export: fixed a bug where exporting to tab format with a header would

not export the first line of data and switch to version 3.0.0b39

python/obitools3/apps/optiongroups/__init__.py

@@ -39,6 +39,12 @@ def __addImportInputOption(optionManager):
                      const=b'fastq',
                      help="Input file is in fastq format")
 
+    group.add_argument('--silva-input',
+                     action="store_const", dest="obi:inputformat",
+                     default=None,
+                     const=b'silva',
+                     help="Input file is in SILVA fasta format")
+
     group.add_argument('--embl-input',
                      action="store_const", dest="obi:inputformat",
                      default=None,

python/obitools3/commands/addtaxids.pyx

@@ -0,0 +1,230 @@
+#cython: language_level=3
+
+from obitools3.apps.progress cimport ProgressBar  # @UnresolvedImport
+from obitools3.dms import DMS
+from obitools3.dms.view.view cimport View, Line_selection
+from obitools3.uri.decode import open_uri
+from obitools3.apps.optiongroups import addMinimalInputOption, addTaxonomyOption, addMinimalOutputOption, addNoProgressBarOption
+from obitools3.dms.view import RollbackException
+from obitools3.dms.column.column cimport Column
+from functools import reduce
+from obitools3.apps.config import logger
+from obitools3.utils cimport tobytes, str2bytes, tostr
+from io import BufferedWriter
+from obitools3.dms.capi.obiview cimport NUC_SEQUENCE_COLUMN, \
+                                        ID_COLUMN, \
+                                        DEFINITION_COLUMN, \
+                                        QUALITY_COLUMN, \
+                                        COUNT_COLUMN, \
+                                        TAXID_COLUMN
+from obitools3.dms.capi.obitypes cimport OBI_INT
+from obitools3.dms.capi.obitaxonomy cimport MIN_LOCAL_TAXID
+import time
+import math 
+import sys
+
+from cpython.exc cimport PyErr_CheckSignals
+
+ 
+__title__="Annotate sequences with their corresponding NCBI taxid found from the taxon scientific name."
+
+
+ 
+def addOptions(parser):
+    
+    addMinimalInputOption(parser)
+    addTaxonomyOption(parser)
+    addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
+
+    group=parser.add_argument_group('obi addtaxids specific options')
+
+    group.add_argument('-t', '--taxid-tag', 
+                       action="store", 
+                       dest="addtaxids:taxid_tag",
+                       metavar="<TAXID_TAG>",
+                       default=b"TAXID",
+                       help="Name of the tag to store the found taxid "
+                            "(default: 'TAXID'.")
+
+    group.add_argument('-n', '--taxon-name-tag', 
+                       action="store", 
+                       dest="addtaxids:taxon_name_tag",
+                       metavar="<SCIENTIFIC_NAME_TAG>",
+                       default=b"SCIENTIFIC_NAME",
+                       help="Name of the tag giving the scientific name of the taxon "
+                            "(default: 'SCIENTIFIC_NAME'.")
+
+    group.add_argument('-g', '--try-genus-match',
+                     action="store_true", dest="addtaxids:try_genus_match",
+                     default=False,
+                     help="Try matching the first word of <SCIENTIFIC_NAME_TAG> when can't find corresponding taxid for a taxon. "
+                          "If there is a match it is added in the 'parent_taxid' tag. (Can be used by 'obi taxonomy' to add the taxon under that taxid).")
+
+    group.add_argument('-a', '--restricting-ancestor', 
+                       action="store", 
+                       dest="addtaxids:restricting_ancestor",
+                       metavar="<RESTRICTING_ANCESTOR>",
+                       default=None,
+                       help="Enables to restrict the search of taxids under an ancestor specified by its taxid.")
+
+    group.add_argument('-l', '--log-file', 
+                       action="store", 
+                       dest="addtaxids:log_file",
+                       metavar="<LOG_FILE>",
+                       default='',
+                       help="Path to a log file to write informations about not found taxids.")
+
+
+def run(config):
+     
+    DMS.obi_atexit()
+    
+    logger("info", "obi addtaxids")
+
+    # Open the input
+    input = open_uri(config['obi']['inputURI'])
+    if input is None:
+        raise Exception("Could not read input view")
+    i_dms = input[0]
+    i_view = input[1]
+    i_view_name = input[1].name
+
+    # Open the output: only the DMS, as the output view is going to be created by cloning the input view
+    # (could eventually be done via an open_uri() argument)
+    output = open_uri(config['obi']['outputURI'],
+                      input=False,
+                      dms_only=True)
+    if output is None:
+        raise Exception("Could not create output view")
+    o_dms = output[0]
+    output_0 = output[0]
+    o_view_name = output[1]
+
+    # stdout output: create temporary view
+    if type(output_0)==BufferedWriter:
+        o_dms = i_dms
+        i=0
+        o_view_name = b"temp"
+        while o_view_name in i_dms: # Making sure view name is unique in output DMS
+            o_view_name = o_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        imported_view_name = o_view_name
+
+    # If the input and output DMS are not the same, import the input view in the output DMS before cloning it to modify it
+    # (could be the other way around: clone and modify in the input DMS then import the new view in the output DMS)
+    if i_dms != o_dms:
+        imported_view_name = i_view_name
+        i=0
+        while imported_view_name in o_dms:  # Making sure view name is unique in output DMS
+            imported_view_name = i_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], i_view_name, imported_view_name)
+        i_view = o_dms[imported_view_name]
+
+    # Clone output view from input view
+    o_view = i_view.clone(o_view_name)
+    if o_view is None:
+        raise Exception("Couldn't create output view")
+    i_view.close()
+
+    # Open taxonomy
+    taxo_uri = open_uri(config['obi']['taxoURI'])
+    if taxo_uri is None or taxo_uri[2] == bytes:
+        raise Exception("Couldn't open taxonomy")
+    taxo = taxo_uri[1]
+
+    # Initialize the progress bar
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(len(o_view), config)
+    else:
+        pb = None
+        
+    try:
+        if config['addtaxids']['log_file']:
+            logfile = open(config['addtaxids']['log_file'], 'w')
+        else:
+            logfile = None
+        if config['addtaxids']['try_genus_match']:
+            try_genus = True
+        else:
+            try_genus = False
+        if 'restricting_ancestor' in config['addtaxids']:
+            res_anc = int(config['addtaxids']['restricting_ancestor'])
+        else:
+            res_anc = None
+        taxid_column_name = config['addtaxids']['taxid_tag']
+        parent_taxid_column_name = "PARENT_TAXID" # TODO macro
+        taxon_name_column_name = config['addtaxids']['taxon_name_tag']
+        taxid_column = Column.new_column(o_view, taxid_column_name, OBI_INT)
+        parent_taxid_column = Column.new_column(o_view, parent_taxid_column_name, OBI_INT)
+        taxon_name_column = o_view[taxon_name_column_name]
+        
+        found_count = 0
+        not_found_count = 0
+        parent_found_count = 0
+        
+        for i in range(len(o_view)):
+            PyErr_CheckSignals()
+            if pb is not None:
+                pb(i)
+            taxon_name = taxon_name_column[i]
+            taxon = taxo.get_taxon_by_name(taxon_name, res_anc)
+            if taxon is not None:
+                taxid_column[i] = taxon.taxid
+                found_count+=1
+            elif try_genus: # try finding genus or other parent taxon from the first word
+                taxon_name_sp = taxon_name.split(b" ")
+                taxon = taxo.get_taxon_by_name(taxon_name_sp[0], res_anc)
+                if taxon is not None:
+                    parent_taxid_column[i] = taxon.taxid
+                    parent_found_count+=1
+                    if logfile:
+                        print("Found parent taxon for", tostr(taxon_name), file=logfile)
+                else:
+                    not_found_count+=1
+                    if logfile:
+                        print("No taxid found for", tostr(taxon_name), file=logfile)
+            else:
+                not_found_count+=1
+                if logfile:
+                    print("No taxid found for", tostr(taxon_name), file=logfile)
+
+    except Exception, e:
+        raise RollbackException("obi addtaxids error, rollbacking view: "+str(e), o_view)
+
+    if pb is not None:
+        pb(i, force=True)
+        print("", file=sys.stderr)
+    
+    logger("info", "\nTaxids found: "+str(found_count)+"/"+str(len(o_view))+" ("+str(round(found_count*100.0/len(o_view), 2))+"%)")
+    if config['addtaxids']['try_genus_match']:
+        logger("info", "\nParent taxids found: "+str(parent_found_count)+"/"+str(len(o_view))+" ("+str(round(parent_found_count*100.0/len(o_view), 2))+"%)")
+    logger("info", "\nTaxids not found: "+str(not_found_count)+"/"+str(len(o_view))+" ("+str(round(not_found_count*100.0/len(o_view), 2))+"%)") 
+    
+    # Save command config in View and DMS comments
+    command_line = " ".join(sys.argv[1:])
+    input_dms_name=[input[0].name]
+    input_view_name=[i_view_name]
+    if 'taxoURI' in config['obi'] and config['obi']['taxoURI'] is not None:
+        input_dms_name.append(config['obi']['taxoURI'].split("/")[-3])
+        input_view_name.append("taxonomy/"+config['obi']['taxoURI'].split("/")[-1])
+    o_view.write_config(config, "addtaxids", command_line, input_dms_name=input_dms_name, input_view_name=input_view_name)
+    o_dms.record_command_line(command_line)
+
+    #print("\n\nOutput view:\n````````````", file=sys.stderr)
+    #print(repr(o_view), file=sys.stderr)
+
+    # stdout output: write to buffer
+    if type(output_0)==BufferedWriter:
+        logger("info", "Printing to output...")
+        o_view.print_to_output(output_0, noprogressbar=config['obi']['noprogressbar'])
+        o_view.close()
+
+    # If the input and the output DMS are different or if stdout output, delete the temporary imported view used to create the final view
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        View.delete_view(o_dms, imported_view_name)
+        o_dms.close(force=True)
+    i_dms.close(force=True)
+
+    logger("info", "Done.")

