements_TestCase):
     def setUp(self):
-        self.data_type_code = 2
+        self.data_type = 'OBI_FLOAT'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
         self.col_name, \
-        self.elts_names, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code,
-                                                multiple_elements_per_line=True)
+        self.elts_names = create_test_column(self.dms, 
+                                             self.data_type,
+                                             multiple_elements_per_line=True)
 
 
 class OBIDMS_Column_OBI_BOOL_TestCase(OBIDMS_Column_TestCase):
     def setUp(self):
-        self.data_type_code = 3
+        self.data_type = 'OBI_BOOL'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
-        self.col_name, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code)
+        self.col_name = create_test_column(self.dms, 
+                                           self.data_type)
        
 
 class OBIDMS_Column_OBI_BOOL_multiple_elements_TestCase(OBIDMS_Column_multiple_elements_TestCase):
     def setUp(self):
-        self.data_type_code = 3
+        self.data_type = 'OBI_BOOL'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
         self.col_name, \
-        self.elts_names, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code,
-                                                multiple_elements_per_line=True)
+        self.elts_names = create_test_column(self.dms, 
+                                             self.data_type,
+                                             multiple_elements_per_line=True)
 
 
 class OBIDMS_Column_OBI_CHAR_TestCase(OBIDMS_Column_TestCase):
     def setUp(self):
-        self.data_type_code = 4
+        self.data_type = 'OBI_CHAR'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
-        self.col_name, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code)
+        self.col_name = create_test_column(self.dms, 
+                                                self.data_type)
        
 
 class OBIDMS_Column_OBI_CHAR_multiple_elements_TestCase(OBIDMS_Column_multiple_elements_TestCase):
     def setUp(self):
-        self.data_type_code = 4
+        self.data_type = 'OBI_CHAR'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
         self.col_name, \
-        self.elts_names, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code,
-                                                multiple_elements_per_line=True)
+        self.elts_names = create_test_column(self.dms, 
+                                             self.data_type,
+                                             multiple_elements_per_line=True)
 
 
 class OBIDMS_Column_OBI_STR_TestCase(OBIDMS_Column_TestCase):
     def setUp(self):
-        self.data_type_code = 5
+        self.data_type = 'OBI_STR'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
-        self.col_name, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code)
+        self.col_name = create_test_column(self.dms, 
+                                           self.data_type)
        
 
 class OBIDMS_Column_OBI_STR_multiple_elements_TestCase(OBIDMS_Column_multiple_elements_TestCase):
     def setUp(self):
-        self.data_type_code = 5
+        self.data_type = 'OBI_STR'
         self.dms, \
         self.dms_name, \
         self.dms_dir_name = create_test_obidms()
         self.col, \
         self.col_name, \
-        self.elts_names, \
-        self.data_type_str = create_test_column(self.dms, 
-                                                self.data_type_code,
-                                                multiple_elements_per_line=True)
+        self.elts_names = create_test_column(self.dms, 
+                                             self.data_type,
+                                             multiple_elements_per_line=True)
 
+class OBIDMS_Column_OBI_SEQ_TestCase(OBIDMS_Column_TestCase):
+    def setUp(self):
+        self.data_type = 'OBI_SEQ'
+        self.dms, \
+        self.dms_name, \
+        self.dms_dir_name = create_test_obidms()
+        self.col, \
+        self.col_name = create_test_column(self.dms, 
+                                           self.data_type)
+       
+
+class OBIDMS_Column_OBI_SEQ_multiple_elements_TestCase(OBIDMS_Column_multiple_elements_TestCase):
+    def setUp(self):
+        self.data_type = 'OBI_SEQ'
+        self.dms, \
+        self.dms_name, \
+        self.dms_dir_name = create_test_obidms()
+        self.col, \
+        self.col_name, \
+        self.elts_names = create_test_column(self.dms, 
+                                             self.data_type,
+                                             multiple_elements_per_line=True)
 
 if __name__ == '__main__': 
      unittest.main(verbosity=2, defaultTest=["OBIDMS_Column_OBI_INT_TestCase",
@@ -266,6 +304,8 @@ if __name__ == '__main__':
                                              "OBIDMS_Column_OBI_CHAR_TestCase",
                                              "OBIDMS_Column_OBI_CHAR_multiple_elements_TestCase",
                                              "OBIDMS_Column_OBI_STR_TestCase",
-                                             "OBIDMS_Column_OBI_STR_multiple_elements_TestCase"])
+                                             "OBIDMS_Column_OBI_STR_multiple_elements_TestCase",
+                                             "OBIDMS_Column_OBI_SEQ_TestCase",
+                                             "OBIDMS_Column_OBI_SEQ_multiple_elements_TestCase"])
 
 

python/obitools3/version.py

@@ -1,5 +1,5 @@
-major = 1
-minor = 1
-serial= '16'
+major = 0
+minor = 0
+serial= '0'
 
 version ="%d.%02d.%s" % (major,minor,serial)

src/encode.c

@@ -0,0 +1,345 @@
+/****************************************************************************
+ * Encoding functions                                                       *
+ ****************************************************************************/
+
+/**
+ * @file encode.c
+ * @author Celine Mercier
+ * @date November 18th 2015
+ * @brief Functions encoding DNA sequences.
+ */
+
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <math.h>
+
+#include "encode.h"
+#include "obierrno.h"
+#include "obitypes.h"	// For byte_t type
+#include "obidebug.h"
+
+
+#define DEBUG_LEVEL 0	// TODO has to be defined somewhere else (cython compil flag?)
+
+
+// TODO: endianness problem?
+
+
+
+bool only_ATGC(char* seq)
+{
+	char* c = seq;
+
+	while (*c)
+	{
+		if (!((*c == 'A') || \
+			  (*c == 'T') || \
+			  (*c == 'G') || \
+			  (*c == 'C') || \
+			  (*c == 'a') || \
+			  (*c == 't') || \
+			  (*c == 'g') || \
+			  (*c == 'c')))
+		{
+			return 0;
+		}
+		else
+		{
+			c++;
+		}
+	}
+	return 1;
+}
+
+
+byte_t* encode_seq_on_2_bits(char* seq, int32_t length)
+{
+	byte_t*  seq_b;
+	uint8_t  modulo;
+	int32_t  length_b;
+	int32_t  i;
+
+	length_b = ceil((double) length / (double) 4.0);
+
+	seq_b = (byte_t*) malloc(length_b * sizeof(byte_t));
+
+	// Initialize all the bits to 0
+	memset(seq_b, 0, length_b);
+
+	for (i=0; i<length; i++)
+	{
+		// Shift of 2 to make place for new nucleotide
+		seq_b[i/4] <<= 2;
+
+		// Add new nucleotide
+		switch (seq[i])
+		{
+		case 'a':
+		case 'A':
+			seq_b[i/4] |= NUC_A_2b;
+			break;
+		case 'c':
+		case 'C':
+			seq_b[i/4] |= NUC_C_2b;
+			break;
+		case 'g':
+		case 'G':
+			seq_b[i/4] |= NUC_G_2b;
+			break;
+		case 't':
+		case 'T':
+			seq_b[i/4] |= NUC_T_2b;
+			break;
+		default:
+			obidebug(1, "\nInvalid nucleotide base when encoding (not [atgcATGC])");
+			return NULL;
+		}
+	}
+
+	// Final shift for the last byte if needed
+	modulo = (length % 4);
+	if (modulo)
+		seq_b[(i-1)/4] <<= (2*(4 - modulo));
+
+	return seq_b;
+}
+
+
+char* decode_seq_on_2_bits(byte_t* seq_b, int32_t length_seq)
+{
+	char*   seq;
+	int32_t i;
+	uint8_t shift;
+	uint8_t mask;
+	uint8_t nuc;
+
+	seq = (char*) malloc((length_seq+1) * sizeof(char));
+
+	for (i=0; i<length_seq; i++)
+	{
+		shift = 6 - 2*(i % 4);
+		mask = NUC_MASK_2B << shift;
+		nuc = (seq_b[i/4] & mask) >> shift;
+
+		switch (nuc)
+		{
+		case NUC_A_2b:
+			seq[i] = 'a';
+			break;
+		case NUC_C_2b:
+			seq[i] = 'c';
+			break;
+		case NUC_G_2b:
+			seq[i] = 'g';
+			break;
+		case NUC_T_2b:
+			seq[i] = 't';
+			break;
+		default:
+			obidebug(1, "\nInvalid nucleotide base when decoding");
+			return NULL;
+		}
+	}
+
+	seq[length_seq] = '\0';
+
+	return seq;
+}
+
+
+byte_t* encode_seq_on_4_bits(char* seq, int32_t length)
+{
+	byte_t*  seq_b;
+	uint8_t  modulo;
+	int32_t  length_b;
+	int32_t  i;
+
+	length_b = ceil((double) length / (double) 2.0);
+
+	seq_b = (byte_t*) malloc(length_b * sizeof(byte_t));
+
+	// Initialize all the bits to 0
+	memset(seq_b, 0, length_b);
+
+	for (i=0; i<length; i++)
+	{
+		// Shift of 4 to make place for new nucleotide
+		seq_b[i/2] <<= 4;
+
+		// Add new nucleotide
+		switch (seq[i])
+		{
+		case 'a':
+		case 'A':
+			seq_b[i/2] |= NUC_A_4b;
+			break;
+		case 'c':
+		case 'C':
+			seq_b[i/2] |= NUC_C_4b;
+			break;
+		case 'g':
+		case 'G':
+			seq_b[i/2] |= NUC_G_4b;
+			break;
+		case 't':
+		case 'T':
+			seq_b[i/2] |= NUC_T_4b;
+			break;
+		case 'r':
+		case 'R':
+			seq_b[i/2] |= NUC_R_4b;
+			break;
+		case 'y':
+		case 'Y':
+			seq_b[i/2] |= NUC_Y_4b;
+			break;
+		case 's':
+		case 'S':
+			seq_b[i/2] |= NUC_S_4b;
+			break;
+		case 'w':
+		case 'W':
+			seq_b[i/2] |= NUC_W_4b;
+			break;
+		case 'k':
+		case 'K':
+			seq_b[i/2] |= NUC_K_4b;
+			break;
+		case 'm':
+		case 'M':
+			seq_b[i/2] |= NUC_M_4b;
+			break;
+		case 'b':
+		case 'B':
+			seq_b[i/2] |= NUC_B_4b;
+			break;
+		case 'd':
+		case 'D':
+			seq_b[i/2] |= NUC_D_4b;
+			break;
+		case 'h':
+		case 'H':
+			seq_b[i/2] |= NUC_H_4b;
+			break;
+		case 'v':
+		case 'V':
+			seq_b[i/2] |= NUC_V_4b;
+			break;
+		case 'n':
+		case 'N':
+			seq_b[i/2] |= NUC_N_4b;
+			break;
+		default:
+			obidebug(1, "\nInvalid nucleotide base when encoding (not IUPAC)");
+			return NULL;
+		}
+	}
+
+	// Final shift for the last byte if needed
+	modulo = (length % 2);
+	if (modulo)
+		seq_b[(i-1)/2] <<= (4*modulo);
+
+	return seq_b;
+}
+
+
+char* decode_seq_on_4_bits(byte_t* seq_b, int32_t length_seq)
+{
+	char*   seq;
+	int32_t i;
+	uint8_t shift;
+	uint8_t mask;
+	uint8_t nuc;
+
+	seq = (char*) malloc((length_seq+1) * sizeof(char));
+
+	for (i=0; i<length_seq; i++)
+	{
+		shift = 4 - 4*(i % 2);
+		mask = NUC_MASK_4B << shift;
+		nuc = (seq_b[i/2] & mask) >> shift;
+
+		switch (nuc)
+		{
+		case NUC_A_4b:
+			seq[i] = 'a';
+			break;
+		case NUC_C_4b:
+			seq[i] = 'c';
+			break;
+		case NUC_G_4b:
+			seq[i] = 'g';
+			break;
+		case NUC_T_4b:
+			seq[i] = 't';
+			break;
+		case NUC_R_4b:
+			seq[i] = 'r';
+			break;
+		case NUC_Y_4b:
+			seq[i] = 'y';
+			break;
+		case NUC_S_4b:
+			seq[i] = 's';
+			break;
+		case NUC_W_4b:
+			seq[i] = 'w';
+			break;
+		case NUC_K_4b:
+			seq[i] = 'k';
+			break;
+		case NUC_M_4b:
+			seq[i] = 'm';
+			break;
+		case NUC_B_4b:
+			seq[i] = 'b';
+			break;
+		case NUC_D_4b:
+			seq[i] = 'd';
+			break;
+		case NUC_H_4b:
+			seq[i] = 'h';
+			break;
+		case NUC_V_4b:
+			seq[i] = 'v';
+			break;
+		case NUC_N_4b:
+			seq[i] = 'n';
+			break;
+		default:
+			obidebug(1, "\nInvalid nucleotide base when decoding");
+			return NULL;
+		}
+	}
+
+	seq[length_seq] = '\0';
+
+	return seq;
+}
+
+
+///////////////////// FOR DEBUGGING ///////////////////////////
+//NOTE: The first byte is printed the first (at the left-most).
+
+void print_bits(void* ptr, int32_t size)
+{
+	uint8_t* b = (uint8_t*) ptr;
+	uint8_t byte;
+    int32_t i, j;
+
+    fprintf(stderr, "\n");
+    for (i=0;i<size;i++)
+    {
+        for (j=7;j>=0;j--)
+        {
+            byte = b[i] & (1<<j);
+            byte >>= j;
+            fprintf(stderr, "%u", byte);
+        }
+        fprintf(stderr, " ");
+    }
+    fprintf(stderr, "\n");
+}

src/encode.h

@@ -0,0 +1,181 @@
+/****************************************************************************
+ * Encoding header file	                                                    *
+ ****************************************************************************/
+
+/**
+ * @file encode.h
+ * @author Celine Mercier
+ * @date November 18th 2015
+ * @brief Header file for encoding DNA sequences.
+ */
+
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <stdbool.h>
+
+#include "obitypes.h"
+
+
+#define NUC_MASK_2B 0x3   /**< Binary: 11 to use when decoding 2 bits sequences */
+#define NUC_MASK_4B 0xF  /**< Binary: 1111 to use when decoding 4 bits sequences */
+
+
+/**
+ * @brief enum for the 2-bits codes for each of the 4 nucleotides.
+ */
+enum
+{
+    NUC_A_2b = 0x0,   /* binary: 00 */
+    NUC_C_2b = 0x1,   /* binary: 01 */
+    NUC_G_2b = 0x2,   /* binary: 10 */
+    NUC_T_2b = 0x3,   /* binary: 11 */
+};
+
+
+/**
+ * @brief enum for the 4-bits codes for each of the 15 IUPAC nucleotides.
+ */
+enum
+{
+    NUC_A_4b = 0x1,   /* binary: 0001 */
+    NUC_C_4b = 0x2,   /* binary: 0010 */
+    NUC_G_4b = 0x3,   /* binary: 0011 */
+    NUC_T_4b = 0x4,   /* binary: 0100 */
+	NUC_R_4b = 0x5,   /* binary: 0101 */
+	NUC_Y_4b = 0x6,   /* binary: 0110 */
+	NUC_S_4b = 0x7,   /* binary: 0111 */
+	NUC_W_4b = 0x8,   /* binary: 1000 */
+	NUC_K_4b = 0x9,   /* binary: 1001 */
+	NUC_M_4b = 0xA,   /* binary: 1010 */
+	NUC_B_4b = 0xB,   /* binary: 1011 */
+	NUC_D_4b = 0xC,   /* binary: 1100 */
+	NUC_H_4b = 0xD,   /* binary: 1101 */
+	NUC_V_4b = 0xE,   /* binary: 1110 */
+	NUC_N_4b = 0xF,   /* binary: 1111 */
+};
+
+
+/**
+ * @brief Checks if there are only 'atgcATGC' characters in a
+ *        character string.
+ *
+ * @param seq The sequence to check.
+ *
+ * @returns A boolean value indicating if there are only
+ *          'atgcATGC' characters in a character string.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+bool only_ATGC(char* seq);
+
+
+/**
+ * @brief Encodes a DNA sequence with each nucleotide coded on 2 bits.
+ *
+ *    A or a : 00
+ *    C or c : 01
+ *    T or t : 10
+ *    G or g : 11
+ *
+ * @warning The DNA sequence must contain only 'atgcATGC' characters.
+ *
+ * @param seq The sequence to encode.
+ * @param length The length of the sequence to encode.
+ *
+ * @returns The encoded sequence.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+byte_t* encode_seq_on_2_bits(char* seq, int32_t length);
+
+
+/**
+ * @brief Decodes a DNA sequence that is coded with each nucleotide on 2 bits.
+ *
+ *    00 -> a
+ *    01 -> c
+ *    10 -> t
+ *    11 -> g
+ *
+ * @param seq The sequence to decode.
+ * @param length_seq The initial length of the sequence before it was encoded.
+ *
+ * @returns The decoded sequence ended with '\0'.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+char* decode_seq_on_2_bits(byte_t* seq_b, int32_t length_seq);
+
+
+/**
+ * @brief Encodes a DNA sequence with each nucleotide coded on 4 bits.
+ *
+ *		A or a : 0001
+ *      C or c : 0010
+ *      G or g : 0011
+ *      T or t : 0100
+ *      R or r : 0101
+ *      Y or y : 0110
+ *      S or s : 0111
+ *      W or w : 1000
+ *      K or k : 1001
+ *      M or m : 1010
+ *      B or b : 1011
+ *      D or d : 1100
+ *      H or h : 1101
+ *      V or v : 1110
+ *      N or n : 1111
+ *
+ * @warning The DNA sequence must contain only IUPAC characters.
+ *
+ * @param seq The sequence to encode.
+ * @param length The length of the sequence to encode.
+ *
+ * @returns The encoded sequence.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+byte_t* encode_seq_on_4_bits(char* seq, int32_t length);
+
+
+/**
+ * @brief Decodes a DNA sequence that is coded with each nucleotide on 4 bits.
+ *
+ *		A or a : 0001
+ *      C or c : 0010
+ *      G or g : 0011
+ *      T or t : 0100
+ *      R or r : 0101
+ *      Y or y : 0110
+ *      S or s : 0111
+ *      W or w : 1000
+ *      K or k : 1001
+ *      M or m : 1010
+ *      B or b : 1011
+ *      D or d : 1100
+ *      H or h : 1101
+ *      V or v : 1110
+ *      N or n : 1111
+ *
+ * @param seq The sequence to decode.
+ * @param length_seq The initial length of the sequence before it was encoded.
+ *
+ * @returns The decoded sequence ended with '\0'.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+char* decode_seq_on_4_bits(byte_t* seq_b, int32_t length_seq);
+
+
+////////// FOR DEBUGGING ///////////
+
+// little endian
+void print_bits(void* ptr, int32_t length);
+