python/obitools3/commands/cat.pyx

@@ -4,7 +4,7 @@ from obitools3.apps.progress cimport ProgressBar  # @UnresolvedImport
 from obitools3.dms import DMS
 from obitools3.dms.view.view cimport View
 from obitools3.uri.decode import open_uri
-from obitools3.apps.optiongroups import addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalOutputOption, addNoProgressBarOption
 from obitools3.dms.view import RollbackException
 from obitools3.apps.config import logger
 from obitools3.utils cimport str2bytes
@@ -28,6 +28,7 @@ __title__="Concatenate views."
 def addOptions(parser):
     
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
 
     group=parser.add_argument_group('obi cat specific options')
 
@@ -47,9 +48,9 @@ def run(config):
     
     logger("info", "obi cat")
 
-    # Open the views to concatenate
+    # Check the views to concatenate
-    iview_list = []
     idms_list = []
+    iview_list = []
     total_len = 0
     remove_qual = False
     remove_rev_qual = False
@@ -67,8 +68,9 @@ def run(config):
         if REVERSE_QUALITY_COLUMN not in i_view: # same as above for reverse quality
             remove_rev_qual = True
         total_len += len(i_view)
-        iview_list.append(i_view)
         idms_list.append(i_dms)
+        iview_list.append(i_view.name)
+        i_view.close()
 
     # Open the output: only the DMS
     output = open_uri(config['obi']['outputURI'],
@@ -97,8 +99,10 @@ def run(config):
     # Initialize multiple elements columns
     if type(output_0)==BufferedWriter: 
         dict_cols = {}
-        for v in iview_list:
+        for v_uri in config["cat"]["views_to_cat"]:
+            v = open_uri(v_uri)[1]
             for coln in v.keys():
+                col = v[coln]
                 if v[coln].nb_elements_per_line > 1:
                     if coln not in dict_cols:
                         dict_cols[coln] = {}
@@ -108,6 +112,7 @@ def run(config):
                     else:
                         dict_cols[coln]['eltnames'] = set(v[coln].elements_names + list(dict_cols[coln]['eltnames']))
                         dict_cols[coln]['nbelts'] = len(dict_cols[coln]['eltnames'])
+            v.close()            
         for coln in dict_cols:
             Column.new_column(o_view, coln, dict_cols[coln]['obitype'], 
                               nb_elements_per_line=dict_cols[coln]['nbelts'], elements_names=list(dict_cols[coln]['eltnames']))
@@ -119,7 +124,8 @@ def run(config):
         pb = None
     
     i = 0
-    for v in iview_list:
+    for v_uri in config["cat"]["views_to_cat"]:
+        v = open_uri(v_uri)[1]
         for entry in v:
             PyErr_CheckSignals()
             if pb is not None:
@@ -130,6 +136,7 @@ def run(config):
             else:
                 o_view[i] = entry
             i+=1
+        v.close()
 
     # Deletes quality columns if needed
     if type(output_0)!=BufferedWriter: 
@@ -144,7 +151,7 @@ def run(config):
     
     # Save command config in DMS comments
     command_line = " ".join(sys.argv[1:])
-    o_view.write_config(config, "cat", command_line, input_dms_name=[d.name for d in idms_list], input_view_name=[v.name for v in iview_list])
+    o_view.write_config(config, "cat", command_line, input_dms_name=[d.name for d in idms_list], input_view_name=[vname for vname in iview_list])
     o_dms.record_command_line(command_line)
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)

python/obitools3/commands/ecotag.pyx

@@ -41,6 +41,17 @@ def addOptions(parser):
                       help="Minimum identity to consider for assignment, as a normalized identity, e.g. 0.95 for an identity of 95%%. "
                            "Default: 0.00 (no threshold).")
 
+    group.add_argument('--minimum-circle','-c',
+                      action="store", dest="ecotag:bubble_threshold",
+                      metavar='<CIRCLE_THRESHOLD>',
+                      default=0.99,
+                      type=float,
+                      help="Minimum identity considered for the assignment circle "
+                           "(sequence is assigned to the LCA of all sequences within a similarity circle of the best matches; "
+                           "the threshold for this circle is the highest value between <CIRCLE_THRESHOLD> and the best assignment score found for the query sequence). "
+                           "Give value as a normalized identity, e.g. 0.95 for an identity of 95%%. "
+                           "Default: 0.99.")
+
 def run(config):
         
     DMS.obi_atexit()
@@ -66,9 +77,8 @@ def run(config):
     ref_view_name = ref[1]
 
     # Check that the threshold demanded is greater than or equal to the threshold used to build the reference database
-    if config['ecotag']['threshold'] < eval(ref_dms[ref_view_name].comments["ref_db_threshold"]) :
+    if config['ecotag']['bubble_threshold'] < eval(ref_dms[ref_view_name].comments["ref_db_threshold"]) :
-        print("Error: The threshold demanded (%f) is lower than the threshold used to build the reference database (%f).", 
+        raise Exception(f"Error: The threshold demanded ({config['ecotag']['bubble_threshold']}) is lower than the threshold used to build the reference database ({float(ref_dms[ref_view_name].comments['ref_db_threshold'])}).")
-              config['ecotag']['threshold'], ref_dms[ref_view_name].comments["ref_db_threshold"])
     
     # Open the output: only the DMS
     output = open_uri(config['obi']['outputURI'],
@@ -113,8 +123,9 @@ def run(config):
     if obi_ecotag(i_dms.name_with_full_path, tobytes(i_view_name), \
                   ref_dms.name_with_full_path, tobytes(ref_view_name), \
                   taxo_dms.name_with_full_path, tobytes(taxonomy_name), \
-                  tobytes(o_view_name), comments, 
+                  tobytes(o_view_name), comments, \
-                  config['ecotag']['threshold']) < 0:
+                  config['ecotag']['threshold'], \
+                  config['ecotag']['bubble_threshold']) < 0:
         raise Exception("Error running ecotag")
     
     # If the input and output DMS are not the same, export result view to output DMS

python/obitools3/commands/grep.pyx

@@ -184,7 +184,7 @@ def Filter_generator(options, tax_filter, i_view):
     invert_selection = options["invert_selection"]
     id_set = None
     if "id_list" in options:
-        id_set = set(x.strip() for x in open(options["id_list"]))
+        id_set = set(x.strip() for x in open(options["id_list"], 'rb'))
 
     # Initialize the regular expression patterns
     seq_pattern = None

python/obitools3/commands/import.pyx

@@ -26,13 +26,15 @@ from obitools3.dms.capi.obiview cimport VIEW_TYPE_NUC_SEQS, \
                                         QUALITY_COLUMN, \
                                         COUNT_COLUMN, \
                                         TAXID_COLUMN, \
-                                        MERGED_PREFIX
+                                        MERGED_PREFIX, \
+                                        SCIENTIFIC_NAME_COLUMN
                                         
 from obitools3.dms.capi.obidms cimport obi_import_view
 
 from obitools3.dms.capi.obitypes cimport obitype_t, \
                                          OBI_VOID, \
-                                         OBI_QUAL
+                                         OBI_QUAL, \
+                                         OBI_STR
 
 from obitools3.dms.capi.obierrno cimport obi_errno
 
@@ -94,6 +96,7 @@ def run(config):
     cdef   obitype_t   new_type
     cdef   bint        get_quality
     cdef   bint        NUC_SEQS_view
+    cdef   bint        silva
     cdef   int         nb_elts
     cdef   object      d
     cdef   View        view
@@ -104,6 +107,8 @@ def run(config):
     cdef   Column      seq_col
     cdef   Column      qual_col
     cdef   Column      old_column
+    cdef   Column      sci_name_col
+    cdef   bytes       sci_name
     cdef   bint        rewrite
     cdef   dict        dcols
     cdef   int         skipping
@@ -203,9 +208,16 @@ def run(config):
         id_col = view[ID_COLUMN]
         def_col = view[DEFINITION_COLUMN]
         seq_col = view[NUC_SEQUENCE_COLUMN]
-        
+
+    # Prepare taxon scientific name if SILVA file
+    if 'inputformat' in config['obi'] and config['obi']['inputformat'] == b"silva":
+        silva = True
+        sci_name_col = Column.new_column(view, SCIENTIFIC_NAME_COLUMN, OBI_STR)
+    else:
+        silva = False
+    
     dcols = {}
-        
+
     # First read through the entries to prepare columns with dictionaries as they are very time-expensive to rewrite
     if config['import']['preread']:
         logger("info", "First readthrough...")
@@ -282,7 +294,7 @@ def run(config):
         try:
              
             if NUC_SEQS_view: 
-                id_col[i] = entry.id
+                id_col[i] = entry.id           
                 def_col[i] = entry.definition
                 seq_col[i] = entry.seq
                 # Check if there is a sequencing quality associated by checking the first entry    # TODO haven't found a more robust solution yet
@@ -293,6 +305,11 @@ def run(config):
                         qual_col = view[QUALITY_COLUMN]
                 if get_quality:
                     qual_col[i] = entry.quality
+                    
+                # Parse taxon scientific name if SILVA file
+                if silva:
+                    sci_name = entry.definition.split(b";")[-1]
+                    sci_name_col[i] = sci_name
              
             for tag in entry :
                 

python/obitools3/commands/ls.pyx

@@ -31,27 +31,11 @@ def run(config):
     input = open_uri(config['obi']['inputURI'])
     if input is None:
         raise Exception("Could not read input")
-    if input[2] == DMS and not config['ls']['longformat']:
+    
-        dms = input[0]
+    # Print representation
-        l = []
+    if config['ls']['longformat']:
-        for viewname in input[0]:
+        print(input[1].repr_longformat())
-            view = dms[viewname]
-            l.append(tostr(viewname) + "\t(Date created: " + str(bytes2str_object(view.comments["Date created"]))+")")
-            view.close()
-        l.sort()
-        for v in l:
-            print(v)
     else:
         print(repr(input[1]))
-    if input[2] == DMS:
-        taxolist = ["\n### Taxonomies:"]
-        for t in Taxonomy.list_taxos(input[0]):
-            taxolist.append("\t"+tostr(t))
-        if len(taxolist) > 1:
-            for t in taxolist:
-                print(t)
-    if config['ls']['longformat'] and len(input[1].comments) > 0:
-        print("\n### Comments:")
-        print(str(input[1].comments))
     
     input[0].close(force=True)

python/obitools3/commands/ngsfilter.pyx

@@ -24,10 +24,6 @@ from cpython.exc cimport PyErr_CheckSignals
 from io import BufferedWriter
 
 
-#REVERSE_SEQ_COLUMN_NAME = b"REVERSE_SEQUENCE"      # used by alignpairedend tool
-#REVERSE_QUALITY_COLUMN_NAME = b"REVERSE_QUALITY"   # used by alignpairedend tool
-
-
 __title__="Assigns sequence records to the corresponding experiment/sample based on DNA tags and primers"
 
 

python/obitools3/dms/capi/obidms.pxd

@@ -34,6 +34,7 @@ cdef extern from "obidms.h" nogil:
     int obi_close_dms(OBIDMS_p dms, bint force)
     char* obi_dms_get_dms_path(OBIDMS_p dms)
     char* obi_dms_get_full_path(OBIDMS_p dms, const_char_p path_name)
+    char* obi_dms_formatted_infos(OBIDMS_p dms, bint detailed)
     void obi_close_atexit()
     
     obiversion_t obi_import_column(const char* dms_path_1, const char* dms_path_2, const char* column_name, obiversion_t version_number)

python/obitools3/dms/capi/obidmscolumn.pxd

@@ -63,10 +63,11 @@ cdef extern from "obidmscolumn.h" nogil:
     
     char* obi_get_elements_names(OBIDMS_column_p column)
     