src/obiarray.h

@@ -1,285 +0,0 @@
-/****************************************************************************
- * OBIDMS array header file	                                                *
- ****************************************************************************/
-
-/**
- * @file obiarray.h
- * @author Celine Mercier
- * @date October 19th 2015
- * @brief Header file for handling arrays for storing and retrieving byte arrays (i.e. coding for character strings).
- */
-
-
-#ifndef OBIARRAY_H_
-#define OBIARRAY_H_
-
-
-#include <stdlib.h>
-#include <stdio.h>
-#include <stdint.h>
-#include <time.h>
-#include <sys/types.h>
-#include <dirent.h>
-
-#include "obidms.h"
-#include "obitypes.h"
-
-
-#define ARRAY_MAX_NAME (2048) 		/**< The maximum length of an array name.
-                              	  	 */
-#define ARRAY_GROWTH_FACTOR (2)		/**< The growth factor when an array is enlarged.
-                                	 */
-#define BYTE_ARRAY_HEADER_SIZE (5)  /**< The size of the header of a byte array.
-                              	  	 */
-
-
-typedef char byte_t;		/**< Defining byte type since data is stored in bits
-							 *	 and char (stored on one byte) is the smallest addressable unit.
- 	 	 	 	 	 	 	 */
-
-
-/**
- * @brief OBIDMS array data header structure.
- */
-typedef struct OBIDMS_array_data_header {
-	int			header_size;						/**< Size of the header in bytes.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	index_t		data_size_used;						/**< Size of the data used in bytes.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	index_t		data_size_max;						/**< Max size of the data in bytes.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	index_t		nb_items;							/**< Number of items.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	char    	array_name[ARRAY_MAX_NAME+1];		/**< The array name as a NULL terminated string.
-													 */
-	time_t		creation_date;			    	    /**< Date of creation of the file.
-												     */
-} OBIDMS_array_data_header_t, *OBIDMS_array_data_header_p;
-
-
-/**
- * @brief OBIDMS array data structure.
- */
-typedef struct OBIDMS_array_data {
-	OBIDMS_array_data_header_p	header; 		 	/**< A pointer to the header of the array data.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	byte_t*                   	data;   		 	/**< A pointer to the beginning of the data.
-													 */
-} OBIDMS_array_data_t, *OBIDMS_array_data_p;
-
-
-/**
- * @brief OBIDMS array header structure.
- */
-typedef struct OBIDMS_array_header {
-	int			header_size;						/**< Size of the header in bytes.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	size_t		array_size;							/**< Size of the array in bytes.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	index_t		nb_items;							/**< Number of items in the array.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	index_t		nb_items_max;						/**< Maximum number of items in the array before it has to be enlarged.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	char    	array_name[ARRAY_MAX_NAME+1];		/**< The array name as a NULL terminated string.
-													 */
-	time_t		creation_date;			    	    /**< Date of creation of the file.
-												     */
-} OBIDMS_array_header_t, *OBIDMS_array_header_p;
-
-
-/**
- * @brief OBIDMS array structure.
- */
-typedef struct OBIDMS_array {
-	OBIDMS_array_header_p		header; 		 	/**< A pointer to the header of the array.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	index_t*                 	first;   		 	/**< A pointer to the beginning of the array itself.
-													 */
-	OBIDMS_array_data_p			data;				/**< A pointer to the structure containing the data
-													 *   that the array references.
-	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	DIR*						directory;			/**< A directory entry usable to
-										 	 	 	 *    refer and scan the array directory.
-										 	 	 	 */
-} OBIDMS_array_t, *OBIDMS_array_p;
-
-
-/**
- * @brief Checks if an obiarray already exists or not.
- *
- * @param dms The OBIDMS to which the obiarray belongs.
- * @param array_name The name of the obiarray.
- *
- * @returns A value indicating whether the obiarray exists or not.
- * @retval 1 if the obiarray exists.
- * @retval 0 if the obiarray does not exist.
- * @retval -1 if an error occurred.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-int obi_array_exists(OBIDMS_p dms, const char* array_name);
-
-
-/**
- * @brief Opens an obiarray and creates it if it does not already exist.
- *
- * Note: An obiarray is made of two files (referred to by two structures).
- * 		 One file contains the indices referring to the data, and the other
- * 		 file contains the data itself. The obiarray as a whole is referred
- * 		 to via the OBIDMS_array structure.
- *
- * @param dms The OBIDMS to which the obiarray belongs.
- * @param array_name The name of the obiarray.
- *
- * @returns A pointer to the obiarray structure.
- * @retval NULL if an error occurred.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-OBIDMS_array_p obi_array(OBIDMS_p dms, const char* array_name);
-
-
-/**
- * @brief Creates an obiarray. Fails if it already exists.
- *
- * Note: An obiarray is made of two files (referred to by two structures).
- * 		 One file contains the indices referring to the data, and the other
- * 		 file contains the data itself. The obiarray as a whole is referred
- * 		 to via the OBIDMS_array structure.
- *
- * @param dms The OBIDMS to which the obiarray belongs.
- * @param array_name The name of the obiarray.
- *
- * @returns A pointer to the newly created obiarray structure.
- * @retval NULL if an error occurred.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-OBIDMS_array_p obi_create_array(OBIDMS_p dms, const char* array_name);
-
-
-/**
- * @brief Opens an obiarray. Fails if it does not already exist.
- *
- * Note: An obiarray is made of two files (referred to by two structures).
- * 		 One file contains the indices referring to the data, and the other
- * 		 file contains the data itself. The obiarray as a whole is referred
- * 		 to via the OBIDMS_array structure.
- *
- * @param dms The OBIDMS to which the obiarray belongs.
- * @param array_name The name of the obiarray.
- *
- * @returns A pointer to the obiarray structure.
- * @retval NULL if an error occurred.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-OBIDMS_array_p obi_open_array(OBIDMS_p dms, const char* array_name);
-
-
-/**
- * @brief Closes an obiarray.
- *
- * Note: An obiarray is made of two files (referred to by two structures).
- * 		 One file contains the indices referring to the data, and the other
- * 		 file contains the data itself. The obiarray as a whole is referred
- * 		 to via the OBIDMS_array structure.
- *
- * @param array A pointer to the obiarray structure to close and free.
- *
- * @retval 0 if the operation was successfully completed.
- * @retval -1 if an error occurred.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-int obi_close_array(OBIDMS_array_p array);
-
-
-/**
- * @brief Adds a value (byte array) in an obiarray, checking first if it is already in it.
- *
- * @warning The byte array to add must already be encoded and contain its header.
- *
- * @param array A pointer to the obiarray.
- * @param value The byte array to add in the obiarray.
- *
- * @returns The index of the value, whether it was added or already in the obiarray.
- * @retval -1 if an error occurred.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-index_t obi_array_add(OBIDMS_array_p array, byte_t* value);
-
-
-/**
- * @brief Recovers a value (byte array) in an obiarray.
- *
- * @warning The byte array recovered is encoded and contains its header.
- *
- * @param array A pointer to the obiarray.
- * @param index The index of the value in the data array.
- *
- * @returns A pointer to the byte array recovered.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-byte_t* obi_array_get(OBIDMS_array_p array, index_t index);
-
-
-/**
- * @brief Searches a value (byte array) in an obiarray performing a binary search.
- *
- * @warning The byte array to search must already be encoded and contain its header.
- *
- * @param array A pointer to the obiarray.
- * @param value The byte array to add in the obiarray.
- *
- * @returns If the value is found, its data index is returned.
- * 			If the value is not found, the array index indicating where the value's data index
- * 			should be in the array is returned in the form (- (index + 1)), as data indices in an
- * 			obiarray are sorted according to the ascending order of the values (byte arrays) themselves.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-index_t obi_array_search(OBIDMS_array_p array, byte_t* value);
-
-
-/**
- * @brief Converts a character string to a byte array with a header.
- *
- * @warning The byte array must be freed by the caller.
- *
- * @param value The character string to convert.
- *
- * @returns A pointer to the byte array created.
- * @retval NULL if an error occurred.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-byte_t* obi_str_to_obibytes(char* value);
-
-
-/**
- * @brief Converts a byte array to a character string.
- *
- * @param value_b The byte array to convert.
- *
- * @returns A pointer to the character string contained in the byte array.
- *
- * @since October 2015
- * @author Celine Mercier (celine.mercier@metabarcoding.org)
- */
-const char* obi_obibytes_to_str(byte_t* value_b);
-
-
-#endif /* OBIARRAY_H_ */
-

src/obiavl.h

@@ -0,0 +1,350 @@
+/****************************************************************************
+ * OBIDMS AVL tree header file	                                            *
+ ****************************************************************************/
+
+/**
+ * @file obiavl.h
+ * @author Celine Mercier
+ * @date December 3rd 2015
+ * @brief Header file for handling AVL trees for storing and retrieving byte arrays (i.e. coding for character strings).
+ */
+
+
+#ifndef OBIAVL_H_
+#define OBIAVL_H_
+
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <time.h>
+#include <sys/types.h>
+#include <dirent.h>
+#include <stdbool.h>
+
+#include "obidms.h"
+#include "obitypes.h"
+
+
+#define AVL_MAX_NAME (1024) 								/**< The maximum length of an AVL tree name.
+                              	  	 	 	 	 	 	 	 */
+#define AVL_GROWTH_FACTOR (2)								/**< The growth factor when an AVL tree is enlarged.
+                                	 	 	 	 	 	 	 */
+#define AVL_MAX_DEPTH (1000)								/**< The maximum depth of an AVL tree.
+                                	 	 	 	 	 	 	 */
+#define LEFT_CHILD(node)	(avl->tree)+(node->left_child)  /**< Pointer to the left child of a node in an AVL tree.
+															 */
+#define RIGHT_CHILD(node)	(avl->tree)+(node->right_child) /**< Pointer to the right child of a node in an AVL tree.
+															 */
+#define BYTE_ARRAY_HEADER_SIZE (9)  						/**< The size of the header of a byte array.
+                              	  	 	 	 	 	 	 	 */
+
+
+/**
+ * @brief AVL tree node structure.
+ */
+typedef struct AVL_node {
+	index_t left_child;		/**< Index of left less child node.
+	  	 	 	 	 	 	 */
+	index_t right_child; 	/**< Index of right greater child node.
+	 	 	 	 	 	 	 */
+	int8_t balance_factor; /**< Balance factor of the node.
+							 */
+	index_t value;			/**< Index of the value associated with the node in the data array.
+							 */
+} AVL_node_t, *AVL_node_p;
+
+
+/**
+ * @brief OBIDMS AVL tree data header structure.
+ */
+typedef struct OBIDMS_avl_data_header {
+	int			header_size;						/**< Size of the header in bytes.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	index_t		data_size_used;						/**< Size of the data used in bytes.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	index_t		data_size_max;						/**< Max size of the data in bytes.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	index_t		nb_items;							/**< Number of items.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	char    	avl_name[AVL_MAX_NAME+1];			/**< The AVL tree name as a NULL terminated string.
+													 */
+	time_t		creation_date;			    	    /**< Date of creation of the file.
+												     */
+} OBIDMS_avl_data_header_t, *OBIDMS_avl_data_header_p;
+
+
+/**
+ * @brief OBIDMS AVL tree data structure.
+ */
+typedef struct OBIDMS_avl_data {
+	OBIDMS_avl_data_header_p	header; 		 	/**< A pointer to the header of the AVL tree data.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	byte_t*                   	data;   		 	/**< A pointer to the beginning of the data.
+													 */
+} OBIDMS_avl_data_t, *OBIDMS_avl_data_p;
+
+
+/**
+ * @brief OBIDMS AVL tree header structure.
+ */
+typedef struct OBIDMS_avl_header {
+	int			header_size;						/**< Size of the header in bytes.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	size_t		avl_size;							/**< Size of the AVL tree in bytes.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	index_t		nb_items;							/**< Number of items in the AVL tree.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	index_t		nb_items_max;						/**< Maximum number of items in the AVL tree before it has to be enlarged.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	index_t		root_idx;							/**< Index of the root of the AVL tree.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	char    	avl_name[AVL_MAX_NAME+1];			/**< The AVL tree name as a NULL terminated string.
+													 */
+	time_t		creation_date;			    	    /**< Date of creation of the file.
+												     */
+} OBIDMS_avl_header_t, *OBIDMS_avl_header_p;
+
+
+/**
+ * @brief OBIDMS AVL tree structure.
+ */
+typedef struct OBIDMS_avl {
+	OBIDMS_p                	dms;			 	/**< A pointer to the OBIDMS structure to which the AVL tree belongs.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	OBIDMS_avl_header_p			header; 	 		/**< A pointer to the header of the AVL tree.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	struct AVL_node*           	tree;   		 	/**< A pointer to the root of the AVL tree.
+													 */
+	index_t                     path_idx[AVL_MAX_DEPTH];	/**< The path taken to a node from the root as an array of node indices.
+													 	     */
+	int8_t                      path_dir[AVL_MAX_DEPTH];	/**< The path taken to a node from the root as an array of directions
+															 * (0 for left, -1 for right).
+															 */
+	OBIDMS_avl_data_p			data;				/**< A pointer to the structure containing the data
+													 *   that the AVL tree references.
+	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	DIR*						directory;			/**< A directory entry usable to
+										 	 	 	 *    refer and scan the AVL tree directory.
+										 	 	 	 */
+	int							dir_fd;				/**< The file descriptor of the directory entry
+													 *    usable to refer and scan the AVL tree directory.
+													 */
+	int							avl_fd;				/**< The file descriptor of the file
+													 *    to refer and scan the AVL tree.
+													 */
+	int							data_fd;				/**< The file descriptor of the file
+													 *    to refer and scan the data referred to by the AVL tree.
+													 */
+	size_t						counter;			/**< Indicates by how many threads/programs (TODO) the AVL tree is used.
+													 */
+} OBIDMS_avl_t, *OBIDMS_avl_p;
+
+
+/**
+ * @brief Checks if an AVL tree already exists or not.
+ *
+ * @param dms The OBIDMS to which the AVL tree belongs.
+ * @param avl_name The name of the AVL tree.
+ *
+ * @returns A value indicating whether the AVL tree exists or not.
+ * @retval 1 if the AVL tree exists.
+ * @retval 0 if the AVL tree does not exist.
+ * @retval -1 if an error occurred.
+ *
+ * @since December 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_avl_exists(OBIDMS_p dms, const char* avl_name);
+
+
+/**
+ * @brief Opens an AVL tree and creates it if it does not already exist.
+ *
+ * Note: An AVL tree is made of two files (referred to by two structures).
+ * 		 One file contains the indices referring to the data, and the other
+ * 		 file contains the data itself. The AVL tree as a whole is referred
+ * 		 to via the OBIDMS_avl structure.
+ *
+ * @param dms The OBIDMS to which the AVL tree belongs.
+ * @param avl_name The name of the AVL tree.
+ *
+ * @returns A pointer to the AVL tree structure.
+ * @retval NULL if an error occurred.
+ *
+ * @since December 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+OBIDMS_avl_p obi_avl(OBIDMS_p dms, const char* avl_name);
+
+
+/**
+ * @brief Creates an AVL tree. Fails if it already exists.
+ *
+ * Note: An AVL tree is made of two files (referred to by two structures).
+ * 		 One file contains the indices referring to the data, and the other
+ * 		 file contains the data itself. The AVL tree as a whole is referred
+ * 		 to via the OBIDMS_avl structure.
+ *
+ * @param dms The OBIDMS to which the AVL tree belongs.
+ * @param avl_name The name of the AVL tree.
+ *
+ * @returns A pointer to the newly created AVL tree structure.
+ * @retval NULL if an error occurred.
+ *
+ * @since December 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+OBIDMS_avl_p obi_create_avl(OBIDMS_p dms, const char* avl_name);
+
+
+/**
+ * @brief Opens an AVL tree. Fails if it does not already exist.
+ *
+ * Note: An AVL tree is made of two files (referred to by two structures).
+ * 		 One file contains the indices referring to the data, and the other
+ * 		 file contains the data itself. The AVL tree as a whole is referred
+ * 		 to via the OBIDMS_avl structure.
+ *
+ * @param dms The OBIDMS to which the AVL tree belongs.
+ * @param avl_name The name of the AVL tree.
+ *
+ * @returns A pointer to the AVL tree structure.
+ * @retval NULL if an error occurred.
+ *
+ * @since December 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+OBIDMS_avl_p obi_open_avl(OBIDMS_p dms, const char* avl_name);
+
+
+/**
+ * @brief Closes an AVL tree.
+ *
+ * Note: An AVL tree is made of two files (referred to by two structures).
+ * 		 One file contains the indices referring to the data, and the other
+ * 		 file contains the data itself. The AVL tree as a whole is referred
+ * 		 to via the OBIDMS_avl structure.
+ *
+ * @param avl A pointer to the AVL tree structure to close and free.
+ *
+ * @retval 0 if the operation was successfully completed.
+ * @retval -1 if an error occurred.
+ *
+ * @since December 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_close_avl(OBIDMS_avl_p avl);
+
+
+/**
+ * @brief Adds a value (byte array) in an AVL tree, checking if it is already in it.
+ *
+ * @warning The byte array to add must already be encoded and contain its header.
+ *
+ * @param avl A pointer to the AVL tree.
+ * @param value The byte array to add in the AVL tree.
+ *
+ * @returns The index of the value, whether it was added or already in the AVL tree.
+ * @retval -1 if an error occurred.
+ *
+ * @since December 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+index_t obi_avl_add(OBIDMS_avl_p avl, byte_t* value);
+
+
+/**
+ * @brief Finds a value (byte array) in an AVL tree, checking first if it is already in it.
+ *
+ * @warning The byte array to add must already be encoded and contain its header.
+ *
+ * @param avl A pointer to the AVL tree.
+ * @param value The byte array to add in the AVL tree.
+ *
+ * @returns The data index of the value.
+ * @retval -1 if the value is not in the tree.
+ *
+ * @since December 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+index_t obi_avl_find(OBIDMS_avl_p avl, byte_t* value);
+
+
+/**
+ * @brief Recovers a value (byte array) in an AVL tree.
+ *
+ * @warning The byte array recovered is encoded and contains its header.
+ *
+ * @param avl A pointer to the AVL tree.
+ * @param index The index of the value in the data array.
+ *
+ * @returns A pointer to the byte array recovered.
+ *
+ * @since December 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+byte_t* obi_avl_get(OBIDMS_avl_p avl, index_t index);
+
+
+/**
+ * @brief Converts a character string to a byte array with a header.
+ *
+ * @warning The byte array must be freed by the caller.
+ *
+ * @param value The character string to convert.
+ *
+ * @returns A pointer to the byte array created.
+ * @retval NULL if an error occurred.
+ *
+ * @since October 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+byte_t* obi_str_to_obibytes(char* value);
+
+
+/**
+ * @brief Converts a byte array to a character string.
+ *
+ * @param value_b The byte array to convert.
+ *
+ * @returns A pointer to the character string contained in the byte array.
+ *
+ * @since October 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+const char* obi_obibytes_to_str(byte_t* value_b);
+
+
+/**
+ * @brief Converts a DNA sequence to a byte array with a header.
+ *
+ * @warning The byte array must be freed by the caller.
+ *
+ * @param value The DNA sequence to convert.
+ *
+ * @returns A pointer to the byte array created.
+ * @retval NULL if an error occurred.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+byte_t* obi_seq_to_obibytes(char* seq);
+
+
+/**
+ * @brief Converts a byte array to a DNA sequence.
+ *
+ * @param value_b The byte array to convert.
+ *
+ * @returns A pointer to the DNA sequence contained in the byte array.
+ * @retval NULL if an error occurred.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+const char* obi_obibytes_to_seq(byte_t* value_b);
+
+
+#endif /* OBIAVL_H_ */
+