-    char* obi_column_formatted_infos(OBIDMS_column_p column)
-
     index_t obi_column_get_element_index_from_name(OBIDMS_column_p column, const char* element_name)
     
     int obi_column_write_comments(OBIDMS_column_p column, const char* comments)
 
     int obi_column_add_comment(OBIDMS_column_p column, const char* key, const char* value)
+
+    char* obi_column_formatted_infos(OBIDMS_column_p column, bint detailed)
+    

python/obitools3/dms/capi/obiecotag.pxd

@@ -11,4 +11,5 @@ cdef extern from "obi_ecotag.h" nogil:
                    const char* taxonomy_name,
                    const char* output_view_name,
                    const char* output_view_comments,
-                   double ecotag_threshold)
+                   double ecotag_threshold,
+                   double bubble_threshold)

python/obitools3/dms/capi/obitaxonomy.pxd

@@ -7,6 +7,8 @@ from libc.stdint   cimport int32_t
 
 cdef extern from "obidms_taxonomy.h" nogil:
 
+    extern int MIN_LOCAL_TAXID
+    
     struct ecotxnode :
         int32_t           taxid
         int32_t           rank
@@ -18,6 +20,13 @@ cdef extern from "obidms_taxonomy.h" nogil:
     ctypedef ecotxnode ecotx_t
 
 
+    struct econame_t : # can't get this struct to be accepted by Cython ('unknown size')
+        char*      name
+        char*      class_name
+        int32_t    is_scientific_name
+        ecotxnode* taxon
+
+
     struct ecotxidx_t :   
         int32_t  count
         int32_t  max_taxid
@@ -30,9 +39,14 @@ cdef extern from "obidms_taxonomy.h" nogil:
         char**   label
 
 
+    struct econameidx_t :   
+        int32_t    count
+        econame_t* names
+
+
     struct OBIDMS_taxonomy_t :
         ecorankidx_t* ranks
-#        econameidx_t* names
+        econameidx_t* names
         ecotxidx_t*   taxa
 
     ctypedef OBIDMS_taxonomy_t* OBIDMS_taxonomy_p
@@ -51,7 +65,11 @@ cdef extern from "obidms_taxonomy.h" nogil:
     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)
+
+    char* obi_taxo_get_name_from_name_idx(OBIDMS_taxonomy_p taxonomy, int32_t idx)
     
+    ecotx_t* obi_taxo_get_taxon_from_name_idx(OBIDMS_taxonomy_p taxonomy, int32_t idx)
+
     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)
@@ -71,4 +89,4 @@ cdef extern from "obidms_taxonomy.h" nogil:
     int obi_taxo_add_preferred_name_with_taxon(OBIDMS_taxonomy_p tax, ecotx_t* taxon, const char* preferred_name)
 
     const char* obi_taxo_rank_index_to_label(int32_t rank_idx, ecorankidx_t* ranks)
-    
+

python/obitools3/dms/capi/obiview.pxd

@@ -27,6 +27,7 @@ cdef extern from "obiview.h" nogil:
     extern const_char_p REVERSE_QUALITY_COLUMN
     extern const_char_p REVERSE_SEQUENCE_COLUMN
     extern const_char_p COUNT_COLUMN
+    extern const_char_p SCIENTIFIC_NAME_COLUMN
     extern const_char_p TAXID_COLUMN
     extern const_char_p MERGED_TAXID_COLUMN
     extern const_char_p MERGED_PREFIX
@@ -103,13 +104,17 @@ cdef extern from "obiview.h" nogil:
                             bint create)
 
     int obi_view_delete_column(Obiview_p view, const_char_p column_name, bint delete_file)
-            
+
     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_view_create_column_alias(Obiview_p view, const_char_p current_name, const_char_p alias)    
-    
+
+    char* obi_view_formatted_infos(Obiview_p view, bint detailed)
+
+    char* obi_view_formatted_infos_one_line(Obiview_p view)
+
     int obi_view_write_comments(Obiview_p view, const_char_p comments)
 
     int obi_view_add_comment(Obiview_p view, const_char_p key, const_char_p value)

python/obitools3/dms/column/column.pyx

@@ -302,15 +302,24 @@ cdef class Column(OBIWrapper) :
     
     @OBIWrapper.checkIsActive
     def __repr__(self) :
-        cdef bytes s
+        cdef str s
-        #cdef char* s_b
+        cdef char* sc
-        #cdef str s_str
+        cdef OBIDMS_column_p pointer = self.pointer()
-        #s_b = obi_column_formatted_infos(self.pointer())
+        sc = obi_column_formatted_infos(pointer, False)
-        #s_str = bytes2str(s_b)
+        s = bytes2str(sc)
-        #free(s_b)
+        free(sc)
-        s = self._alias + b", data type: " + self.data_type
+        return s
-        #return s_str
+
-        return bytes2str(s)
+
+    @OBIWrapper.checkIsActive
+    def repr_longformat(self) :
+        cdef str s
+        cdef char* sc
+        cdef OBIDMS_column_p pointer = self.pointer()
+        sc = obi_column_formatted_infos(pointer, True)
+        s = bytes2str(sc)
+        free(sc)
+        return s
     
          
     def close(self):  # TODO discuss, can't be called bc then bug when closing view that tries to close it in C

python/obitools3/dms/dms.pyx

@@ -10,7 +10,8 @@ from .capi.obidms cimport obi_open_dms, \
                           obi_dms_exists, \
                           obi_dms_get_full_path, \
                           obi_close_atexit, \
-                          obi_dms_write_comments
+                          obi_dms_write_comments, \
+                          obi_dms_formatted_infos
                                                       
 from .capi.obitypes cimport const_char_p
                             
@@ -32,6 +33,8 @@ from .object import OBIWrapper
 import json
 import time
 
+from libc.stdlib cimport free
+
 
 cdef class DMS(OBIWrapper):    
     
@@ -223,13 +226,24 @@ cdef class DMS(OBIWrapper):
 
 
     @OBIWrapper.checkIsActive
-    def __repr__(self):
+    def __repr__(self) :
-        cdef str s 
+        cdef str s
-        s=""
+        cdef char* sc
-        for view_name in self.keys():
+        cdef OBIDMS_p pointer = self.pointer()
-            view = self.get_view(view_name)
+        sc = obi_dms_formatted_infos(pointer, False)
-            s = s + repr(view) + "\n"
+        s = bytes2str(sc)
-            view.close()
+        free(sc)
+        return s
+
+
+    @OBIWrapper.checkIsActive
+    def repr_longformat(self) :
+        cdef str s
+        cdef char* sc
+        cdef OBIDMS_p pointer = self.pointer()
+        sc = obi_dms_formatted_infos(pointer, True)
+        s = bytes2str(sc)
+        free(sc)
         return s
     
 

python/obitools3/dms/taxo/taxo.pxd

@@ -11,11 +11,14 @@ cdef class Taxonomy(OBIWrapper) :
     cdef bytes  _name
     cdef DMS    _dms
     cdef list   _ranks
+    cdef dict   _name_dict
     
     cdef inline OBIDMS_taxonomy_p pointer(self)
+    cdef fill_name_dict(self)
 
     cpdef Taxon get_taxon_by_idx(self, int idx)
     cpdef Taxon get_taxon_by_taxid(self, int taxid)
+    cpdef Taxon get_taxon_by_name(self, object taxon_name, object restricting_taxid=*)
     cpdef write(self, object prefix)
     cpdef int add_taxon(self, str name, str rank_name, int parent_taxid, int min_taxid=*)
     cpdef object get_species(self, int taxid)

python/obitools3/dms/taxo/taxo.pyx

@@ -15,7 +15,11 @@ from ..capi.obitaxonomy cimport obi_taxonomy_exists, \
                                 obi_taxo_get_species, \
                                 obi_taxo_get_genus, \
                                 obi_taxo_get_family, \
-                                ecotx_t                
+                                ecotx_t, \
+                                econame_t, \
+                                obi_taxo_get_name_from_name_idx, \
+                                obi_taxo_get_taxon_from_name_idx
+                                
                                            
 from cpython.pycapsule cimport PyCapsule_New, PyCapsule_GetPointer
 import tarfile
@@ -24,11 +28,29 @@ from libc.stdlib  cimport free
 
 
 cdef class Taxonomy(OBIWrapper) :
-    # TODO function to import taxonomy?
+    # TODO function to import taxonomy?    
-    
+     
     cdef inline OBIDMS_taxonomy_p pointer(self) :
         return <OBIDMS_taxonomy_p>(self._pointer)        
 
+    cdef fill_name_dict(self):
+        print("Indexing taxon names...")
+        
+        cdef OBIDMS_taxonomy_p pointer = self.pointer()
+        cdef ecotx_t*     taxon_p
+        cdef object       taxon_capsule
+        cdef bytes        name
+        cdef int          count
+        cdef int          n
+        
+        count = (<OBIDMS_taxonomy_p>pointer).names.count
+        
+        for n in range(count) :
+            name = obi_taxo_get_name_from_name_idx(pointer, n)
+            taxon_p = obi_taxo_get_taxon_from_name_idx(pointer, n)
+            taxon_capsule = PyCapsule_New(taxon_p, NULL, NULL)
+            self._name_dict[name] = Taxon(taxon_capsule, self)
+
 
     @staticmethod
     def exists(DMS dms, object name) :
@@ -75,7 +97,8 @@ cdef class Taxonomy(OBIWrapper) :
 
         taxo._dms = dms
         taxo._name = tobytes(name)
-        
+        taxo._name_dict = {}
+        taxo.fill_name_dict()
         taxo._ranks = []
         for r in range((<OBIDMS_taxonomy_p>pointer).ranks.count) :
             taxo._ranks.append(obi_taxo_rank_index_to_label(r, (<OBIDMS_taxonomy_p>pointer).ranks))
@@ -118,7 +141,8 @@ cdef class Taxonomy(OBIWrapper) :
 
         taxo._dms = dms
         taxo._name = folder_path
-        
+        taxo._name_dict = {}
+        taxo.fill_name_dict()
         taxo._ranks = []
         for r in range((<OBIDMS_taxonomy_p>pointer).ranks.count) :
             taxo._ranks.append(obi_taxo_rank_index_to_label(r, (<OBIDMS_taxonomy_p>pointer).ranks))
@@ -129,8 +153,8 @@ cdef class Taxonomy(OBIWrapper) :
     def __getitem__(self, object ref):     
         if type(ref) == int :
             return self.get_taxon_by_taxid(ref)
-        else :
+        elif type(ref) == str or type(ref) == bytes :
-            raise NotImplementedError()
+            return self.get_taxon_by_name(ref)
 
 
     cpdef Taxon get_taxon_by_taxid(self, int taxid):
@@ -143,6 +167,19 @@ cdef class Taxonomy(OBIWrapper) :
         return Taxon(taxon_capsule, self)
 
 
+    cpdef Taxon get_taxon_by_name(self, object taxon_name, object restricting_taxid=None):            
+        taxon = self._name_dict.get(tobytes(taxon_name), None)
+        if not taxon:
+            return None
+        elif restricting_taxid:
+            if self.is_ancestor(restricting_taxid, taxon.taxid):
+                return taxon
+            else:
+                return None
+        else:
+            return taxon
+
+
     cpdef Taxon get_taxon_by_idx(self, int idx):
         cdef ecotx_t* taxa
         cdef ecotx_t* taxon_p
@@ -232,7 +269,7 @@ cdef class Taxonomy(OBIWrapper) :
  
         taxa = self.pointer().taxa.taxon
  
-        # Yield each taxid
+        # Yield each taxon
         for t in range(self.pointer().taxa.count):
             taxon_p = <ecotx_t*> (taxa+t)
             taxon_capsule = PyCapsule_New(taxon_p, NULL, NULL)

python/obitools3/dms/view/view.pyx

@@ -19,7 +19,9 @@ from ..capi.obiview cimport Alias_column_pair_p, \
                             obi_view_delete_column, \
                             obi_view_create_column_alias, \
                             obi_view_write_comments, \
-                            obi_delete_view
+                            obi_delete_view, \
+                            obi_view_formatted_infos, \
+                            obi_view_formatted_infos_one_line
                            
 from ..capi.obidmscolumn cimport OBIDMS_column_p
 from ..capi.obidms cimport OBIDMS_p
@@ -59,6 +61,8 @@ import pkgutil
 import json
 import sys
 
+from libc.stdlib cimport free
+
 
 cdef class View(OBIWrapper) :
 
@@ -186,15 +190,22 @@ cdef class View(OBIWrapper) :
     @OBIWrapper.checkIsActive
     def __repr__(self) :
         cdef str s
-        if self.read_only:   # can read date   
+        cdef char* sc
-            s = "#View name:\n{name:s}\n#Date created:\n{date:s}\n#Line count:\n{line_count:d}\n#Columns:\n".format(name       = bytes2str(self.name),
+        cdef Obiview_p pointer = self.pointer()
-                                                                                                                    line_count = self.line_count,
+        sc = obi_view_formatted_infos(pointer, False)
-                                                                                                                    date       = str(bytes2str_object(self.comments["Date created"])))        
+        s = bytes2str(sc)
-        else:
+        free(sc)
-            s = "#View name:\n{name:s}\n#Line count:\n{line_count:d}\n#Columns:\n".format(name       = bytes2str(self.name),
+        return s
-                                                                                          line_count = self.line_count)
+
-        for column_name in self.keys() :
+
-            s = s + repr(self[column_name]) + '\n'
+    @OBIWrapper.checkIsActive
+    def repr_longformat(self) :
+        cdef str s
+        cdef char* sc
+        cdef Obiview_p pointer = self.pointer()
+        sc = obi_view_formatted_infos(pointer, True)
+        s = bytes2str(sc)
+        free(sc)
         return s
 
 

python/obitools3/format/tab.pyx

@@ -23,9 +23,8 @@ cdef class TabFormat:
         if self.first_line:
             self.tags = [k for k in data.keys()]
         