src/obidms.c

@@ -13,6 +13,7 @@
 #include <stdlib.h>
 #include <stdio.h>
 #include <string.h>
+#include <stdbool.h>
 #include <sys/stat.h>
 #include <fcntl.h>
 #include <sys/types.h>
@@ -23,6 +24,7 @@
 #include "obidebug.h"
 #include "obidmscolumn.h"
 #include "private_at_functions.h"
+#include "obilittlebigman.h"
 
 
 #define DEBUG_LEVEL 0	// TODO has to be defined somewhere else (cython compil flag?)
@@ -54,6 +56,44 @@
 static char* build_directory_name(const char* dms_name);
 
 
+/**
+ * Internal function building the informations file name from an OBIDMS name.
+ *
+ * The function builds the file name for the informations file of an OBIDMS.
+ *
+ * @warning The returned pointer has to be freed by the caller.
+ *
+ * @param dms_name The name of the OBIDMS.
+ *
+ * @returns A pointer to the file name.
+ * @retval NULL if an error occurred.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+static char* build_infos_file_name(const char* dms_name);
+
+
+/**
+ * Internal function creating the file containing basic informations on the OBIDMS.
+ *
+ * This file contains:
+ * 		- The endianness of the platform
+ *
+ * @warning The returned pointer has to be freed by the caller.
+ *
+ * @param dms_file_descriptor The file descriptor for the OBIDMS directory.
+ * @param dms_name The name of the OBIDMS.
+ *
+ * @retval 0 if the operation was successfully completed.
+ * @retval -1 if an error occurred.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int create_dms_infos_file(int dms_file_descriptor, const char* dms_name);
+
+
 /************************************************************************
  *
  * D E F I N I T I O N   O F   T H E   P R I V A T E   F U N C T I O N S
@@ -85,6 +125,74 @@ static char* build_directory_name(const char* dms_name)
 }
 
 
+static char* build_infos_file_name(const char* dms_name)
+{
+	char* file_name;
+
+	// Build file name
+	if (asprintf(&file_name, "%s_infos", dms_name) < 0)
+	{
+		obi_set_errno(OBIDMS_MEMORY_ERROR);
+		obidebug(1, "\nProblem building an informations file name");
+		return NULL;
+	}
+
+	return file_name;
+}
+
+
+int create_dms_infos_file(int dms_file_descriptor, const char* dms_name)
+{
+	char* 	file_name;
+	int 	infos_file_descriptor;
+	off_t 	file_size;
+	bool	little_endian;
+
+	file_size = sizeof(bool);
+
+	// Create file name
+	file_name = build_infos_file_name(dms_name);
+	if (file_name == NULL)
+		return -1;
+
+    // Create file
+	infos_file_descriptor = openat(dms_file_descriptor, file_name, O_RDWR | O_CREAT | O_EXCL, 0777);
+	if (infos_file_descriptor < 0)
+	{
+		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
+		obidebug(1, "\nError creating an informations file");
+		free(file_name);
+		return -1;
+	}
+
+	free(file_name);
+
+	// Truncate the infos file to the right size
+	if (ftruncate(infos_file_descriptor, file_size) < 0)
+	{
+		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
+		obidebug(1, "\nError truncating an informations file");
+		close(infos_file_descriptor);
+		return -1;
+	}
+
+	// Write endianness
+	little_endian = obi_is_little_endian();
+	if (write(infos_file_descriptor, &little_endian, sizeof(bool)) < ((ssize_t) sizeof(bool)))
+	{
+		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
+		obidebug(1, "\nError writing the endianness in an informations file");
+		close(infos_file_descriptor);
+		return -1;
+	}
+
+	// Close file
+	close(infos_file_descriptor);
+
+	return 0;
+}
+
+
 /**********************************************************************
  *
  * D E F I N I T I O N   O F   T H E   P U B L I C   F U N C T I O N S
@@ -94,8 +202,8 @@ static char* build_directory_name(const char* dms_name)
 int obi_dms_exists(const char* dms_name)
 {
     struct stat buffer;
-	char *directory_name;
-    int check_dir;
+	char*       directory_name;
+    int 		check_dir;
 
 	// Build and check the directory name
     directory_name = build_directory_name(dms_name);
@@ -116,8 +224,8 @@ int obi_dms_exists(const char* dms_name)
 OBIDMS_p obi_create_dms(const char* dms_name)
 {
 	char* directory_name;
-	DIR* dms_dir;
-	int dms_file_descriptor;
+	DIR*  dms_dir;
+	int   dms_file_descriptor;
 
 	// Build and check the directory name
 	directory_name = build_directory_name(dms_name);
@@ -128,7 +236,10 @@ OBIDMS_p obi_create_dms(const char* dms_name)
 	if (mkdir(directory_name, 00777) < 0)
 	{
 		if (errno == EEXIST)
+		{
 			obi_set_errno(OBIDMS_EXIST_ERROR);
+			obidebug(1, "\nAn OBIDMS directory with the same name already exists in this directory.");
+		}
 		else
 			obi_set_errno(OBIDMS_UNKNOWN_ERROR);
 		obidebug(1, "\nProblem creating an OBIDMS directory");
@@ -136,7 +247,7 @@ OBIDMS_p obi_create_dms(const char* dms_name)
 		return NULL;
 	}
 
-	// Get file descriptor of DMS directory to create the arrays directory
+	// Get file descriptor of DMS directory to create the AVL trees directory
 	dms_dir = opendir(directory_name);
 	if (dms_dir == NULL)
 	{
@@ -145,25 +256,28 @@ OBIDMS_p obi_create_dms(const char* dms_name)
 		free(directory_name);
 		return NULL;
 	}
+
+	free(directory_name);
+
 	dms_file_descriptor = dirfd(dms_dir);
 	if (dms_file_descriptor < 0)
 	{
 		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
 		obidebug(1, "\nProblem getting the file descriptor of a newly created OBIDMS directory");
-		free(directory_name);
 		return NULL;
 	}
 
-	// Create the arrays directory
-	if (mkdirat(dms_file_descriptor, ARRAY_DIR_NAME, 00777) < 0)
+	// Create the AVL trees directory
+	if (mkdirat(dms_file_descriptor, AVL_TREES_DIR_NAME, 00777) < 0)
 	{
-		obi_set_errno(OBI_ARRAY_ERROR);
-		obidebug(1, "\nProblem creating an arrays directory");
-		free(directory_name);
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nProblem creating an AVL trees directory");
 		return NULL;
 	}
 
-	free(directory_name);
+	// Create the informations file
+	if (create_dms_infos_file(dms_file_descriptor, dms_name) < 0)
+		return NULL;
 
 	return obi_open_dms(dms_name);
 }
@@ -173,19 +287,36 @@ OBIDMS_p obi_open_dms(const char* dms_name)
 {
 	OBIDMS_p	dms;
 	char*	 	directory_name;
-	DIR*		directory;
-	int			dms_file_descriptor;
+	char* 		infos_file_name;
+	int			infos_file_descriptor;
+	bool		little_endian_dms;
+	bool 		little_endian_platform;
 
 	dms = NULL;
 
+	// Allocate the data structure
+	dms = (OBIDMS_p) malloc(sizeof(OBIDMS_t));
+	if (dms == NULL)
+	{
+		obi_set_errno(OBIDMS_MEMORY_ERROR);
+		obidebug(1, "\nError allocating the memory for the OBIDMS structure");
+		return NULL;
+	}
+
 	// Build and check the directory name
 	directory_name = build_directory_name(dms_name);
 	if (directory_name == NULL)
+	{
+		free(dms);
 		return NULL;
+	}
+
+	strncpy(dms->directory_name, directory_name, OBIDMS_MAX_NAME);
+	free(directory_name);
 
 	// Try to open the directory
-	directory = opendir(directory_name);
-	if (directory == NULL)
+	dms->directory = opendir(dms->directory_name);
+	if (dms->directory == NULL)
 	{
 		switch (errno)
 		{
@@ -205,45 +336,92 @@ OBIDMS_p obi_open_dms(const char* dms_name)
 				obi_set_errno(OBIDMS_UNKNOWN_ERROR);
 		}
 		obidebug(1, "\nCan't open OBIDMS directory");
-		free(directory_name);
+		free(dms);
 		return NULL;
 	}
 
-	// Allocate the data structure
-	dms = (OBIDMS_p) malloc(sizeof(OBIDMS_t));
-	if (dms == NULL)
-	{
-		obi_set_errno(OBIDMS_MEMORY_ERROR);
-		obidebug(1, "\nError allocating the memory for the OBIDMS structure");
-		free(directory_name);
-		return NULL;
-	}
-
-	// Initialize the data structure
-	strcpy(dms->directory_name, directory_name);
-	dms->directory = directory;
-
-	// Get file descriptor of DMS directory to open the arrays directory
-	dms_file_descriptor = dirfd(directory);
-	if (dms_file_descriptor < 0)
+	// Get and store file descriptor of DMS directory to open the informations file
+	dms->dir_fd = dirfd(dms->directory);
+	if (dms->dir_fd < 0)
 	{
 		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
 		obidebug(1, "\nError getting the file descriptor for a newly created OBIDMS directory");
-		free(directory_name);
+		closedir(dms->directory);
+		free(dms);
 		return NULL;
 	}
 
-	// Open the arrays directory
-	dms->array_directory = private_opendirat(dms_file_descriptor, ARRAY_DIR_NAME);
-	if (dms->array_directory == NULL)
+	// Open informations file to check endianness
+	infos_file_name = build_infos_file_name(dms_name);
+	infos_file_descriptor = openat(dms->dir_fd, infos_file_name, O_RDONLY, 0777);
+	if (infos_file_descriptor < 0)
 	{
 		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
-		obidebug(1, "\nError opening the arrays directory");
-		free(directory_name);
+		obidebug(1, "\nError opening an informations file");
+		closedir(dms->directory);
+		free(dms);
 		return NULL;
 	}
 
-	free(directory_name);
+	free(infos_file_name);
+
+	// Check endianness of the platform and DMS
+	little_endian_platform = obi_is_little_endian();
+	if (read(infos_file_descriptor, &little_endian_dms, sizeof(bool)) < ((ssize_t) sizeof(bool)))
+	{
+		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
+		obidebug(1, "\nError reading the endianness in an informations file");
+		close(infos_file_descriptor);
+		closedir(dms->directory);
+		free(dms);
+		return NULL;
+	}
+	if (little_endian_platform != little_endian_dms)
+	{
+		obi_set_errno(OBIDMS_BAD_ENDIAN_ERROR);
+		obidebug(1, "\nError: The DMS and the platform have different endianness");
+		close(infos_file_descriptor);
+		closedir(dms->directory);
+		free(dms);
+		return NULL;
+	}
+
+	close(infos_file_descriptor);
+
+	dms->little_endian = little_endian_dms;
+
+	// Open the AVL trees directory
+	dms->avl_directory = private_opendirat(dms->dir_fd, AVL_TREES_DIR_NAME);
+	if (dms->avl_directory == NULL)
+	{
+		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
+		obidebug(1, "\nError opening the AVL trees directory");
+		closedir(dms->directory);
+		free(dms);
+		return NULL;
+	}
+
+	// Store the AVL trees directory's file descriptor
+	dms->avl_dir_fd = dirfd(dms->avl_directory);
+	if (dms->avl_dir_fd < 0)
+	{
+		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
+		obidebug(1, "\nError getting the file descriptor of the AVL trees directory");
+		closedir(dms->avl_directory);
+		closedir(dms->directory);
+		free(dms);
+		return NULL;
+	}
+
+	// Initialize the list of opened columns
+	dms->opened_columns = (Opened_columns_list_p) malloc(sizeof(Opened_columns_list_t));
+	(dms->opened_columns)->columns = (OBIDMS_column_p*) malloc(MAX_NB_OPENED_COLUMNS*sizeof(OBIDMS_column_p));
+	(dms->opened_columns)->nb_opened_columns = 0;
+
+	// Initialize the list of opened AVL trees
+	dms->opened_avls = (Opened_avls_list_p) malloc(sizeof(Opened_avls_list_t));
+	(dms->opened_avls)->avls = (OBIDMS_avl_p*) malloc(MAX_NB_OPENED_AVL_TREES*sizeof(OBIDMS_avl_p));
+	(dms->opened_avls)->nb_opened_avls = 0;
 
 	return dms;
 }
@@ -272,6 +450,11 @@ int obi_close_dms(OBIDMS_p dms)
 {
 	if (dms != NULL)
 	{
+		// Close all columns
+		while ((dms->opened_columns)->nb_opened_columns > 0)
+			obi_close_column(*((dms->opened_columns)->columns));
+
+		// Close dms and AVL trees directories
 		if (closedir(dms->directory) < 0)
 		{
 			obi_set_errno(OBIDMS_MEMORY_ERROR);
@@ -279,10 +462,10 @@ int obi_close_dms(OBIDMS_p dms)
 			free(dms);
 			return -1;
 		}
-		if (closedir(dms->array_directory) < 0)
+		if (closedir(dms->avl_directory) < 0)
 		{
-			obi_set_errno(OBI_ARRAY_ERROR);
-			obidebug(1, "\nError closing an array directory");
+			obi_set_errno(OBI_AVL_ERROR);
+			obidebug(1, "\nError closing an AVL trees directory");
 			free(dms);
 			return -1;
 		}

src/obidms.h

@@ -20,14 +20,37 @@
 #include <dirent.h>
 #include <string.h>
 #include <stdio.h>
+#include <stdbool.h>
 
 #include "obierrno.h"
 
 
-#define OBIDMS_MAX_NAME (2048) 	   /**< The maximum length of an OBIDMS name.
-                                	*/
-#define ARRAY_DIR_NAME "arrays"	   /**< The name of the arrays directory.
-                                	*/
+#define OBIDMS_MAX_NAME (2048) 	    		/**< The maximum length of an OBIDMS name.
+                                	 	 	 */
+#define AVL_TREES_DIR_NAME "AVL_trees"    	/**< The name of the AVL trees directory.
+                                	 	 	 */
+#define TAXONOMY_DIR_NAME "TAXONOMY"    	/**< The name of the taxonomy directory.
+                                	 	 	 */
+#define MAX_NB_OPENED_COLUMNS (100) 		/**< The maximum number of columns open at the same time.
+                                	 	 	 */
+#define MAX_NB_OPENED_AVL_TREES (100)  		/**< The maximum number of AVL trees open at the same time.
+                                	 	 	 */
+
+
+struct OBIDMS_column;		// TODO
+
+typedef struct Opened_columns_list {
+	size_t	 		 		nb_opened_columns;
+	struct OBIDMS_column** 	columns;				// TODO allocate array size here
+} Opened_columns_list_t, *Opened_columns_list_p;
+
+
+struct OBIDMS_avl;		// TODO
+
+typedef struct Opened_avls_list {
+	size_t	 		 		nb_opened_avls;
+	struct OBIDMS_avl** 	avls;
+} Opened_avls_list_t, *Opened_avls_list_p;
 
 
 /**
@@ -37,15 +60,27 @@
  * and opening of an OBITools Data Management System (DMS)
  */
 typedef struct OBIDMS {
-	char	directory_name[OBIDMS_MAX_NAME+1];	/**< The name of the directory
-	                                     	 	 *   containing the DMS.
-	                                     	 	 */
-	DIR*	directory;							/**< A directory entry usable to
-										 	 	 *   refer and scan the database directory.
-										 	 	 */
-	DIR*	array_directory;					/**< A directory entry usable to
-										 	 	 *   refer and scan the array directory.
-										 	 	 */
+	char					directory_name[OBIDMS_MAX_NAME+1];	/**< The name of the directory
+																 *   containing the DMS.
+																 */
+	DIR*					directory;							/**< A directory entry usable to
+																 *   refer and scan the database directory.
+																 */
+	int     				dir_fd;								/**< The file descriptor of the directory entry
+																 *   usable to refer and scan the database directory.
+																 */
+	DIR*					avl_directory;						/**< A directory entry usable to
+																 *   refer and scan the AVL trees directory.
+																 */
+	int						avl_dir_fd;							/**< The file descriptor of the directory entry
+																 *   usable to refer and scan the AVL trees directory.
+																 */
+	bool					little_endian;						/**< Endianness of the database.
+	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	Opened_columns_list_p 	opened_columns;						/**< List of opened columns.
+	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	Opened_avls_list_p 		opened_avls;						/**< List of opened AVL trees.
+	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
 } OBIDMS_t, *OBIDMS_p;
 