-        for k in self.tags:
+        if self.header and self.first_line:
-            
+            for k in self.tags:
-            if self.header and self.first_line:
                 if isinstance(data.view[k], Column_multi_elts):
                     keys = data.view[k].keys()
                     keys.sort()
@@ -33,27 +32,36 @@ cdef class TabFormat:
                         line.append(tobytes(k)+b':'+tobytes(k2))
                 else:
                     line.append(tobytes(k))
-            else:
+            r = self.sep.join(value for value in line)
-                value = data[k]
+            r += b'\n'
-                if isinstance(data.view[k], Column_multi_elts):
+            line = []
-                    keys = data.view[k].keys()
+                    
-                    keys.sort()
+        for k in self.tags:
-                    if value is None:  # all keys at None
+            value = data[k]
-                        for k2 in keys: # TODO could be much more efficient
+            if isinstance(data.view[k], Column_multi_elts):
-                            line.append(self.NAString)
+                keys = data.view[k].keys()
-                    else:
+                keys.sort()
-                        for k2 in keys: # TODO could be much more efficient
+                if value is None:  # all keys at None
-                            if value[k2] is not None:
+                    for k2 in keys: # TODO could be much more efficient
-                                line.append(str2bytes(str(bytes2str_object(value[k2]))))  # genius programming
-                            else:
-                                line.append(self.NAString)
-                else:
-                    if value is not None:
-                        line.append(str2bytes(str(bytes2str_object(value))))
-                    else:
                         line.append(self.NAString)
-                  	
+                else:
+                    for k2 in keys: # TODO could be much more efficient
+                        if value[k2] is not None:
+                            line.append(str2bytes(str(bytes2str_object(value[k2]))))  # genius programming
+                        else:
+                            line.append(self.NAString)
+            else:
+                if value is not None:
+                    line.append(str2bytes(str(bytes2str_object(value))))
+                else:
+                    line.append(self.NAString)
+                  	      	
+        if self.header and self.first_line:
+            r += self.sep.join(value for value in line)
+        else:
+            r = self.sep.join(value for value in line)
+
         if self.first_line:
             self.first_line = False
-      		
+
-        return self.sep.join(value for value in line)
+        return r

python/obitools3/uri/decode.pyx

@@ -464,7 +464,7 @@ def open_uri(uri,
         if format is not None:
             if seqtype==b"nuc":
                 objclass = Nuc_Seq    # Nuc_Seq_Stored? TODO
-                if format==b"fasta":
+                if format==b"fasta" or format==b"silva":
                     if input:
                         iseq = fastaNucIterator(file, 
                                                 skip=skip, 

python/obitools3/version.py

@@ -1,5 +1,5 @@
 major = 3
 minor = 0
-serial= '0b32'
+serial= '0b39'
 
 version ="%d.%d.%s" % (major,minor,serial)

src/encode.c

@@ -36,10 +36,12 @@ bool only_ATGC(const char* seq)
 	{
 		if (!((*c == 'A') || \
 			  (*c == 'T') || \
+			  (*c == 'U') || \
 			  (*c == 'G') || \
 			  (*c == 'C') || \
 			  (*c == 'a') || \
 			  (*c == 't') || \
+			  (*c == 'u') || \
 			  (*c == 'g') || \
 			  (*c == 'c')))
 		{
@@ -182,6 +184,8 @@ byte_t* encode_seq_on_2_bits(const char* seq, int32_t length)
 			break;
 		case 't':
 		case 'T':
+		case 'u':
+		case 'U':
 			seq_b[i/4] |= NUC_T_2b;
 			break;
 		default:
@@ -288,6 +292,8 @@ byte_t* encode_seq_on_4_bits(const char* seq, int32_t length)
 			break;
 		case 't':
 		case 'T':
+		case 'u': // discussable
+		case 'U':
 			seq_b[i/2] |= NUC_T_4b;
 			break;
 		case 'r':

src/encode.h

@@ -64,7 +64,7 @@ enum
 
 
 /**
- * @brief Checks if there are only 'atgcATGC' characters in a
+ * @brief Checks if there are only 'atgcuATGCU' characters in a
  *        character string.
  *
  * @param seq The sequence to check.
@@ -129,12 +129,13 @@ byte_t get_nucleotide_from_encoded_seq(byte_t* seq, int32_t idx, uint8_t encodin
 /**
  * @brief Encodes a DNA sequence with each nucleotide coded on 2 bits.
  *
- *    A or a : 00
+ *    A or a :           00
- *    C or c : 01
+ *    C or c :           01
- *    T or t : 10
+ *    T or t or U or u : 10
- *    G or g : 11
+ *    G or g :           11
  *
- * @warning The DNA sequence must contain only 'atgcATGC' characters.
+ * @warning The DNA sequence must contain only 'atgcuATGCU' characters.
+ * @warning Uracil ('U') bases are encoded as Thymine ('T') bases.
  *
  * @param seq The sequence to encode.
  * @param length The length of the sequence to encode.
@@ -169,23 +170,24 @@ 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
+ *		A or a :           0001
- *      C or c : 0010
+ *      C or c :           0010
- *      G or g : 0011
+ *      G or g :           0011
- *      T or t : 0100
+ *      T or t or U or u : 0100
- *      R or r : 0101
+ *      R or r :           0101
- *      Y or y : 0110
+ *      Y or y :           0110
- *      S or s : 0111
+ *      S or s :           0111
- *      W or w : 1000
+ *      W or w :           1000
- *      K or k : 1001
+ *      K or k :           1001
- *      M or m : 1010
+ *      M or m :           1010
- *      B or b : 1011
+ *      B or b :           1011
- *      D or d : 1100
+ *      D or d :           1100
- *      H or h : 1101
+ *      H or h :           1101
- *      V or v : 1110
+ *      V or v :           1110
- *      N or n : 1111
+ *      N or n :           1111
  *
  * @warning The DNA sequence must contain only IUPAC characters.
+ * @warning Uracil ('U') bases are encoded as Thymine ('T') bases.
  *
  * @param seq The sequence to encode.
  * @param length The length of the sequence to encode.

src/obi_ecotag.c

@@ -218,7 +218,8 @@ int obi_ecotag(const char* dms_name,
 			   const char* taxonomy_name,
 			   const char* output_view_name,
 			   const char* output_view_comments,
-			   double ecotag_threshold)    // TODO different threshold for the similarity sphere around ref seqs
+			   double ecotag_threshold,
+			   double bubble_threshold)
 {
 
 	// For each sequence
@@ -239,6 +240,7 @@ int obi_ecotag(const char* dms_name,
 	index_t         query_seq_idx, ref_seq_idx;
 	double          score, best_score;
 	double			threshold;
+	double			lca_threshold;
 	int             lcs_length;
 	int             ali_length;
 	Kmer_table_p    ktable;
@@ -389,10 +391,10 @@ int obi_ecotag(const char* dms_name,
 		return -1;
 	}
 	free(db_threshold_str);
-	if (ecotag_threshold < db_threshold)
+	if (bubble_threshold < db_threshold)
 	{
 		fprintf(stderr, "\nError: The threshold demanded (%f) is lower than the threshold used to build the reference database (%f).\n\n",
-				ecotag_threshold, db_threshold);
+				bubble_threshold, db_threshold);
 		return -1;
 	}
 
@@ -597,11 +599,16 @@ int obi_ecotag(const char* dms_name,
 		{
 			best_match_idx = best_match_array[j];
 
-			// Find the LCA for the chosen threshold
+			// Find the LCA for the highest threshold between best_score and the chosen bubble threshold
 			score_array = obi_get_array_with_col_p_in_view(ref_view, score_a_column, best_match_idx, &lca_array_length);
 