 
@@ -72,7 +107,7 @@ int obi_dms_exists(const char* dms_name);
  * if a directory with this name does not already exist
  * before creating the new database.
  *
- * A directory to store obiarrays is also created.
+ * A directory to store AVL trees is also created.
  *
  * @param dms_name A pointer to a C string containing the name of the database.
  *                 The actual directory name used to store the DMS will be

src/obidms_taxonomy.c

@@ -0,0 +1,681 @@
+/********************************************************************
+ * OBIDMS taxonomy functions                                        *
+ ********************************************************************/
+
+/**
+ * @file obidms_taxonomy.c
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ * @date March 2nd 2016
+ * @brief Functions for reading binary taxonomy files.
+ */
+
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <stdbool.h>
+#include <fcntl.h>
+
+#include "obidms_taxonomy.h"
+#include "obidms.h"
+#include "obidebug.h"
+#include "obierrno.h"
+#include "private_at_functions.h"
+
+
+#define DEBUG_LEVEL 0	// TODO has to be defined somewhere else (cython compil flag?)
+
+// TODO : the malloc aren't checked but won't exist for long because mapping instead
+
+
+int compareRankLabel(const void *label1, const void *label2)
+{
+	return strcmp((const char*)label1,*(const char**)label2);
+}
+
+
+int32_t rank_index(const char* label, ecorankidx_t* ranks)
+{
+	char **rep;
+
+	rep = bsearch(label, ranks->label, ranks->count, sizeof(char*), compareRankLabel);
+
+	if (rep)
+		return rep-ranks->label;	// TODO what???
+
+	return -1;
+}
+
+
+void* read_ecorecord(FILE* f, int32_t* record_size)
+{
+	static void* buffer = NULL;
+	int32_t      buffer_size = 0;
+	int32_t      read;
+
+	if (!record_size)
+	{
+ 		obi_set_errno(OBI_TAXONOMY_ERROR);
+ 		obidebug(1, "\nError reading a taxonomy file: record_size can not be NULL");
+ 		return NULL;
+	}
+
+	read = fread(record_size,
+	      		 1,
+	      		 sizeof(int32_t),
+	             f);
+
+	if (feof(f))
+		return NULL;
+
+	if (read != sizeof(int32_t))
+	{
+ 		obi_set_errno(OBI_TAXONOMY_ERROR);
+ 		obidebug(1, "\nError reading a taxonomy file: error reading record size");
+ 		return NULL;
+	}
+
+//	if (is_big_endian())						// TODO
+//		*recordSize=swap_int32_t(*recordSize);
+
+	if (buffer_size < *record_size)
+	{
+		if (buffer)
+			buffer = realloc(buffer, *record_size);
+		else
+			buffer = malloc(*record_size);
+		if (buffer == NULL)
+		{
+	 		obi_set_errno(OBI_TAXONOMY_ERROR);
+	 		obidebug(1, "\nError reading a taxonomy file: error allocating memory");
+	 		return NULL;
+		}
+	}
+
+	read = fread(buffer,
+	             1,
+				 *record_size,
+				 f);
+
+	if (read != *record_size)
+	{
+ 		obi_set_errno(OBI_TAXONOMY_ERROR);
+ 		obidebug(1, "\nError reading a taxonomy file: error reading a record %d, %d", read, *record_size);
+ 		return NULL;
+	}
+
+	return buffer;
+};
+
+
+ecotx_t* readnext_ecotaxon(FILE* f, ecotx_t* taxon)
+{
+	ecotxformat_t* raw;
+	int32_t        record_length;
+
+	raw = read_ecorecord(f, &record_length);
+
+	if (!raw)
+		return NULL;
+
+//	if (is_big_endian())		// TODO
+//	{
+//		raw->namelength = swap_int32_t(raw->namelength);
+//		raw->parent     = swap_int32_t(raw->parent);
+//		raw->rank       = swap_int32_t(raw->rank);
+//		raw->taxid      = swap_int32_t(raw->taxid);
+//	}
+
+	taxon->parent = (ecotx_t*) ((size_t) raw->parent);
+	taxon->taxid  = raw->taxid;
+	taxon->rank   = raw->rank;
+	taxon->farest = -1;
+
+	taxon->name   = malloc((raw->name_length+1) * sizeof(char));
+
+	strncpy(taxon->name, raw->name, raw->name_length);
+
+	return taxon;
+}
+
+
+FILE* open_ecorecorddb(const char* file_name,
+                       int32_t*    count,
+                       int32_t     abort_on_open_error)
+{
+    FILE*		f;
+	int32_t     read;
+
+	fprintf(stderr, "\n%s\n", file_name);
+
+	f = fopen(file_name, "rb");
+
+	if (!f)
+	{
+		if (abort_on_open_error)
+		{
+	 		obi_set_errno(OBI_TAXONOMY_ERROR);
+	 		obidebug(1, "\nCouldn't open a taxonomy file");
+	 		return NULL;
+		}
+	 	else
+	 	{
+	 		*count = 0;
+	 		return NULL;
+	 	}
+	}
+
+	read = fread(count,
+	      		 1,
+	      		 sizeof(int32_t),
+	      		 f);
+
+	if (read != sizeof(int32_t))
+	{
+ 		obi_set_errno(OBI_TAXONOMY_ERROR);
+ 		obidebug(1, "\nError reading taxonomy record size");
+ 		return NULL;
+	}
+
+//	if (!obi_is_little_endian())		// TODO
+//		*count = swap_int32_t(*count);
+
+	return f;
+}
+
+
+ecorankidx_t* read_rankidx(const char* ranks_file_name)
+{
+	int32_t      			count;
+	FILE*        			ranks_file;
+	ecorankidx_t*           ranks_index;
+	int32_t      			i;
+	int32_t      			rank_length;
+	char*					buffer;
+
+	ranks_file = open_ecorecorddb(ranks_file_name, &count, 0);
+
+	if (ranks_file==NULL)
+		return NULL;
+
+	ranks_index = (ecorankidx_t*) malloc(sizeof(ecorankidx_t) + sizeof(char*) * (count-1));
+
+	ranks_index->count = count;
+
+	for (i=0; i < count; i++)
+	{
+		buffer = read_ecorecord(ranks_file, &rank_length);
+		ranks_index->label[i] = (char*) malloc(rank_length+1);
+		strncpy(ranks_index->label[i], buffer, rank_length);
+	}
+
+	return ranks_index;
+}
+
+
+ecotxidx_t* read_taxonomyidx(const char* taxa_file_name, const char* local_taxa_file_name)
+{
+	int32_t      	  count_taxa;
+	int32_t      	  count_local_taxa;
+	FILE*			  f_taxa;
+	FILE* 			  f_local_taxa;
+	ecotxidx_t*		  taxa_index;
+	struct ecotxnode* t;
+	int32_t      	  i;
+	int32_t      	  j;
+
+	f_taxa = open_ecorecorddb(taxa_file_name, &count_taxa,0);
+
+	if (f_taxa == NULL)
+	{
+ 		obidebug(1, "\nError reading taxonomy taxa file");
+ 		return NULL;
+	}
+
+	f_local_taxa = open_ecorecorddb(local_taxa_file_name, &count_local_taxa, 0);
+
+	taxa_index = (ecotxidx_t*) malloc(sizeof(ecotxidx_t) + sizeof(ecotx_t) * (count_taxa + count_local_taxa - 1));
+
+	taxa_index->count = count_taxa + count_local_taxa;
+	taxa_index->buffer_size = taxa_index->count;
+
+	taxa_index->max_taxid = 0;
+	printf("Reading %d taxa...\n", count_taxa);
+	for (i=0; i<count_taxa; i++)
+	{
+		readnext_ecotaxon(f_taxa, &(taxa_index->taxon[i]));
+		taxa_index->taxon[i].parent = taxa_index->taxon + (size_t) taxa_index->taxon[i].parent;
+		taxa_index->taxon[i].parent->farest = 0;
+		if (taxa_index->taxon[i].taxid > taxa_index->max_taxid)
+			taxa_index->max_taxid = taxa_index->taxon[i].taxid;
+	}
+
+	if (count_local_taxa > 0)
+		printf("Reading %d local taxa...\n", count_local_taxa);
+	else
+		printf("No local taxa\n");
+
+	count_taxa = taxa_index->count;
+
+	for (; i < count_taxa; i++){
+		readnext_ecotaxon(f_local_taxa, &(taxa_index->taxon[i]));
+		taxa_index->taxon[i].parent = taxa_index->taxon + (size_t) taxa_index->taxon[i].parent;
+		taxa_index->taxon[i].parent->farest=0;
+		if (taxa_index->taxon[i].taxid > taxa_index->max_taxid)
+			taxa_index->max_taxid = taxa_index->taxon[i].taxid;
+	}
+
+	printf("Computing longest branches...\n");
+
+	for (i=0; i < count_taxa; i++)
+	{
+		t = taxa_index->taxon+i;
+		if (t->farest == -1)
+		{
+			t->farest=0;
+            while (t->parent != t)
+            {
+            	j = t->farest + 1;
+            	if (j > t->parent->farest)
+            	{
+            		t->parent->farest = j;
+            		t=t->parent;
+            	}
+            	else
+            		t = taxa_index->taxon;
+            }
+		}
+	}
+
+	return taxa_index;
+}
+
+
+econame_t* readnext_econame(FILE* f, econame_t* name, OBIDMS_taxonomy_p taxonomy)
+{
+
+	econameformat_t* raw;
+	int32_t record_length;
+
+	raw = read_ecorecord(f, &record_length);
+
+	if (!raw)
+		return NULL;
+
+//	if (is_big_endian())		// TODO
+//	{
+//		raw->is_scientificname 	= swap_int32_t(raw->is_scientificname);
+//		raw->namelength 	    = swap_int32_t(raw->namelength);
+//		raw->classlength        = swap_int32_t(raw->classlength);
+//		raw->taxid  	        = swap_int32_t(raw->taxid);
+//	}
+
+	name->is_scientific_name = raw->is_scientific_name;
+
+	name->name = malloc((raw->name_length + 1) * sizeof(char));
+	strncpy(name->name, raw->names, raw->name_length);
+	name->name[raw->name_length] = 0;
+
+	name->class_name = malloc((raw->class_length+1) * sizeof(char));
+	strncpy(name->class_name,(raw->names + raw->name_length), raw->class_length);
+	name->class_name[raw->class_length] = 0;
+
+	name->taxon = taxonomy->taxa->taxon + raw->taxid;
+
+	return name;
+}
+
+
+econameidx_t* read_nameidx(const char *file_name, OBIDMS_taxonomy_p taxonomy)
+{
+	int32_t      		count;
+	FILE*				f;
+	econameidx_t*		index_names;
+	int32_t      		i;
+
+	f = open_ecorecorddb(file_name, &count, 0);
+
+	if (f == NULL)
+		return NULL;
+
+	index_names = (econameidx_t*) malloc(sizeof(econameidx_t) + sizeof(econame_t) * (count-1));
+
+	index_names->count = count;
+
+	for (i=0; i < count; i++)
+		readnext_econame(f, (index_names->names)+i, taxonomy);
+
+	return index_names;
+}
+
+
+static int bcomptaxon (const void* ptaxid, const void* ptaxon)
+{
+  ecotx_t* current_taxon = (ecotx_t*) ptaxon;
+  int32_t  taxid = (int32_t) ((size_t) ptaxid);
+  return taxid - current_taxon->taxid;
+}
+
+
+/////// PUBLIC /////////
+
+
+OBIDMS_taxonomy_p obi_read_taxonomy(OBIDMS_p dms, const char* taxonomy_name, bool read_alternative_names)
+{
+	OBIDMS_taxonomy_p  tax;
+	char*			   main_taxonomy_dir_path;
+	char*			   taxonomy_path;
+	char*  			   ranks_file_name;
+	char*              taxa_file_name;
+	char*			   local_taxa_file_name;
+	char*			   alter_names_file_name;
+	int                buffer_size;
+
+	tax = (OBIDMS_taxonomy_p) malloc(sizeof(OBIDMS_taxonomy_t));
+
+	tax->ranks = NULL;
+	tax->taxa  = NULL;
+	tax->names = NULL;
+
+	buffer_size = 2048;		// TODO
+
+	main_taxonomy_dir_path = get_full_path(dms->dir_fd, TAXONOMY_DIR_NAME);
+	if (asprintf(&taxonomy_path, "%s/%s/%s", main_taxonomy_dir_path, taxonomy_name, taxonomy_name) < 0)
+	{
+		free(main_taxonomy_dir_path);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+
+	free(main_taxonomy_dir_path);
+
+	// Read ranks
+	ranks_file_name = (char*) malloc(buffer_size*sizeof(char));
+	if (ranks_file_name == NULL)
+	{
+		free(taxonomy_path);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+	if (snprintf(ranks_file_name, buffer_size, "%s.rdx", taxonomy_path) < 0)
+	{
+		free(taxonomy_path);
+		free(ranks_file_name);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+	tax->ranks = read_rankidx(ranks_file_name);
+	if (tax->ranks == NULL)
+	{
+		free(ranks_file_name);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+	free(ranks_file_name);
+
+	// Read taxa
+	taxa_file_name = (char*) malloc(buffer_size*sizeof(char));
+	if (taxa_file_name == NULL)
+	{
+		free(taxonomy_path);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+	if (snprintf(taxa_file_name, buffer_size,"%s.tdx", taxonomy_path) < 0)
+	{
+		free(taxonomy_path);
+		free(taxa_file_name);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+	local_taxa_file_name = (char*) malloc(buffer_size*sizeof(char));
+	if (local_taxa_file_name == NULL)
+	{
+		free(taxonomy_path);
+		free(taxa_file_name);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+	if (snprintf(local_taxa_file_name, buffer_size,"%s.ldx", taxonomy_path) < 0)
+	{
+		free(taxonomy_path);
+		free(taxa_file_name);
+		free(local_taxa_file_name);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+	tax->taxa = read_taxonomyidx(taxa_file_name, local_taxa_file_name);
+	if (tax->taxa == NULL)
+	{
+		free(taxonomy_path);
+		free(taxa_file_name);
+		free(local_taxa_file_name);
+		obi_close_taxonomy(tax);
+		return NULL;
+	}
+	free(taxa_file_name);
+	free(local_taxa_file_name);
+
+	// Read alternative names
+	if (read_alternative_names)
+	{
+		alter_names_file_name = (char*) malloc(buffer_size*sizeof(char));
+		if (alter_names_file_name == NULL)
+		{
+			free(taxonomy_path);
+			obi_close_taxonomy(tax);
+			return NULL;
+		}
+		if (snprintf(alter_names_file_name, buffer_size,"%s.ndx", taxonomy_path) < 0)
+		{
+			free(taxonomy_path);
+			free(alter_names_file_name);
+			obi_close_taxonomy(tax);
+			return NULL;
+		}
+        tax->names = read_nameidx(alter_names_file_name, tax);
+    	if (tax->names == NULL)
+    	{
+    		free(alter_names_file_name);
+    		obi_close_taxonomy(tax);
+    		return NULL;
+    	}
+        free(alter_names_file_name);
+	}
+
+	free(taxonomy_path);
+
+	return tax;
+}
+
+
+int obi_close_taxonomy(OBIDMS_taxonomy_p taxonomy)
+{
+	if (taxonomy)
+	{
+		if (taxonomy->ranks)
+			free(taxonomy->ranks);	// TODO those don't free everything but mapping will replace anyway
+
+		if (taxonomy->names)
+			free(taxonomy->names);
+
+		if (taxonomy->taxa)
+			free(taxonomy->taxa);
+
+		free(taxonomy);
+
+		return 0;
+	}
+
+	// TODO no closing files?
+
+	return 1;
+}
+
+
+//////////////////////////////////////////////////////////////////////////
+
+
+ecotx_t* obi_taxo_get_parent_at_rank(ecotx_t* taxon, int32_t rankidx)
+{
+	ecotx_t* current_taxon;
+	ecotx_t* next_taxon;
+
+	current_taxon = taxon;
+	next_taxon    = current_taxon->parent;
+
+	while ((current_taxon != next_taxon) &&  // root node
+		   (current_taxon->rank != rankidx))
+	{
+		current_taxon = next_taxon;
+		next_taxon    = current_taxon->parent;
+	}
+
+	if (current_taxon->rank == rankidx)
+		return current_taxon;
+	else
+		return NULL;
+}
+
+
+ecotx_t* obi_taxo_get_taxon_with_taxid(OBIDMS_taxonomy_p taxonomy, int32_t taxid)
+{
+	ecotx_t    *current_taxon;
+	int32_t     count;
+
+	count = taxonomy->taxa->count;
+
+	current_taxon = (ecotx_t*) bsearch((const void *) ((size_t) taxid),
+                                       (const void *) taxonomy->taxa->taxon,
+                                       count,
+                                       sizeof(ecotx_t),
+                                       bcomptaxon);
+	return current_taxon;
+}
+
+
+bool obi_taxo_is_taxon_under_taxid(ecotx_t* taxon, int32_t other_taxid)
+{
+	ecotx_t* next_parent;
+
+	next_parent = taxon->parent;
+
+	while ((other_taxid != next_parent->taxid) && (strcmp(next_parent->name, "root")))
+		next_parent = next_parent->parent;
+
+	if (other_taxid == next_parent->taxid)
+		return 1;
+	else
+		return 0;
+}
+
+
+ecotx_t* obi_taxo_get_species(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+{
+	static OBIDMS_taxonomy_p tax = NULL;
+	static int32_t		  rankindex = -1;
+
+	if (taxonomy && (tax != taxonomy))
+	{
+		rankindex = rank_index("species", taxonomy->ranks);
+		tax = taxonomy;
+	}
+
+	if (!tax || (rankindex < 0))
+	{
+		obi_set_errno(OBI_TAXONOMY_ERROR);
+		obidebug(1, "\nError trying to get the species associated with a taxon: No taxonomy defined");
+		return NULL;
+	}
+
+	return obi_taxo_get_parent_at_rank(taxon, rankindex);
+}
+
+
+ecotx_t* obi_taxo_get_genus(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+{
+	static OBIDMS_taxonomy_p tax = NULL;
+	static int32_t		  rankindex = -1;
+
+	if (taxonomy && (tax != taxonomy))
+	{
+		rankindex = rank_index("genus", taxonomy->ranks);
+		tax = taxonomy;
+	}
+
+	if (!tax || (rankindex < 0))
+	{
+		obi_set_errno(OBI_TAXONOMY_ERROR);
+		obidebug(1, "\nError trying to get the genus associated with a taxon: No taxonomy defined");
+		return NULL;
+	}
+
+	return obi_taxo_get_parent_at_rank(taxon, rankindex);
+}
+
+
+ecotx_t* obi_taxo_get_family(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+{
+	static OBIDMS_taxonomy_p tax = NULL;
+	static int32_t		  rankindex = -1;
+
+	if (taxonomy && (tax != taxonomy))
+	{
+		rankindex = rank_index("family", taxonomy->ranks);
+		tax = taxonomy;
+	}
+
+	if (!tax || (rankindex < 0))
+	{
+		obi_set_errno(OBI_TAXONOMY_ERROR);
+		obidebug(1, "\nError trying to get the family associated with a taxon: No taxonomy defined");
+		return NULL;
+	}
+
+	return obi_taxo_get_parent_at_rank(taxon, rankindex);
+}
+
+
+ecotx_t* obi_taxo_get_kingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+{
+	static OBIDMS_taxonomy_p tax = NULL;
+	static int32_t		  rankindex = -1;
+
+	if (taxonomy && (tax != taxonomy))
+	{
+		rankindex = rank_index("kingdom", taxonomy->ranks);
+		tax = taxonomy;
+	}
+
+	if (!tax || (rankindex < 0))
+	{
+		obi_set_errno(OBI_TAXONOMY_ERROR);
+		obidebug(1, "\nError trying to get the kingdom associated with a taxon: No taxonomy defined");
+		return NULL;
+	}
+
+	return obi_taxo_get_parent_at_rank(taxon, rankindex);
+}
+
+
+ecotx_t* obi_taxo_get_superkingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
+{
+	static OBIDMS_taxonomy_p tax = NULL;
+	static int32_t		  rankindex = -1;
+
+	if (taxonomy && (tax != taxonomy))
+	{
+		rankindex = rank_index("superkingdom", taxonomy->ranks);
+		tax = taxonomy;
+	}
+
+	if (!tax || (rankindex < 0))
+	{
+		obi_set_errno(OBI_TAXONOMY_ERROR);
+		obidebug(1, "\nError trying to get the superkingdom associated with a taxon: No taxonomy defined");
+		return NULL;
+	}
+
+	return obi_taxo_get_parent_at_rank(taxon, rankindex);
+}
+

src/obidms_taxonomy.h

@@ -0,0 +1,101 @@
+/********************************************************************
+ * OBIDMS taxonomy headeer file                                     *
+ ********************************************************************/
+
+/**
+ * @file obidms_taxonomy.h
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ * @date March 2nd 2016
+ * @brief Header file for the functions handling the reading of binary taxonomy files.
+ */
+
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdbool.h>
+
+#include "obidms.h"
+
+
+typedef struct {
+	int32_t  taxid;
+	int32_t  rank;
+	int32_t	 parent;
+	int32_t  name_length;
+	char     name[1];
+} ecotxformat_t;
+
+
+typedef struct ecotxnode {
+	int32_t           taxid;
+	int32_t           rank;
+	int32_t  		  farest;
+	struct ecotxnode* parent;
+	char*			  name;
+} ecotx_t;
+
+
+typedef struct {
+	int32_t count;
+	int32_t max_taxid;
+	int32_t buffer_size;
+	ecotx_t taxon[1];
+} ecotxidx_t;
+
+
+typedef struct {
+	int32_t count;
+	char*   label[1];
+} ecorankidx_t;
+
+
+typedef struct {
+ 	int32_t is_scientific_name;
+	int32_t  name_length;
+	int32_t  class_length;
+	int32_t  taxid;
+	char     names[1];
+} econameformat_t;
+
+
+typedef struct {
+	char*   		  name;
+	char*   		  class_name;
+	int32_t 		  is_scientific_name;
+	struct ecotxnode* taxon;
+} econame_t;
+
+
+typedef struct {
+	int32_t 	count;
+	econame_t   names[1];
+} econameidx_t;
+
+
+typedef struct OBIDMS_taxonomy_t {
+	ecorankidx_t* ranks;
+	econameidx_t* names;
+	ecotxidx_t*   taxa;
+} OBIDMS_taxonomy_t, *OBIDMS_taxonomy_p;
+
+
+OBIDMS_taxonomy_p obi_read_taxonomy(OBIDMS_p dms, const char* taxonomy_name, bool read_alternative_names);
+
+int obi_close_taxonomy(OBIDMS_taxonomy_p taxonomy);
+
+ecotx_t* obi_taxo_get_parent_at_rank(ecotx_t* taxon, int32_t rankidx);
+
+ecotx_t* obi_taxo_get_taxon_with_taxid(OBIDMS_taxonomy_p taxonomy, int32_t taxid);
+
+bool obi_taxo_is_taxon_under_taxid(ecotx_t* taxon, int32_t other_taxid);
+
+ecotx_t* obi_taxo_get_species(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
+
+ecotx_t* obi_taxo_get_genus(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
+
+ecotx_t* obi_taxo_get_family(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
+
+ecotx_t* obi_taxo_get_kingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
+
+ecotx_t* obi_taxo_get_superkingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
+

src/obidmscolumn.h

@@ -25,7 +25,7 @@
 #include "obierrno.h"
 #include "obilittlebigman.h"
 #include "obidmscolumndir.h"
-#include "obiarray.h"
+#include "obiavl.h"
 
 
 #define ONE_IF_ZERO(x) (((x)==0)?1:(x))	/**< If x is equal to 0, x takes the value 1.
@@ -34,11 +34,12 @@
                                 	     */
 #define COLUMN_GROWTH_FACTOR (2)	 	/**< The growth factor when a column is enlarged.
                                 	   	 */
-#define MAXIMUM_LINE_COUNT (1000000)    /**< The maximum line count for the data of a column. //TODO
+#define MAXIMUM_LINE_COUNT (1000000000) /**< The maximum line count for the data of a column. //TODO
                                 	     */
-#define FORMATTED_TIME_LENGTH (1024)    /**< The length allocated for the character string containing a formatted date
+#define FORMATTED_TIME_LENGTH (1024)    /**< The length allocated for the character string containing a formatted date.
+ 	 	 	 	 	 	 	 	 	     */
+#define COMMENTS_MAX_LENGTH (2048)      /**< The maximum length for comments.
  	 	 	 	 	 	 	 	 	     */
-
 
 typedef int32_t obiversion_t; 	/**< Used to store the column version number
                                  */
@@ -48,23 +49,24 @@ typedef int32_t obiversion_t; 	/**< Used to store the column version number
  * @brief OBIDMS column header structure.
  */
 typedef struct OBIDMS_column_header {
-	bool			little_endian;		        			/**< Endianness of the column:
-											     	 	 	 *   	- `true`  on little endian platforms
-											     	 	 	 *		- `false` on big endian platforms
-											     	 	 	 *   @see obi_is_little_endian()
-											     	 	 	 */
 	size_t			header_size;		   				    /**< Size of the header in bytes.
 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	size_t			data_size;			   				    /**< Size of the data in bytes.
+	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
 	index_t			line_count;							    /**< Number of lines of data allocated.
 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
 	index_t			lines_used;							    /**< Number of lines of data used.
 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	index_t			nb_elements_per_line;       			/**< Number of elements per line (default: 1).
+	index_t			nb_elements_per_line;   			   	/**< Number of elements per line.
 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
 	char			elements_names[ELEMENTS_NAMES_MAX+1];	/**< Names of the line elements with ';' as separator
 															 *	 (should be the column name if one element per line).
 															 */
-	OBIType_t		data_type;				    			/**< Type of the data.
+	OBIType_t		returned_data_type;		    			/**< Type of the data that is returned when getting an
+															 *   element from the column.
+	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
+	OBIType_t		stored_data_type;		    			/**< Type of the data that is actually stored in the data
+															 *   part of the column.
 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
 	time_t			creation_date;			    			/**< Date of creation of the file.
 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
@@ -76,10 +78,9 @@ typedef struct OBIDMS_column_header {
 															 */
 	char            name[OBIDMS_COLUMN_MAX_NAME+1]; 	    /**< The column name as a NULL terminated string.
 	                                             	 	 	 */
-	char            array_name[ARRAY_MAX_NAME+1]; 	    	/**< If there is one, the obi_array name as a NULL terminated string.
+	char            avl_name[AVL_MAX_NAME+1]; 	    		/**< If there is one, the AVL tree name as a NULL terminated string.
 	                                             	 	 	 */
-	char 			comments[1];							/**< Comments stored as a classical zero end C string.
-												 	 	 	 *	 The size of the comment is only limited by the header size.
+	char 			comments[COMMENTS_MAX_LENGTH+1];		/**< Comments stored as a classical zero end C string.
 												 	 	 	 */
 } OBIDMS_column_header_t, *OBIDMS_column_header_p;
 
@@ -97,20 +98,22 @@ typedef struct OBIDMS_column {
 	 	 	 	 	 	 	 	 	 	 	 	 	 */
 	OBIDMS_column_header_p		header; 		 	/**< A pointer to the header of the column.
 	 	 	 	 	 	 	 	 	 	 	 	 	 */
-	OBIDMS_array_p 				array;				/**< A pointer to the array associated with the column if there is one.
+	OBIDMS_avl_p 				avl;				/**< A pointer to the AVL tree associated with the column if there is one.
 	 	 	 	 	 	 	 	 	 	 	 	 	 */
 	void*                   	data;   		 	/**< A `void` pointer to the beginning of the data.
 	                                 	 	         *
 													 *   @warning Never use this member directly outside of the code of the
 													 *            low level functions of the OBIDMS.
 													 */
-	bool						writable;	     	/**< Indicates if the column is writable or not.	TODO delete?
+	bool						writable;	     	/**< Indicates if the column is writable or not.
 													 *       - `true` the column is writable
 													 *       - `false` the column is read-only
 													 *
 													 * A column is writable only by its creator
 													 * until it closes it.
 													 */
+	size_t						counter;			/**< Indicates by how many threads/programs (TODO) the column is used.
+													 */
 } OBIDMS_column_t, *OBIDMS_column_p;
 
 
@@ -161,7 +164,7 @@ size_t obi_get_platform_header_size();
  * @brief Creates a column.
  *
  * The minimum data size allocated is one memory page, and the data is initialized to the NA value of the OBIType.
- * If there is an array associated with the column, it is opened or created if it does not already exist.
+ * If there is an AVL tree associated with the column, it is opened or created if it does not already exist.
  *
  * @warning If there is one element per line, elements_names should be equal to column_name.	// TODO change this condition?
  *
@@ -169,9 +172,10 @@ size_t obi_get_platform_header_size();
  * @param column_name The name of the new column.
  * @param data_type The OBIType code of the data.
  * @param nb_lines The number of lines to be stored.
- * @param nb_elements_per_line The number of elements per line.
+ * @param nb_elements_per_line The number of elements per line.								// TODO talk about default values
  * @param elements_names The names of the elements with ';' as separator.
- * @param array_name The name of the array if there is one associated with the column.
+ * @param avl_name The name of the AVL tree if there is one associated with the column.
+ * @param comments Optional comments associated with the column.
  *
  * @returns A pointer on the newly created column structure.
  * @retval NULL if an error occurred.
@@ -185,7 +189,9 @@ OBIDMS_column_p obi_create_column(OBIDMS_p dms,
 								  index_t nb_lines,
 								  index_t nb_elements_per_line,
 								  const char* elements_names,
-								  const char* array_name);
+								  const char* avl_name,
+								  const char* comments
+								 );
 
 
 /**
@@ -218,7 +224,7 @@ OBIDMS_column_p obi_open_column(OBIDMS_p dms, const char* column_name, obiversio
  * @since August 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
  */
-OBIDMS_column_p obi_clone_column(OBIDMS_p dms, const char* column_name, obiversion_t version_number, bool clone_data);
+OBIDMS_column_p obi_clone_column(OBIDMS_p dms, OBIDMS_column_p line_selection, const char* column_name, obiversion_t version_number, bool clone_data);
 
 
 /**
@@ -289,7 +295,7 @@ int obi_truncate_and_close_column(OBIDMS_column_p column);
  * @since August 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
  */
-void obi_ini_to_NA_values(OBIDMS_column_p column, index_t first_line_nb, index_t nb_lines);	// TO make private?
+void obi_ini_to_NA_values(OBIDMS_column_p column, index_t first_line_nb, index_t nb_lines);	// TODO make private?
 
 
 /**
@@ -299,6 +305,8 @@ void obi_ini_to_NA_values(OBIDMS_column_p column, index_t first_line_nb, index_t
  *
  * @param dms A pointer on an OBIDMS.
  * @param column_name The name of an OBIDMS column.
+ * @param version_number The version of the column from which the header should be
+ *        retrieved (-1: latest version).
  *
  * @returns A pointer on the mmapped header of the column.
  * @retval NULL if an error occurred.
@@ -306,7 +314,7 @@ void obi_ini_to_NA_values(OBIDMS_column_p column, index_t first_line_nb, index_t
  * @since October 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
  */
-OBIDMS_column_header_p obi_column_get_header_from_name(OBIDMS_p dms, const char* column_name);
+OBIDMS_column_header_p obi_column_get_header_from_name(OBIDMS_p dms, const char* column_name, obiversion_t version_number);
 
 
 /**
@@ -330,7 +338,7 @@ int obi_unmap_header(OBIDMS_column_header_p header);
  * @param element_name The name of the element.
  *
  * @returns The index of the element in a line of the column.
- * @retval SIZE_MAX if an error occurred.
+ * @retval OBIIdx_NA if an error occurred.						// TODO not sure if this is "clean".
  *
  * @since July 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
@@ -341,6 +349,8 @@ index_t obi_column_get_element_index_from_name(OBIDMS_column_p column, const cha
 /**
  * @brief Formats a date in a way that is easy to read.
  *
+ * @warning The pointer returned must be freed by the caller.
+ *
  * @param date A date.
  *
  * @returns The date formatted in a way that is easy to read.
@@ -351,4 +361,8 @@ index_t obi_column_get_element_index_from_name(OBIDMS_column_p column, const cha
 char* obi_column_format_date(time_t date);
 
 
+int obi_select(OBIDMS_column_p lines_column, index_t line_to_grep);
+
+
+
 #endif /* OBIDMSCOLUMN_H_ */

src/obidmscolumn_bool.c

@@ -14,6 +14,7 @@
 #include <stdio.h>
 
 #include "obidmscolumn.h"
+#include "obiview.h"
 #include "obitypes.h"
 #include "obierrno.h"
 #include "obidebug.h"
@@ -28,6 +29,7 @@
  *
  **********************************************************************/
 
+
 int obi_column_set_obibool_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx, obibool_t value)
 {
 	// Check that the line number is not greater than the maximum allowed
@@ -57,18 +59,72 @@ int obi_column_set_obibool_with_elt_idx(OBIDMS_column_p column, index_t line_nb,
 }
 
 
+int obi_column_set_obibool_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, obibool_t value)
+{
+	// Check that the view is not read-only
+	if (view->read_only)
+	{
+		obi_set_errno(OBIVIEW_ERROR);
+		obidebug(1, "\nError trying to set a value in a column in a read-only view");
+		return -1;
+	}
+
+	if ((view->line_selection != NULL) || (!(column->writable)))
+	{
+		// Get the right line number
+		if (column->writable)
+			line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+		column = obi_view_clone_column(view, (column->header)->name);
+		if (column == NULL)
+		{
+			obidebug(1, "\nError trying to clone a column to modify it");
+			return -1;
+		}
+	}
+
+	if ((line_nb+1) > view->line_count)
+	{
+		if (obi_view_update_lines(view, (line_nb+1)) < 0)
+			return -1;
+	}
+
+	if (obi_column_set_obibool_with_elt_idx(column, line_nb, element_idx, value) < 0)
+		return -1;
+
+	return 0;
+}
+
+
 obibool_t obi_column_get_obibool_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx)
 {
-	if ((line_nb+1) > (column->header)->lines_used)
+	if ((line_nb+1) > ((column->header)->line_count))
 	{
 		obi_set_errno(OBICOL_UNKNOWN_ERROR);
-		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used");
+		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used in the column");
 		return OBIBool_NA;
 	}
+
 	return *(((obibool_t*) (column->data)) + (line_nb * ((column->header)->nb_elements_per_line)) + element_idx);
 }
 
 
+obibool_t obi_column_get_obibool_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx)
+{
+	if ((line_nb+1) > (view->line_count))
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to get a value that is beyond the current line count of the view");
+		return OBIBool_NA;
+	}
+
+	if (view->line_selection)
+		line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+	return obi_column_get_obibool_with_elt_idx(column, line_nb, element_idx);
+}
+
+
 int obi_column_set_obibool_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name, obibool_t value)
 {
 	index_t element_idx;
@@ -80,6 +136,17 @@ int obi_column_set_obibool_with_elt_name(OBIDMS_column_p column, index_t line_nb
 }
 
 
+int obi_column_set_obibool_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, obibool_t value)
+{
+	index_t element_idx;
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return -1;
+	obi_column_set_obibool_with_elt_idx_in_view(view, column, line_nb, element_idx, value);
+	return 0;
+}
+
+
 obibool_t obi_column_get_obibool_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name)
 {
 	index_t element_idx;
@@ -90,3 +157,13 @@ obibool_t obi_column_get_obibool_with_elt_name(OBIDMS_column_p column, index_t l
 	return obi_column_get_obibool_with_elt_idx(column, line_nb, element_idx);
 }
 