+			if (bubble_threshold < best_score)
+				lca_threshold = best_score;
+			else
+				lca_threshold = bubble_threshold;
+
 			k = 0;
-			while ((k < lca_array_length) && (score_array[k] >= best_score))
+			while ((k < lca_array_length) && (score_array[k] >= lca_threshold))
 				k++;
 
 			if (k>0)

src/obi_ecotag.h

@@ -42,12 +42,14 @@
  * @param output_view_name The name to give to the output view.
  * @param output_view_comments The comments to associate to the output view.
  * @param ecotag_threshold The threshold at which to assign.
+ * @param bubble_threshold The threshold at which to look for an LCA (i.e. minimum identity considered for the assignment circle);
+ *                         the threshold actually used will be the highest between this value and the best assignment score found.
  *
  * 	The algorithm works like this:
  * 		For each query sequence:
  *			Align with reference database
  *			Keep the indices of all the best matches
- *			For each kept index, get the LCA at that threshold as stored in the reference database, then the LCA of those LCAs
+ *			For each kept index, get the LCA at the highest threshold between bubble_threshold and the best assignment score found (as stored in the reference database), then the LCA of those LCAs
  *			Write result (max score, threshold, taxid and scientific name of the LCA assigned, list of the ids of the best matches)
  *
  * @returns A value indicating the success of the operation.
@@ -65,7 +67,8 @@ int obi_ecotag(const char* dms_name,
 			   const char* taxonomy_name,
 			   const char* output_view_name,
 			   const char* output_view_comments,
-			   double ecotag_threshold);
+			   double ecotag_threshold,
+			   double bubble_threshold);
 
 
 #endif /* OBI_ECOTAG_H_ */

src/obidms.c

@@ -1409,6 +1409,107 @@ DIR* opendir_in_dms(OBIDMS_p dms, const char* path_name)
 }
 
 
+char* obi_dms_formatted_infos(OBIDMS_p dms, bool detailed)
+{
+	char*          dms_infos = NULL;
+	char*   	   view_infos = NULL;
+	char*          view_name = NULL;
+	char*          tax_name = NULL;
+	char*          all_tax_dir_path = NULL;
+	int            i;
+	struct dirent* dp;
+	Obiview_p      view;
+
+	// DMS name
+	dms_infos = (char*) malloc((strlen("# DMS name: ")+strlen(dms->dms_name)+strlen("\n# Views:\n")+1) * sizeof(char));
+	if (dms_infos == NULL)
+	{
+		obidebug(1, "\nError allocating memory for DMS formatted infos");
+		return NULL;
+	}
+	strcpy(dms_infos, "# DMS name: ");
+	strcat(dms_infos, dms->dms_name);
+	strcat(dms_infos, "\n# Views:\n");
+
+	// Go through views and get their infos
+	rewinddir(dms->view_directory);
+	while ((dp = readdir(dms->view_directory)) != NULL)
+	{
+		if ((dp->d_name)[0] == '.')
+			continue;
+		i=0;
+		while ((dp->d_name)[i] != '.')
+			i++;
+		view_name = (char*) malloc((i+1) * sizeof(char));
+		if (view_name == NULL)
+		{
+			obi_set_errno(OBI_MALLOC_ERROR);
+			obidebug(1, "\nError allocating memory for a view name when getting formatted DMS infos: file %s", dp->d_name);
+			return NULL;
+		}
+		strncpy(view_name, dp->d_name, i);
+		view_name[i] = '\0';
+		view = obi_open_view(dms, view_name);
+		if (view == NULL)
+		{
+			obidebug(1, "\nError opening a view to get DMS formatted infos");
+			return NULL;
+		}
+		if (detailed)
+			view_infos = obi_view_formatted_infos(view, detailed);
+		else
+			view_infos = obi_view_formatted_infos_one_line(view);
+		if (view_infos == NULL)
+		{
+			obidebug(1, "\nError getting a view infos to get DMS formatted infos");
+			return NULL;
+		}
+		dms_infos = realloc(dms_infos, (strlen(dms_infos)+strlen(view_infos)+1) * sizeof(char));
+		if (dms_infos == NULL)
+		{
+			obidebug(1, "\nError reallocating memory for DMS formatted infos");
+			return NULL;
+		}
+		strcat(dms_infos, view_infos);
+		if (obi_save_and_close_view(view) < 0)
+		{
+			obidebug(1, "\nError closing view while getting DMS formatted infos");
+			return NULL;
+		}
+		if (detailed)
+		{
+			dms_infos = realloc(dms_infos, (strlen(dms_infos)+2) * sizeof(char));
+			strcat(dms_infos, "\n");
+		}
+	}
+
+	// Add taxonomies
+	dms_infos = realloc(dms_infos, (strlen(dms_infos)+strlen("\n# Taxonomies:\n")+1) * sizeof(char));
+	if (dms_infos == NULL)
+	{
+		obidebug(1, "\nError reallocating memory for DMS formatted infos");
+		return NULL;
+	}
+	strcat(dms_infos, "# Taxonomies:\n");
+	rewinddir(dms->tax_directory);
+	while ((dp = readdir(dms->tax_directory)) != NULL)
+	{
+		if ((dp->d_name)[0] == '.')
+			continue;
+		tax_name = dp->d_name;
+		dms_infos = realloc(dms_infos, (strlen(dms_infos)+strlen("  # ")+strlen(view_infos)+1) * sizeof(char));
+		if (dms_infos == NULL)
+		{
+			obidebug(1, "\nError reallocating memory for DMS formatted infos");
+			return NULL;
+		}
+		strcat(dms_infos, "  # ");
+		strcat(dms_infos, tax_name);
+	}
+	return dms_infos;
+}
+
+
 // TODO move somewhere else maybe
 // TODO discuss arguments
 obiversion_t obi_import_column(const char* dms_path_1, const char* dms_path_2, const char* column_name, obiversion_t version_number)

src/obidms.h

@@ -459,6 +459,23 @@ char* obi_dms_get_full_path(OBIDMS_p dms, const char* path_name);
 DIR* opendir_in_dms(OBIDMS_p dms, const char* path_name);
 
 
+/**
+ * @brief Returns the informations of a DMS with a human readable format (dms name, taxonomies and view infos).
+ *
+ * @warning The returned pointer has to be freed by the caller.
+ *
+ * @param column A pointer on a DMS.
+ * @param detailed Whether the informations should contain detailed view infos.
+ *
+ * @returns A pointer on a character array where the formatted DMS informations are stored.
+ * @retval NULL if an error occurred.
+ *
+ * @since September 2020
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+char* obi_dms_formatted_infos(OBIDMS_p dms, bool detailed);
+
+
 /**
  * @brief Imports a column, copying it from a DMS to another DMS, and returns the version of the column in the destination DMS.
  *

src/obidms_taxonomy.c

@@ -3649,6 +3649,18 @@ ecotx_t* obi_taxo_get_taxon_with_taxid(OBIDMS_taxonomy_p taxonomy, int32_t taxid
 }
 
 
+char* obi_taxo_get_name_from_name_idx(OBIDMS_taxonomy_p taxonomy, int32_t idx)
+{
+	return (((taxonomy->names)->names)[idx]).name;
+}
+
+
+ecotx_t* obi_taxo_get_taxon_from_name_idx(OBIDMS_taxonomy_p taxonomy, int32_t idx)
+{
+	return (((taxonomy->names)->names)[idx]).taxon;
+}
+
+
 int obi_taxo_is_taxon_under_taxid(ecotx_t* taxon, int32_t other_taxid)		// TODO discuss that this doesn't work with deprecated taxids
 {
 	ecotx_t* next_parent;

src/obidms_taxonomy.h

@@ -447,8 +447,51 @@ ecotx_t* obi_taxo_get_superkingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
 const char* obi_taxo_rank_index_to_label(int32_t rank_idx, ecorankidx_t* ranks);
 