+
+obibool_t obi_column_get_obibool_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name)
+{
+	index_t element_idx;
+
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return OBIBool_NA;
+	return obi_column_get_obibool_with_elt_idx_in_view(view, column, line_nb, element_idx);
+}

src/obidmscolumn_bool.h

@@ -19,6 +19,7 @@
 
 #include "obidmscolumn.h"
 #include "obitypes.h"
+#include "obiview.h"
 
 
 /**
@@ -38,7 +39,7 @@
  * @since July 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
  */
-int obi_column_set_obibool_with_elt_idx(OBIDMS_column_p column, index_t line_nb, size_t element_idx, obibool_t value);
+int obi_column_set_obibool_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx, obibool_t value);
 
 
 /**
@@ -54,7 +55,7 @@ int obi_column_set_obibool_with_elt_idx(OBIDMS_column_p column, index_t line_nb,
  * @since July 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
  */
-obibool_t obi_column_get_obibool_with_elt_idx(OBIDMS_column_p column, index_t line_nb, size_t element_idx);
+obibool_t obi_column_get_obibool_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx);
 
 
 /**
@@ -95,5 +96,79 @@ int obi_column_set_obibool_with_elt_name(OBIDMS_column_p column, index_t line_nb
 obibool_t obi_column_get_obibool_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name);
 
 
+/**
+ * @brief Sets a value in an OBIDMS column containing data with the type OBI_BOOL, using the index of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_idx The index of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obibool_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, obibool_t value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data with the type OBI_BOOL.
+ *
+ * @param column A pointer as returned by obi_create_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_idx The index of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval OBIBool_NA the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+obibool_t obi_column_get_obibool_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx);
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data with the type OBI_BOOL,
+ *        using the name of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_name The name of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obibool_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, obibool_t value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data with the type OBI_BOOL,
+ *        using the name of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_name The name of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval OBIBool_NA the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+obibool_t obi_column_get_obibool_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name);
+
+
 #endif /* OBIDMSCOLUMN_BOOL_H_ */
 

src/obidmscolumn_char.c

@@ -14,6 +14,7 @@
 #include <stdio.h>
 
 #include "obidmscolumn.h"
+#include "obiview.h"
 #include "obitypes.h"
 #include "obierrno.h"
 #include "obidebug.h"
@@ -28,6 +29,7 @@
  *
  **********************************************************************/
 
+
 int obi_column_set_obichar_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx, obichar_t value)
 {
 	// Check that the line number is not greater than the maximum allowed
@@ -57,18 +59,72 @@ int obi_column_set_obichar_with_elt_idx(OBIDMS_column_p column, index_t line_nb,
 }
 
 
+int obi_column_set_obichar_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, obichar_t value)
+{
+	// Check that the view is not read-only
+	if (view->read_only)
+	{
+		obi_set_errno(OBIVIEW_ERROR);
+		obidebug(1, "\nError trying to set a value in a column in a read-only view");
+		return -1;
+	}
+
+	if ((view->line_selection != NULL) || (!(column->writable)))
+	{
+		// Get the right line number
+		if (column->writable)
+			line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+		column = obi_view_clone_column(view, (column->header)->name);
+		if (column == NULL)
+		{
+			obidebug(1, "\nError trying to clone a column to modify it");
+			return -1;
+		}
+	}
+
+	if ((line_nb+1) > view->line_count)
+	{
+		if (obi_view_update_lines(view, (line_nb+1)) < 0)
+			return -1;
+	}
+
+	if (obi_column_set_obichar_with_elt_idx(column, line_nb, element_idx, value) < 0)
+		return -1;
+
+	return 0;
+}
+
+
 obichar_t obi_column_get_obichar_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx)
 {
-	if ((line_nb+1) > (column->header)->lines_used)
+	if ((line_nb+1) > ((column->header)->line_count))
 	{
 		obi_set_errno(OBICOL_UNKNOWN_ERROR);
-		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used");
+		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used in the column");
 		return OBIChar_NA;
 	}
+
 	return *(((obichar_t*) (column->data)) + (line_nb * ((column->header)->nb_elements_per_line)) + element_idx);
 }
 
 
+obichar_t obi_column_get_obichar_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx)
+{
+	if ((line_nb+1) > (view->line_count))
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to get a value that is beyond the current line count of the view");
+		return OBIChar_NA;
+	}
+
+	if (view->line_selection)
+		line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+	return obi_column_get_obichar_with_elt_idx(column, line_nb, element_idx);
+}
+
+
 int obi_column_set_obichar_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name, obichar_t value)
 {
 	index_t element_idx;
@@ -80,6 +136,17 @@ int obi_column_set_obichar_with_elt_name(OBIDMS_column_p column, index_t line_nb
 }
 
 
+int obi_column_set_obichar_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, obichar_t value)
+{
+	index_t element_idx;
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return -1;
+	obi_column_set_obichar_with_elt_idx_in_view(view, column, line_nb, element_idx, value);
+	return 0;
+}
+
+
 obichar_t obi_column_get_obichar_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name)
 {
 	index_t element_idx;
@@ -90,3 +157,13 @@ obichar_t obi_column_get_obichar_with_elt_name(OBIDMS_column_p column, index_t l
 	return obi_column_get_obichar_with_elt_idx(column, line_nb, element_idx);
 }
 
+
+obichar_t obi_column_get_obichar_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name)
+{
+	index_t element_idx;
+
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return OBIChar_NA;
+	return obi_column_get_obichar_with_elt_idx_in_view(view, column, line_nb, element_idx);
+}

src/obidmscolumn_char.h

@@ -19,6 +19,7 @@
 
 #include "obidmscolumn.h"
 #include "obitypes.h"
+#include "obiview.h"
 
 
 /**
@@ -49,7 +50,7 @@ int obi_column_set_obichar_with_elt_idx(OBIDMS_column_p column, index_t line_nb,
  * @param element_idx The index of the element that should be recovered in the line.
  *
  * @returns The recovered value.
- * @retval OBIBool_NA the NA value of the type if an error occurred and obi_errno is set.
+ * @retval OBIChar_NA the NA value of the type if an error occurred and obi_errno is set.
  *
  * @since July 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
@@ -87,7 +88,7 @@ int obi_column_set_obichar_with_elt_name(OBIDMS_column_p column, index_t line_nb
  * @param element_name The name of the element that should be recovered in the line.
  *
  * @returns The recovered value.
- * @retval OBIBool_NA the NA value of the type if an error occurred and obi_errno is set.
+ * @retval OBIChar_NA the NA value of the type if an error occurred and obi_errno is set.
  *
  * @since August 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
@@ -95,5 +96,79 @@ int obi_column_set_obichar_with_elt_name(OBIDMS_column_p column, index_t line_nb
 obichar_t obi_column_get_obichar_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name);
 
 
+/**
+ * @brief Sets a value in an OBIDMS column containing data with the type OBI_CHAR, using the index of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_idx The index of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obichar_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, obichar_t value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data with the type OBI_CHAR.
+ *
+ * @param column A pointer as returned by obi_create_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_idx The index of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval OBIChar_NA the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+obichar_t obi_column_get_obichar_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx);
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data with the type OBI_CHAR,
+ *        using the name of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_name The name of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obichar_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, obichar_t value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data with the type OBI_CHAR,
+ *        using the name of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_name The name of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval OBIChar_NA the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+obichar_t obi_column_get_obichar_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name);
+
+
 #endif /* OBIDMSCOLUMN_CHAR_H_ */
 

src/obidmscolumn_float.c

@@ -14,6 +14,7 @@
 #include <stdio.h>
 
 #include "obidmscolumn.h"
+#include "obiview.h"
 #include "obitypes.h"
 #include "obierrno.h"
 #include "obidebug.h"
@@ -57,18 +58,72 @@ int obi_column_set_obifloat_with_elt_idx(OBIDMS_column_p column, index_t line_nb
 }
 
 
+int obi_column_set_obifloat_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, obifloat_t value)
+{
+	// Check that the view is not read-only
+	if (view->read_only)
+	{
+		obi_set_errno(OBIVIEW_ERROR);
+		obidebug(1, "\nError trying to set a value in a column in a read-only view");
+		return -1;
+	}
+
+	if ((view->line_selection != NULL) || (!(column->writable)))
+	{
+		// Get the right line number
+		if (column->writable)
+			line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+		column = obi_view_clone_column(view, (column->header)->name);
+		if (column == NULL)
+		{
+			obidebug(1, "\nError trying to clone a column to modify it");
+			return -1;
+		}
+	}
+
+	if ((line_nb+1) > view->line_count)
+	{
+		if (obi_view_update_lines(view, (line_nb+1)) < 0)
+			return -1;
+	}
+
+	if (obi_column_set_obifloat_with_elt_idx(column, line_nb, element_idx, value) < 0)
+		return -1;
+
+	return 0;
+}
+
+
 obifloat_t obi_column_get_obifloat_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx)
 {
-	if ((line_nb+1) > (column->header)->lines_used)
+	if ((line_nb+1) > ((column->header)->line_count))
 	{
 		obi_set_errno(OBICOL_UNKNOWN_ERROR);
-		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used");
+		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used in the column");
 		return OBIFloat_NA;
 	}
+
 	return *(((obifloat_t*) (column->data)) + (line_nb * ((column->header)->nb_elements_per_line)) + element_idx);
 }
 
 
+obifloat_t obi_column_get_obifloat_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx)
+{
+	if ((line_nb+1) > (view->line_count))
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to get a value that is beyond the current line count of the view");
+		return OBIFloat_NA;
+	}
+
+	if (view->line_selection)
+		line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+	return obi_column_get_obifloat_with_elt_idx(column, line_nb, element_idx);
+}
+
+
 int obi_column_set_obifloat_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name, obifloat_t value)
 {
 	index_t element_idx;
@@ -80,6 +135,17 @@ int obi_column_set_obifloat_with_elt_name(OBIDMS_column_p column, index_t line_n
 }
 
 
+int obi_column_set_obifloat_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, obifloat_t value)
+{
+	index_t element_idx;
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return -1;
+	obi_column_set_obifloat_with_elt_idx_in_view(view, column, line_nb, element_idx, value);
+	return 0;
+}
+
+
 obifloat_t obi_column_get_obifloat_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name)
 {
 	index_t element_idx;
@@ -90,3 +156,13 @@ obifloat_t obi_column_get_obifloat_with_elt_name(OBIDMS_column_p column, index_t
 	return obi_column_get_obifloat_with_elt_idx(column, line_nb, element_idx);
 }
 
+
+obifloat_t obi_column_get_obifloat_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name)
+{
+	index_t element_idx;
+
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return OBIFloat_NA;
+	return obi_column_get_obifloat_with_elt_idx_in_view(view, column, line_nb, element_idx);
+}

src/obidmscolumn_float.h

@@ -19,6 +19,7 @@
 
 #include "obidmscolumn.h"
 #include "obitypes.h"
+#include "obiview.h"
 
 
 /**
@@ -49,7 +50,7 @@ int obi_column_set_obifloat_with_elt_idx(OBIDMS_column_p column, index_t line_nb
  * @param element_idx The index of the element that should be recovered in the line.
  *
  * @returns The recovered value.
- * @retval OBIBool_NA the NA value of the type if an error occurred and obi_errno is set.
+ * @retval OBIFloat_NA the NA value of the type if an error occurred and obi_errno is set.
  *
  * @since July 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
@@ -87,7 +88,7 @@ int obi_column_set_obifloat_with_elt_name(OBIDMS_column_p column, index_t line_n
  * @param element_name The name of the element that should be recovered in the line.
  *
  * @returns The recovered value.
- * @retval OBIBool_NA the NA value of the type if an error occurred and obi_errno is set.
+ * @retval OBIFloat_NA the NA value of the type if an error occurred and obi_errno is set.
  *
  * @since August 2015
  * @author Celine Mercier (celine.mercier@metabarcoding.org)
@@ -95,5 +96,79 @@ int obi_column_set_obifloat_with_elt_name(OBIDMS_column_p column, index_t line_n
 obifloat_t obi_column_get_obifloat_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name);
 
 
+/**
+ * @brief Sets a value in an OBIDMS column containing data with the type OBI_FLOAT, using the index of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_idx The index of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obifloat_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, obifloat_t value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data with the type OBI_FLOAT.
+ *
+ * @param column A pointer as returned by obi_create_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_idx The index of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval OBIFloat_NA the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+obifloat_t obi_column_get_obifloat_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx);
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data with the type OBI_FLOAT,
+ *        using the name of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_name The name of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obifloat_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, obifloat_t value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data with the type OBI_FLOAT,
+ *        using the name of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_name The name of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval OBIFloat_NA the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+obifloat_t obi_column_get_obifloat_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name);
+
+
 #endif /* OBIDMSCOLUMN_FLOAT_H_ */
 

src/obidmscolumn_idx.c

@@ -0,0 +1,72 @@
+/****************************************************************************
+ * OBIDMS_column_idx functions                                              *
+ ****************************************************************************/
+
+/**
+ * @file obidsmcolumn_idx.c
+ * @author Celine Mercier
+ * @date February 14th 2016
+ * @brief Functions handling OBIColumns containing data with the index_t type.
+ */
+
+
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "obidmscolumn.h"
+#include "obitypes.h"
+#include "obierrno.h"
+#include "obidebug.h"
+
+
+#define DEBUG_LEVEL 0	// TODO has to be defined somewhere else (cython compil flag?)
+
+
+/**********************************************************************
+ *
+ * D E F I N I T I O N   O F   T H E   P U B L I C   F U N C T I O N S
+ *
+ **********************************************************************/
+
+
+int obi_column_set_index(OBIDMS_column_p column, index_t line_nb, index_t value)
+{
+	// Check that the line number is not greater than the maximum allowed
+	if (line_nb >= MAXIMUM_LINE_COUNT)
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to set a value at a line number greater than the maximum allowed");
+    	return -1;
+	}
+
+	// Check if the file needs to be enlarged
+	while ((line_nb+1) > (column->header)->line_count)
+	{
+		// Enlarge the file
+		if (obi_enlarge_column(column) < 0)
+	    	return -1;
+	}
+
+	// Update lines used
+	if ((line_nb+1) > (column->header)->lines_used)
+		(column->header)->lines_used = line_nb+1;
+
+	// Set the value
+	*(((index_t*) (column->data)) + line_nb) = value;
+
+	return 0;
+}
+
+
+index_t obi_column_get_index(OBIDMS_column_p column, index_t line_nb)
+{
+	if ((line_nb+1) > ((column->header)->line_count))
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used in the column");
+		return OBIIdx_NA;
+	}
+
+	return *(((index_t*) (column->data)) + line_nb);
+}
+

src/obidmscolumn_idx.h

@@ -0,0 +1,30 @@
+/****************************************************************************
+ * OBIDMS_column_idx header file                                            *
+ ****************************************************************************/
+
+/**
+ * @file obidsmcolumn_idx.h
+ * @author Celine Mercier
+ * @date February 14th 2016
+ * @brief Header file for the functions handling OBIColumns containing data with the OBIType OBI_IDX.
+ */
+
+
+#ifndef OBIDMSCOLUMN_IDX_H_
+#define OBIDMSCOLUMN_IDX_H_
+
+
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "obidmscolumn.h"
+#include "obitypes.h"
+
+
+int obi_column_set_index(OBIDMS_column_p column, index_t line_nb, index_t value);
+
+index_t obi_column_get_index(OBIDMS_column_p column, index_t line_nb);
+
+
+#endif /* OBIDMSCOLUMN_IDX_H_ */
+

src/obidmscolumn_int.c

@@ -14,6 +14,7 @@
 #include <stdio.h>
 
 #include "obidmscolumn.h"
+#include "obiview.h"
 #include "obitypes.h"
 #include "obierrno.h"
 #include "obidebug.h"
@@ -28,6 +29,7 @@
  *
  **********************************************************************/
 
+
 int obi_column_set_obiint_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx, obiint_t value)
 {
 	// Check that the line number is not greater than the maximum allowed
@@ -57,18 +59,72 @@ int obi_column_set_obiint_with_elt_idx(OBIDMS_column_p column, index_t line_nb,
 }
 
 
+int obi_column_set_obiint_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, obiint_t value)
+{
+	// Check that the view is not read-only
+	if (view->read_only)
+	{
+		obi_set_errno(OBIVIEW_ERROR);
+		obidebug(1, "\nError trying to set a value in a column in a read-only view");
+		return -1;
+	}
+
+	if ((view->line_selection != NULL) || (!(column->writable)))
+	{
+		// Get the right line number
+		if (column->writable)
+			line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+		column = obi_view_clone_column(view, (column->header)->name);
+		if (column == NULL)
+		{
+			obidebug(1, "\nError trying to clone a column to modify it");
+			return -1;
+		}
+	}
+
+	if ((line_nb+1) > view->line_count)
+	{
+		if (obi_view_update_lines(view, (line_nb+1)) < 0)
+			return -1;
+	}
+
+	if (obi_column_set_obiint_with_elt_idx(column, line_nb, element_idx, value) < 0)
+		return -1;
+
+	return 0;
+}
+
+
 obiint_t obi_column_get_obiint_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx)
 {
-	if ((line_nb+1) > (column->header)->lines_used)
+	if ((line_nb+1) > ((column->header)->line_count))
 	{
 		obi_set_errno(OBICOL_UNKNOWN_ERROR);
-		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used");
+		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used in the column");
 		return OBIInt_NA;
 	}
+
 	return *(((obiint_t*) (column->data)) + (line_nb * ((column->header)->nb_elements_per_line)) + element_idx);
 }
 
 
+obiint_t obi_column_get_obiint_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx)
+{
+	if ((line_nb+1) > (view->line_count))
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to get a value that is beyond the current line count of the view");
+		return OBIInt_NA;
+	}
+
+	if (view->line_selection)
+		line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+	return obi_column_get_obiint_with_elt_idx(column, line_nb, element_idx);
+}
+
+
 int obi_column_set_obiint_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name, obiint_t value)
 {
 	index_t element_idx;
@@ -80,6 +136,17 @@ int obi_column_set_obiint_with_elt_name(OBIDMS_column_p column, index_t line_nb,
 }
 
 
+int obi_column_set_obiint_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, obiint_t value)
+{
+	index_t element_idx;
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return -1;
+	obi_column_set_obiint_with_elt_idx_in_view(view, column, line_nb, element_idx, value);
+	return 0;
+}
+
+
 obiint_t obi_column_get_obiint_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name)
 {
 	index_t element_idx;
@@ -90,3 +157,14 @@ obiint_t obi_column_get_obiint_with_elt_name(OBIDMS_column_p column, index_t lin
 	return obi_column_get_obiint_with_elt_idx(column, line_nb, element_idx);
 }
 
+
+obiint_t obi_column_get_obiint_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name)
+{
+	index_t element_idx;
+
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return OBIInt_NA;
+	return obi_column_get_obiint_with_elt_idx_in_view(view, column, line_nb, element_idx);
+}
+

src/obidmscolumn_int.h

@@ -19,6 +19,7 @@
 
 #include "obidmscolumn.h"
 #include "obitypes.h"
+#include "obiview.h"
 
 
 /**
@@ -95,5 +96,79 @@ int obi_column_set_obiint_with_elt_name(OBIDMS_column_p column, index_t line_nb,
 obiint_t obi_column_get_obiint_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name);
 
 
+/**
+ * @brief Sets a value in an OBIDMS column containing data with the type OBI_INT, using the index of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_idx The index of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obiint_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, obiint_t value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data with the type OBI_INT.
+ *
+ * @param column A pointer as returned by obi_create_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_idx The index of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval OBIBool_NA the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+obiint_t obi_column_get_obiint_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx);
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data with the type OBI_INT,
+ *        using the name of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_name The name of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obiint_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, obiint_t value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data with the type OBI_INT,
+ *        using the name of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_name The name of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval OBIBool_NA the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+obiint_t obi_column_get_obiint_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name);
+
+
 #endif /* OBIDMSCOLUMN_INT_H_ */
 

src/obidmscolumn_seq.c

@@ -0,0 +1,196 @@
+/****************************************************************************
+ * OBIDMS_column_seq functions                                              *
+ ****************************************************************************/
+
+/**
+ * @file obidsmcolumn_seq.c
+ * @author Celine Mercier
+ * @date November 18th 2015
+ * @brief Functions handling OBIColumns containing data in the form of indices referring to DNA sequences.
+ */
+
+
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "obidmscolumn.h"
+#include "obiview.h"
+#include "obitypes.h"
+#include "obierrno.h"
+#include "obidebug.h"
+#include "obiavl.h"
+
+
+#define DEBUG_LEVEL 0	// TODO has to be defined somewhere else (cython compil flag?)
+
+
+/**********************************************************************
+ *
+ * D E F I N I T I O N   O F   T H E   P U B L I C   F U N C T I O N S
+ *
+ **********************************************************************/
+
+
+int obi_column_set_obiseq_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx, char* value)
+{
+	byte_t* value_b;
+	index_t idx;
+
+	// Check that the line number is not greater than the maximum allowed
+	if (line_nb >= MAXIMUM_LINE_COUNT)
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to set a value at a line number greater than the maximum allowed");
+    	return -1;
+	}
+
+	// Check if the file needs to be enlarged
+	while ((line_nb+1) > (column->header)->line_count)
+	{
+		// Enlarge the file
+		if (obi_enlarge_column(column) < 0)
+	    	return -1;
+	}
+
+	// Update lines used
+	if ((line_nb+1) > (column->header)->lines_used)
+		(column->header)->lines_used = line_nb+1;
+
+	// Encode the value on a byte array with a header
+	value_b = obi_seq_to_obibytes(value);
+	if (value_b == NULL)
+		return -1;
+
+	// Add in the AVL tree
+	idx = obi_avl_add(column->avl, value_b);
+	if (idx == -1)
+		return -1;
+
+	// Add the value's index in the column
+	*(((index_t*) (column->data)) + (line_nb * ((column->header)->nb_elements_per_line)) + element_idx) = idx;
+
+	free(value_b);
+
+	return 0;
+}
+
+
+int obi_column_set_obiseq_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, char* value)
+{
+	// Check that the view is not read-only
+	if (view->read_only)
+	{
+		obi_set_errno(OBIVIEW_ERROR);
+		obidebug(1, "\nError trying to set a value in a column in a read-only view");
+		return -1;
+	}
+
+	if ((view->line_selection != NULL) || (!(column->writable)))
+	{
+		// Get the right line number
+		if (column->writable)
+			line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+		column = obi_view_clone_column(view, (column->header)->name);
+		if (column == NULL)
+		{
+			obidebug(1, "\nError trying to clone a column to modify it");
+			return -1;
+		}
+	}
+
+	if ((line_nb+1) > view->line_count)
+	{
+		if (obi_view_update_lines(view, (line_nb+1)) < 0)
+			return -1;
+	}
+
+	if (obi_column_set_obiseq_with_elt_idx(column, line_nb, element_idx, value) < 0)
+		return -1;
+
+	return 0;
+}
+
+
+const char* obi_column_get_obiseq_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx)
+{
+	index_t idx;
+	byte_t* value_b;
+
+	if ((line_nb+1) > ((column->header)->line_count))
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used in the column");
+		return OBISeq_NA;
+	}
+
+	idx = *(((index_t*) (column->data)) + (line_nb * ((column->header)->nb_elements_per_line)) + element_idx);
+
+	// Check NA
+	if (idx == OBIIdx_NA)
+		return OBISeq_NA;
+
+	value_b = obi_avl_get(column->avl, idx);
+	return obi_obibytes_to_seq(value_b);
+}
+
+
+const char* obi_column_get_obiseq_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx)
+{
+	if ((line_nb+1) > (view->line_count))
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to get a value that is beyond the current line count of the view");
+		return OBISeq_NA;
+	}
+
+	if (view->line_selection)
+		line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+	return obi_column_get_obiseq_with_elt_idx(column, line_nb, element_idx);
+}
+
+
+int obi_column_set_obiseq_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name, char* value)
+{
+	index_t element_idx;
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return -1;
+	obi_column_set_obiseq_with_elt_idx(column, line_nb, element_idx, value);
+	return 0;
+}
+
+
+int obi_column_set_obiseq_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, char* value)
+{
+	index_t element_idx;
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return -1;
+	obi_column_set_obiseq_with_elt_idx_in_view(view, column, line_nb, element_idx, value);
+	return 0;
+}
+
+
+const char* obi_column_get_obiseq_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name)
+{
+	index_t element_idx;
+
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return OBISeq_NA;
+	return obi_column_get_obiseq_with_elt_idx(column, line_nb, element_idx);
+}
+
+
+const char* obi_column_get_obiseq_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name)
+{
+	index_t element_idx;
+
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return OBISeq_NA;
+	return obi_column_get_obiseq_with_elt_idx_in_view(view, column, line_nb, element_idx);
+}
+

src/obidmscolumn_seq.h

@@ -0,0 +1,178 @@
+/****************************************************************************
+ * OBIDMS_column_seq header file                                            *
+ ****************************************************************************/
+
+/**
+ * @file obidsmcolumn_seq.h
+ * @author Celine Mercier
+ * @date Novemeber 18th 2015
+ * @brief Header file for the functions handling OBIColumns containing data in the form of indices referring to DNA sequences.
+ */
+
+
+#ifndef OBIDMSCOLUMN_SEQ_H_
+#define OBIDMSCOLUMN_SEQ_H_
+
+
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "obidmscolumn.h"
+#include "obitypes.h"
+#include "obiview.h"
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data in the form of indices referring
+ * to DNA sequences in an AVL tree, using the index of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_idx The index of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obiseq_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx, char* value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data in the form of indices referring
+ * to DNA sequences in an AVL tree, using the index of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_idx The index of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval '\0' the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+const char* obi_column_get_obiseq_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx);
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data in the form of indices referring
+ * to DNA sequences in an AVL tree, using the name of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_name The name of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obiseq_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name, char* value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data in the form of indices referring
+ * to DNA sequences in an AVL tree, using the name of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_name The name of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval '\0' the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since November 2015
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+const char* obi_column_get_obiseq_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name);
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data in the form of indices referring
+ * to DNA sequences in an AVL tree, using the index of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_idx The index of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obiseq_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, char* value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data in the form of indices referring
+ * to DNA sequences in an AVL tree, using the index of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_idx The index of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval '\0' the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+const char* obi_column_get_obiseq_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx);
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data in the form of indices referring
+ * to DNA sequences in an AVL tree, using the name of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_name The name of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obiseq_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, char* value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data in the form of indices referring
+ * to DNA sequences in an AVL tree, using the name of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_name The name of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval '\0' the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+const char* obi_column_get_obiseq_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name);
+
+
+#endif /* OBIDMSCOLUMN_SEQ_H_ */
+

src/obidmscolumn_str.c

@@ -12,12 +12,14 @@
 
 #include <stdlib.h>
 #include <stdio.h>
+#include <sys/mman.h>
 
 #include "obidmscolumn.h"
+#include "obiview.h"
 #include "obitypes.h"
 #include "obierrno.h"
 #include "obidebug.h"
-#include "obiarray.h"
+#include "obiavl.h"
 
 
 #define DEBUG_LEVEL 0	// TODO has to be defined somewhere else (cython compil flag?)
@@ -59,8 +61,32 @@ int obi_column_set_obistr_with_elt_idx(OBIDMS_column_p column, index_t line_nb,
 	if (value_b == NULL)
 		return -1;
 
-	// Add in the obiarray
-	idx = obi_array_add(column->array, value_b);
+	if (!(line_nb%100000))
+	{
+		//fprintf(stderr, "\nbleeeh");
+		// Unmap and remap the AVL
+		if (munmap((column->avl)->tree, ((((column->avl)->header)->nb_items_max) * sizeof(AVL_node_t))) < 0)
+			return -1;
+		if (munmap(((column->avl)->data)->data, (((column->avl)->data)->header)->data_size_max) < 0)
+			return -1;
+		(column->avl)->tree = mmap(NULL,
+							  (((column->avl)->header)->nb_items_max) * sizeof(AVL_node_t),	// TODO store avl_size_max
+							  PROT_READ | PROT_WRITE,
+							  MAP_SHARED,
+							  (column->avl)->avl_fd,
+							  ((column->avl)->header)->header_size
+							  );
+		((column->avl)->data)->data = mmap(NULL,
+									  (((column->avl)->data)->header)->data_size_max,
+									  PROT_READ | PROT_WRITE,
+									  MAP_SHARED,
+									  (column->avl)->data_fd,
+									  (((column->avl)->data)->header)->header_size
+									  );
+	}
+
+	// Add in the AVL tree
+	idx = obi_avl_add(column->avl, value_b);
 	if (idx == -1)
 		return -1;
 
@@ -73,37 +99,100 @@ int obi_column_set_obistr_with_elt_idx(OBIDMS_column_p column, index_t line_nb,
 }
 
 
+int obi_column_set_obistr_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, char* value)
+{
+	// Check that the view is not read-only
+	if (view->read_only)
+	{
+		obi_set_errno(OBIVIEW_ERROR);
+		obidebug(1, "\nError trying to set a value in a column in a read-only view");
+		return -1;
+	}
+
+	if ((view->line_selection != NULL) || (!(column->writable)))
+	{
+		// Get the right line number
+		if (column->writable)
+			line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+		column = obi_view_clone_column(view, (column->header)->name);
+		if (column == NULL)
+		{
+			obidebug(1, "\nError trying to clone a column to modify it");
+			return -1;
+		}
+	}
+
+	if ((line_nb+1) > view->line_count)
+	{
+		if (obi_view_update_lines(view, (line_nb+1)) < 0)
+			return -1;
+	}
+
+	if (obi_column_set_obistr_with_elt_idx(column, line_nb, element_idx, value) < 0)
+		return -1;
+
+	return 0;
+}
+
+
 const char* obi_column_get_obistr_with_elt_idx(OBIDMS_column_p column, index_t line_nb, index_t element_idx)
 {
 	index_t idx;
 	byte_t* value_b;
 
-	if ((line_nb+1) > (column->header)->lines_used)
+	if ((line_nb+1) > ((column->header)->line_count))
 	{
 		obi_set_errno(OBICOL_UNKNOWN_ERROR);
-		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used");
-		return "\0";		// TODO
+		obidebug(1, "\nError trying to get a value that is beyond the current number of lines used in the column");
+		return OBIStr_NA;
 	}
 
 	idx = *(((index_t*) (column->data)) + (line_nb * ((column->header)->nb_elements_per_line)) + element_idx);
 
 	// Check NA
 	if (idx == OBIIdx_NA)
-		return "\0";		// TODO
+		return OBIStr_NA;
 
-	value_b = obi_array_get(column->array, idx);
+	value_b = obi_avl_get(column->avl, idx);
 	return obi_obibytes_to_str(value_b);
 }
 
 
+const char* obi_column_get_obistr_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx)
+{
+	if ((line_nb+1) > (view->line_count))
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError trying to get a value that is beyond the current line count of the view");
+		return OBIStr_NA;
+	}
+
+	if (view->line_selection)
+		line_nb = *(((index_t*) ((view->line_selection)->data)) + line_nb);
+
+	return obi_column_get_obistr_with_elt_idx(column, line_nb, element_idx);
+}
+
+
 int obi_column_set_obistr_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name, char* value)
 {
 	index_t element_idx;
 	element_idx = obi_column_get_element_index_from_name(column, element_name);
 	if (element_idx == OBIIdx_NA)
 		return -1;
-	if (obi_column_set_obistr_with_elt_idx(column, line_nb, element_idx, value) < 0)
+	obi_column_set_obistr_with_elt_idx(column, line_nb, element_idx, value);
+	return 0;
+}
+
+
+int obi_column_set_obistr_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, char* value)
+{
+	index_t element_idx;
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
 		return -1;
+	obi_column_set_obistr_with_elt_idx_in_view(view, column, line_nb, element_idx, value);
 	return 0;
 }
 
@@ -114,7 +203,18 @@ const char* obi_column_get_obistr_with_elt_name(OBIDMS_column_p column, index_t
 
 	element_idx = obi_column_get_element_index_from_name(column, element_name);
 	if (element_idx == OBIIdx_NA)
-		return "\0";
+		return OBIStr_NA;
 	return obi_column_get_obistr_with_elt_idx(column, line_nb, element_idx);
 }
 
+
+const char* obi_column_get_obistr_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name)
+{
+	index_t element_idx;
+
+	element_idx = obi_column_get_element_index_from_name(column, element_name);
+	if (element_idx == OBIIdx_NA)
+		return OBIStr_NA;
+	return obi_column_get_obistr_with_elt_idx_in_view(view, column, line_nb, element_idx);
+}
+

src/obidmscolumn_str.h

@@ -19,11 +19,12 @@
 
 #include "obidmscolumn.h"
 #include "obitypes.h"
+#include "obiview.h"
 
 
 /**
  * @brief Sets a value in an OBIDMS column containing data in the form of indices referring
- * to character strings in an obiarray, using the index of the element in the line.
+ * to character strings in an AVL tree, using the index of the element in the line.
  *
  * @warning Pointers returned by obi_open_column() don't allow writing.
  *
@@ -44,7 +45,7 @@ int obi_column_set_obistr_with_elt_idx(OBIDMS_column_p column, index_t line_nb,
 
 /**
  * @brief Recovers a value in an OBIDMS column containing data in the form of indices referring
- * to character strings in an obiarray, using the index of the element in the line.
+ * to character strings in an AVL tree, using the index of the element in the line.
  *
  * @param column A pointer as returned by obi_create_column().
  * @param line_nb The number of the line where the value should be recovered.
@@ -61,7 +62,7 @@ const char* obi_column_get_obistr_with_elt_idx(OBIDMS_column_p column, index_t l
 
 /**
  * @brief Sets a value in an OBIDMS column containing data in the form of indices referring
- * to character strings in an obiarray, using the name of the element in the line.
+ * to character strings in an AVL tree, using the name of the element in the line.
  *
  * @warning Pointers returned by obi_open_column() don't allow writing.
  *
@@ -82,7 +83,7 @@ int obi_column_set_obistr_with_elt_name(OBIDMS_column_p column, index_t line_nb,
 
 /**
  * @brief Recovers a value in an OBIDMS column containing data in the form of indices referring
- * to character strings in an obiarray, using the name of the element in the line.
+ * to character strings in an AVL tree, using the name of the element in the line.
  *
  * @param column A pointer as returned by obi_create_column() or obi_clone_column().
  * @param line_nb The number of the line where the value should be recovered.
@@ -97,5 +98,81 @@ int obi_column_set_obistr_with_elt_name(OBIDMS_column_p column, index_t line_nb,
 const char* obi_column_get_obistr_with_elt_name(OBIDMS_column_p column, index_t line_nb, const char* element_name);
 
 
-#endif /* OBIDMSCOLUMN_IDX_H_ */
+/**
+ * @brief Sets a value in an OBIDMS column containing data in the form of indices referring
+ * to character strings in an AVL tree, using the index of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_idx The index of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obistr_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx, char* value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data in the form of indices referring
+ * to character strings in an AVL tree, using the index of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_idx The index of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval '\0' the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+const char* obi_column_get_obistr_with_elt_idx_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, index_t element_idx);
+
+
+/**
+ * @brief Sets a value in an OBIDMS column containing data in the form of indices referring
+ * to character strings in an AVL tree, using the name of the element in the line.
+ *
+ * @warning Pointers returned by obi_open_column() don't allow writing.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be set.
+ * @param element_name The name of the element that should be set in the line.
+ * @param value The value that should be set.
+ *
+ * @returns An integer value indicating the success of the operation.
+ * @retval 0 on success.
+ * @retval -1 if an error occurred.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+int obi_column_set_obistr_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name, char* value);
+
+
+/**
+ * @brief Recovers a value in an OBIDMS column containing data in the form of indices referring
+ * to character strings in an AVL tree, using the name of the element in the line.
+ *
+ * @param column A pointer as returned by obi_create_column() or obi_clone_column().
+ * @param line_nb The number of the line where the value should be recovered.
+ * @param element_name The name of the element that should be recovered in the line.
+ *
+ * @returns The recovered value.
+ * @retval '\0' the NA value of the type if an error occurred and obi_errno is set.
+ *
+ * @since February 2016
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+const char* obi_column_get_obistr_with_elt_name_in_view(Obiview_p view, OBIDMS_column_p column, index_t line_nb, const char* element_name);
+
+
+#endif /* OBIDMSCOLUMN_STR_H_ */
 

src/obidmscolumndir.c

@@ -97,25 +97,14 @@ int obi_column_directory_exists(OBIDMS_p dms, const char* column_name)
 	char* 	column_directory_name;
 	char* 	full_path;
     int 	check_dir;
-	int 	dms_file_descriptor;
 
 	// Build and check the directory name
     column_directory_name = build_column_directory_name(column_name);
 	if (column_directory_name == NULL)
 		return -1;
 
-	// Get the file descriptor for the dms
-	dms_file_descriptor = dirfd(dms->directory);
-	if (dms_file_descriptor < 0)
-	{
-		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
-		obidebug(1, "\nError getting a file descriptor for an OBIDMS directory");
-		free(column_directory_name);
-		return -1;
-	}
-
 	// Get the full path for the column directory
-	full_path = get_full_path(dms_file_descriptor, column_directory_name);
+	full_path = get_full_path(dms->dir_fd, column_directory_name);
 	if (full_path == NULL)
 	{
 		obi_set_errno(OBICOLDIR_UNKNOWN_ERROR);
@@ -139,7 +128,6 @@ int obi_column_directory_exists(OBIDMS_p dms, const char* column_name)
 OBIDMS_column_directory_p obi_create_column_directory(OBIDMS_p dms, const char* column_name)
 {
 	char* 	column_directory_name;
-	int 	dms_file_descriptor;
 
 	// Build and check the directory name
 	column_directory_name = build_column_directory_name(column_name);
@@ -149,18 +137,8 @@ OBIDMS_column_directory_p obi_create_column_directory(OBIDMS_p dms, const char*
 		return NULL;
 	}
 
-	// Get the file descriptor for the dms
-	dms_file_descriptor = dirfd(dms->directory);
-	if (dms_file_descriptor < 0)
-	{
-		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
-		obidebug(1, "\nError getting a file descriptor for an OBIDMS directory");
-		free(column_directory_name);
-		return NULL;
-	}
-
 	// Try to create the directory
-	if (mkdirat(dms_file_descriptor, column_directory_name, 00777) < 0)
+	if (mkdirat(dms->dir_fd, column_directory_name, 00777) < 0)
 	{
 		if (errno == EEXIST)
 			obi_set_errno(OBICOLDIR_EXIST_ERROR);
@@ -182,7 +160,6 @@ OBIDMS_column_directory_p obi_open_column_directory(OBIDMS_p dms, const char* co
 	OBIDMS_column_directory_p	column_directory;
 	char*	 					column_directory_name;
 	DIR*						directory;
-	int 						dms_file_descriptor;
 
 	column_directory = NULL;
 
@@ -191,18 +168,8 @@ OBIDMS_column_directory_p obi_open_column_directory(OBIDMS_p dms, const char* co
 	if (column_directory_name == NULL)
 		return NULL;
 
-	// Get the file descriptor for the dms
-	dms_file_descriptor = dirfd(dms->directory);
-	if (dms_file_descriptor < 0)
-	{
-		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
-		obidebug(1, "\nError getting a file descriptor for an OBIDMS directory");
-		free(column_directory_name);
-		return NULL;
-	}
-
 	// Try to open the column directory
-	directory = private_opendirat(dms_file_descriptor, column_directory_name);
+	directory = private_opendirat(dms->dir_fd, column_directory_name);
 	if (directory == NULL) {
 		switch (errno)
 		{
@@ -241,6 +208,15 @@ OBIDMS_column_directory_p obi_open_column_directory(OBIDMS_p dms, const char* co
 	strcpy(column_directory->column_name, column_name);
 	column_directory->directory = directory;
 
+	column_directory->dir_fd = dirfd(directory);
+	if (column_directory->dir_fd < 0)
+	{
+		obi_set_errno(OBICOLDIR_MEMORY_ERROR);
+		obidebug(1, "\nError allocating the memory for an OBIDMS column directory structure");
+		free(column_directory_name);
+		free(column_directory);
+	}
+
 	free(column_directory_name);
 
 	return column_directory;

src/obidmscolumndir.h

@@ -21,7 +21,7 @@
 #include "obidms.h"
 
 
-#define OBIDMS_COLUMN_MAX_NAME (2048) /**< The maximum length of an OBIDMS column name.
+#define OBIDMS_COLUMN_MAX_NAME (1024) /**< The maximum length of an OBIDMS column name.
                                        */
 
 
@@ -33,16 +33,19 @@
  */
 typedef struct OBIDMS_column_directory {
 	OBIDMS_p	dms;	 										 /**< A pointer to a DMS instance.
-     	 	 	 	 	 	 	 	 	 	 	 	 		 	 */
+     	 	 	 	 	 	 	 	 	 	 	 	 		 	  */
 	char 		column_name[OBIDMS_COLUMN_MAX_NAME+1];		     /**< The name of the column
-																 *    contained in the directory.
-																 */
+																  *    contained in the directory.
+																  */
 	char		directory_name[OBIDMS_COLUMN_MAX_NAME+1];	     /**< The name of the directory
-	                                     	 	 	 	 	 	 *    containing the column.
-	                                     	 	 	 	 	 	 */
+	                                     	 	 	 	 	 	  *    containing the column.
+	                                     	 	 	 	 	 	  */
 	DIR*		directory;										 /**< A directory entry usable to
-										 	 	 	 	 	 	 *    refer and scan the database directory.
-										 	 	 	 	 	 	 */
+										 	 	 	 	 	 	  *    refer and scan the column directory.
+										 	 	 	 	 	 	  */
+	int			dir_fd;											 /**< The file descriptor of the directory entry
+																  *   usable to refer and scan the column directory.
+																  */
 } OBIDMS_column_directory_t, *OBIDMS_column_directory_p;
 
 

src/obierrno.h

@@ -65,25 +65,25 @@ extern int obi_errno;
                                     		 */
 #define OBIDMS_ACCESS_ERROR    		(6) 	/**< Permission error trying to access the database
                                     		 */
-
-#define OBICOLDIR_EXIST_ERROR     	(7) 	/**< Trying to create an OBIDMS column directory with a name
-                                    		 *   that corresponds to an existing one
-                                    		 */
-#define OBICOLDIR_NOT_EXIST_ERROR 	(8) 	/**< Trying to open a non-existing OBIDMS column directory
-                                    		 */
-#define OBICOLDIR_LONG_NAME_ERROR 	(9) 	/**< The specified OBIDMS column directory name is too long
-                                    		 */
-#define OBICOLDIR_MEMORY_ERROR    	(10) 	/**< A memory error occurred during allocation while handling
-											 *	 an OBIDMS column directory
-                                    	 	 */
-#define OBICOLDIR_UNKNOWN_ERROR   	(11) 	/**< Undetermined error while handling an OBIDMS column directory
-                                    	 	 */
-#define OBICOLDIR_ACCESS_ERROR    	(12) 	/**< Permission error trying to access an OBIDSM column directory
-                                    		 */
-
-#define OBICOL_BAD_ENDIAN_ERROR   	(13) 	/**< The opened data structure does not corresponds
+#define OBIDMS_BAD_ENDIAN_ERROR   	(7) 	/**< The opened database does not correspond
                                     		 *   to the endianness of the platform.
                                     		 */
+
+#define OBICOLDIR_EXIST_ERROR     	(8) 	/**< Trying to create an OBIDMS column directory with a name
+                                    		 *   that corresponds to an existing one
+                                    		 */
+#define OBICOLDIR_NOT_EXIST_ERROR 	(9) 	/**< Trying to open a non-existing OBIDMS column directory
+                                    		 */
+#define OBICOLDIR_LONG_NAME_ERROR 	(10) 	/**< The specified OBIDMS column directory name is too long
+                                    		 */
+#define OBICOLDIR_MEMORY_ERROR    	(11) 	/**< A memory error occurred during allocation while handling
+											 *	 an OBIDMS column directory
+                                    	 	 */
+#define OBICOLDIR_UNKNOWN_ERROR   	(12) 	/**< Undetermined error while handling an OBIDMS column directory
+                                    	 	 */
+#define OBICOLDIR_ACCESS_ERROR    	(13) 	/**< Permission error trying to access an OBIDSM column directory
+                                    		 */
+
 #define OBICOL_EXIST_ERROR     		(14) 	/**< Trying to create an OBIDMS column with a name
                                     		 *   that corresponds to an existing one
                                     		 */
@@ -98,8 +98,12 @@ extern int obi_errno;
                                     	 	 */
 #define OBICOL_ACCESS_ERROR    		(19) 	/**< Permission error trying to access an OBIDSM column directory
                                     		 */
-#define OBI_ARRAY_ERROR				(20)    /** Error while handling an array
+#define OBI_AVL_ERROR				(20)    /** Error while handling an AVL tree
 											 */
+#define OBIVIEW_ERROR				(21)    /** Error while handling an OBIView
+											 */
+#define OBI_TAXONOMY_ERROR			(22)    /** Error while handling binary taxonomy files
+ 	 	 	 	 	 	 	 	 	 	 	 */
 /**@}*/
 
 #endif /* OBIERRNO_H_ */

src/obitypes.c

@@ -40,7 +40,9 @@ size_t obi_sizeof(OBIType_t type)
 	case OBI_CHAR:  size = sizeof(obichar_t);
 					break;
 
-	case OBI_IDX:   size = sizeof(index_t);
+	case OBI_STR:   // fallthrough
+	case OBI_SEQ:   // fallthrough
+	case OBI_IDX:	size = sizeof(index_t);
 					break;
 
 	default:        size = 0;
@@ -90,6 +92,12 @@ char* name_data_type(int data_type)
 		case OBI_CHAR:  name = strdup("OBI_CHAR");
 						break;
 
+		case OBI_STR:   name = strdup("OBI_STR");
+						break;
+
+		case OBI_SEQ:   name = strdup("OBI_SEQ");
+						break;
+
 		case OBI_IDX:   name = strdup("OBI_IDX");
 						break;
 	}

src/obitypes.h

@@ -23,6 +23,8 @@
 #define OBIFloat_NA (float_NA())	/**< NA value for the type OBI_FLOAT */
 #define OBIChar_NA (0)				/**< NA value for the type OBI_CHAR */
 										 // TODO not sure about this one as it can be impossible to distinguish from uninitialized values
+#define OBISeq_NA ("\0")			// TODO
+#define OBIStr_NA ("\0")			// TODO
 
 
 /**
@@ -32,7 +34,7 @@ typedef enum OBIBool {
     FALSE      = 0,
     TRUE       = 1,
     OBIBool_NA = 2
-} obibool_t, *obibool_p; 		/**< a boolean true/false value */
+} obibool_t, *obibool_p; 		/**< a boolean true/false value */	// TODO check name convention?
 
 
 /**
@@ -44,7 +46,9 @@ typedef enum OBIType {
 	OBI_FLOAT,					/**< a floating value (C type : double) */
 	OBI_BOOL,					/**< a boolean true/false value, see obibool_t enum */
 	OBI_CHAR,					/**< a character (C type : char) */
-	OBI_IDX					    /**< an index in a data structure (C type : int64_t) */
+	OBI_STR,				    /**< an index in a data structure (C type : int64_t) referring to a character string */
+	OBI_SEQ,					/**< an index in a data structure (C type : int64_t) referring to a DNA sequence */
+	OBI_IDX					    /**< an index referring to a line in another column (C type : int64_t) */	// TODO delete?
 } OBIType_t, *OBIType_p;
 
 
@@ -52,6 +56,12 @@ typedef int64_t index_t;
 typedef int32_t obiint_t;
 typedef double obifloat_t;
 typedef char obichar_t;
+// TODO same for obistr_t and obiseq_t ?
+
+
+typedef char byte_t;		/**< Defining byte type since the data referred to by AVL trees is stored in bits
+							 *	 and char (stored on one byte) is the smallest addressable unit.
+ 	 	 	 	 	 	 	 */
 
 
 /**

src/obiview.h

@@ -0,0 +1,189 @@
+/********************************************************************
+ * Obiview header file                                              *
+ ********************************************************************/
+
+/**
+ * @file obiview.h
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ * @date 16 December 2015
+ * @brief Header file for the OBIDMS view functions and structures.
+ */
+
+
+#ifndef OBIVIEW_H_
+#define OBIVIEW_H_
+
+
+#include <stdlib.h>
+#include <errno.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <time.h>
+#include <math.h>
+
+#include "obidms.h"
+#include "obidmscolumn.h"
+#include "obierrno.h"
+
+
+#define OBIVIEW_NAME_MAX_LENGTH (1000)   /**< The maximum length of an OBIDMS view name.
+                                	 	  */
+#define OBIVIEW_COMMENTS_MAX_LENGTH (10000)
+
+#define OBIVIEW_FILE_NAME "obiviews"
+
+#define VIEW_TYPE_MAX_NAME (1024)
+
+#define VIEW_TYPE_NUC_SEQS "NUC_SEQS_VIEW"
+
+#define NUC_SEQUENCE_COLUMN "NUC_SEQ"
+#define NUC_SEQUENCE_AVL "NUC_SEQ_AVL"
+
+#define ID_COLUMN "ID"
+#define ID_AVL "ID_AVL"
+
+#define DESCRIPTION_COLUMN "DESCRIPTION"
+#define DESCRIPTION_AVL "DESCRIPTION_AVL"
+
+#define LINES_COLUMN_NAME "LINES"
+
+
+/**
+ * @brief .
+ */
+typedef struct Column_reference {
+	char 	   	 column_name[OBIDMS_COLUMN_MAX_NAME+1];    /**< Name of the column.
+	 	 	 	 	 	 	 	 	 	 	 	 	 	    */
+	obiversion_t version;				   		    	   /**< Version of the column.
+	 	 	 	 	 	 	 	 	 	 	 	 	 	    */
+} Column_reference_t, *Column_reference_p;
+
+
+/**
+ * @brief .
+ */
+typedef struct Obiview {
+
+	OBIDMS_p         	dms;
+
+	bool             	read_only;
+
+	OBIDMS_column_p  	line_selection;
+
+	OBIDMS_column_p  	new_line_selection;
+
+	index_t			 	line_count;
+
+	int    			 	column_count;
+
+	OBIDMS_column_p  	columns[MAX_NB_OPENED_COLUMNS];
+
+	char   				name[OBIVIEW_NAME_MAX_LENGTH+1];
+
+	char				created_from[OBIVIEW_NAME_MAX_LENGTH+1];
+
+	int					view_number;
+
+	char				view_type[VIEW_TYPE_MAX_NAME+1];
+
+	char 				comments[OBIVIEW_COMMENTS_MAX_LENGTH+1];
+
+} Obiview_t, *Obiview_p;
+
+
+/**
+ * @brief .
+ */
+typedef struct Obiview_infos {
+	int					view_number;
+
+	int 				column_count;
+
+	index_t             line_count;
+
+	char   				name[OBIVIEW_NAME_MAX_LENGTH+1];
+
+	char				created_from[OBIVIEW_NAME_MAX_LENGTH+1];
+
+	time_t 		 		creation_date;
+
+	bool				all_lines;
+
+	Column_reference_t	line_selection;
+
+	Column_reference_t	column_references[MAX_NB_OPENED_COLUMNS];
+
+	char				view_type[VIEW_TYPE_MAX_NAME+1];
+
+	char 				comments[OBIVIEW_COMMENTS_MAX_LENGTH+1];
+
+} Obiview_infos_t, *Obiview_infos_p;
+
+
+/**
+ * @brief .
+ */
+typedef struct Obiviews_header {
+	size_t header_size;
+	size_t views_size;
+	int    view_count;
+} Obiviews_header_t, *Obiviews_header_p;
+
+
+/**
+ * @brief .
+ */
+typedef struct Obiviews_infos_all {
+	Obiviews_header_p	header;
+
+	Obiview_infos_p 	view_infos;
+
+} Obiviews_infos_all_t, *Obiviews_infos_all_p;
+
+
+Obiview_p obi_new_view_nuc_seqs(OBIDMS_p dms, const char* view_name, Obiview_p view_to_clone, index_t* line_selection, const char* comments);
+
+Obiview_p obi_new_view(OBIDMS_p dms, const char* view_name, Obiview_p view_to_clone, index_t* line_selection, const char* comments);
+
+Obiview_p obi_new_view_cloned_from_name(OBIDMS_p dms, const char* view_name, const char* view_to_clone_name, index_t* line_selection, const char* comments);
+
+Obiview_p obi_new_view_nuc_seqs_cloned_from_name(OBIDMS_p dms, const char* view_name, const char* view_to_clone_name, index_t* line_selection, const char* comments);
+
+Obiview_p obi_open_view(OBIDMS_p dms, const char* view_name);
+
+Obiviews_infos_all_p obi_read_views(OBIDMS_p dms);
+
+int obi_unmap_read_views(Obiviews_infos_all_p views);
+
+int obi_view_add_column(Obiview_p view,
+						const char* column_name,
+						obiversion_t version_number,
+						OBIType_t data_type,
+						index_t nb_lines,
+						index_t nb_elements_per_line,
+						const char* elements_names,
+						const char* avl_name,
+						const char* comments,
+						bool create);
+
+int obi_view_delete_column(Obiview_p view, const char* column_name);
+
+int obi_select_line(Obiview_p view, index_t line_nb);
+
+int obi_select_lines(Obiview_p view, index_t* line_nbs);
+
+int obi_view_update_lines(Obiview_p view, index_t line_count);
+
+OBIDMS_column_p obi_view_clone_column(Obiview_p view, const char* column_name);
+
+OBIDMS_column_p obi_view_get_column(Obiview_p view, const char* column_name);
+
+OBIDMS_column_p* obi_view_get_pointer_on_column_in_view(Obiview_p view, const char* column_name);
+
+int obi_save_view(Obiview_p view);
+
+int obi_close_view(Obiview_p view);
+
+int obi_save_and_close_view(Obiview_p view);
+
+#endif /* OBIVIEW_H_ */