 
-// TODO
+/**
+ * @brief Function checking whether a taxid is included in a subset of the taxonomy.
+ *
+ * @param taxonomy A pointer on the taxonomy structure.
+ * @param restrict_to_taxids An array of taxids. The researched taxid must be under at least one of those array taxids.
+ * @param count Number of taxids in restrict_to_taxids.
+ * @param taxid The taxid to check.
+ *
+ * @returns A value indicating whether the taxid is included in the chosen subset of the taxonomy.
+ * @retval 0 if the taxid is not included in the subset of the taxonomy.
+ * @retval 1 if the taxid is included in the subset of the taxonomy.
+ *
+ * @since October 2020
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
 int obi_taxo_is_taxid_included(OBIDMS_taxonomy_p taxonomy,
 			     			   int32_t* restrict_to_taxids,
 				    		   int32_t count,
 					    	   int32_t taxid);
+
+
+/**
+ * @brief Function returning the name of a taxon from its index in the taxonomy name index (econameidx_t).
+ *
+ * @param taxonomy A pointer on the taxonomy structure.
+ * @param idx The index at which the name is in the taxonomy name index (econameidx_t).
+ *
+ * @returns The taxon name.
+ *
+ * @since October 2020
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+char* obi_taxo_get_name_from_name_idx(OBIDMS_taxonomy_p taxonomy, int32_t idx);
+
+
+/**
+ * @brief Function returning a taxon structure from its index in the taxonomy name index (econameidx_t).
+ *
+ * @param taxonomy A pointer on the taxonomy structure.
+ * @param idx The index at which the taxon is in the taxonomy name index (econameidx_t).
+ *
+ * @returns The taxon structure.
+ *
+ * @since October 2020
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+ecotx_t* obi_taxo_get_taxon_from_name_idx(OBIDMS_taxonomy_p taxonomy, int32_t idx);
+

src/obidmscolumn.c

@@ -2435,17 +2435,77 @@ char* obi_column_formatted_infos(OBIDMS_column_p column, bool detailed)
 {
 	char* column_infos = NULL;
 	char* elt_names = NULL;
-	char* column_name = NULL;
+	char* data_type_str = NULL;
-	 // should be in view.c because alias exists in the context of view
+	char* comments = NULL;
-	column_infos = malloc(2048 * sizeof(char)); // TODO
 
+	// Get element names informations
 	elt_names = obi_get_formatted_elements_names(column);
+	if (elt_names == NULL)
+	{
+		obidebug(1, "\nError getting formatted elements names for formatted columns infos");
+		return NULL;
+	}
 
+	// Get data type informations
+	data_type_str = name_data_type((column->header)->returned_data_type);
+	if (data_type_str == NULL)
+	{
+		obidebug(1, "\nError getting formatted data type for formatted columns infos");
+		return NULL;
+	}
 
-//  "column_name, data type: OBI_TYPE, element names: [formatted element names](, all comments)"
+	// Get commments if detailed informations required
+	if (detailed)
+		comments = (column->header)->comments;
 
+	// Build the string of formatted infos, allocating memory as needed
+
+	// Data type
+	column_infos = (char*) malloc((strlen("data type: ")+strlen(data_type_str)+1) * sizeof(char));
+	if (column_infos == NULL)
+	{
+		obi_set_errno(OBI_MALLOC_ERROR);
+		obidebug(1, "\nError allocating memory for formatted column infos");
+		return NULL;
+	}
+
+	strcpy(column_infos, "data type: ");
+	strcat(column_infos, data_type_str);
+
+	// Element names if more than 1
+	if ((column->header)->nb_elements_per_line > 1)
+	{
+		column_infos = realloc(column_infos, (strlen(column_infos)+strlen(", elements: ")+strlen(elt_names)+1) * sizeof(char));
+		if (column_infos == NULL)
+		{
+			obi_set_errno(OBI_MALLOC_ERROR);
+			obidebug(1, "\nError allocating memory for formatted column infos");
+			return NULL;
+		}
+
+		strcat(column_infos, ", elements: ");
+		strcat(column_infos, elt_names);
+	}
+
+	if (detailed && (strlen(comments)>2)) // Add all comments if required and not empty
+	{
+		column_infos = realloc(column_infos, (strlen(column_infos)+strlen("\nComments:\n")+strlen(comments)+1) * sizeof(char));
+		if (column_infos == NULL)
+		{
+			obi_set_errno(OBI_MALLOC_ERROR);
+			obidebug(1, "\nError allocating memory for formatted column infos");
+			return NULL;
+		}
+
+		strcat(column_infos, "\nComments:\n");
+		strcat(column_infos, comments);
+	}
+
+//  "data type: OBI_TYPE, element names: [formatted element names](, all comments)"
 
 	free(elt_names);
+	free(data_type_str);
+
 	return column_infos;
 }
 
@@ -2492,7 +2552,6 @@ int obi_column_prepare_to_set_value(OBIDMS_column_p column, index_t line_nb, ind
 }
 
 
-
 int obi_column_prepare_to_get_value(OBIDMS_column_p column, index_t line_nb)
 {
 	if ((line_nb+1) > ((column->header)->line_count))

src/obidmscolumn.h

@@ -505,12 +505,37 @@ index_t obi_column_get_element_index_from_name(OBIDMS_column_p column, const cha
 char* obi_get_elements_names(OBIDMS_column_p column);
 
 
-// TODO
+/**
-//char* obi_get_formatted_elements_names(OBIDMS_column_p column);
+ * @brief Recovers the elements names of the lines of a column with a human readable format ("0; 1; 2; ...; n\0").
+ *
+ * @warning The returned pointer has to be freed by the caller.
+ *
+ * @param column A pointer on an OBIDMS column.
+ *
+ * @returns A pointer on a character array where the elements names are stored.
+ * @retval NULL if an error occurred.
+ *
+ * @since September 2020
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+char* obi_get_formatted_elements_names(OBIDMS_column_p column);
 
 
-// TODO
+/**
-//char* obi_column_formatted_infos(OBIDMS_column_p column);
+ * @brief Returns the informations of a column with a human readable format (data type, element names, comments).
+ *
+ * @warning The returned pointer has to be freed by the caller.
+ *
+ * @param column A pointer on an OBIDMS column.
+ * @param detailed Whether the informations should contain column comments or just data type and element names.
+ *
+ * @returns A pointer on a character array where the formatted column informations are stored.
+ * @retval NULL if an error occurred.
+ *
+ * @since September 2020
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+char* obi_column_formatted_infos(OBIDMS_column_p column, bool detailed);
 
 
 /**

src/obiview.c

@@ -17,6 +17,7 @@
 #include <sys/mman.h>
 #include <inttypes.h>
 #include <math.h>
+#include <time.h>
 //#include <ctype.h>
 
 #include "obiview.h"
@@ -1185,6 +1186,7 @@ static int close_view(Obiview_p view)
 			obidebug(1, "\nError getting a column to close from the linked list of column pointers of a view");
 			return -1;
 		}
+
 		if (obi_close_column(column) < 0)
 		{
 			obidebug(1, "\nError closing a column while closing a view");
@@ -2603,6 +2605,144 @@ int obi_view_create_column_alias(Obiview_p view, const char* current_name, const
 }
 
 
+char* obi_view_formatted_infos(Obiview_p view, bool detailed)
+{
+	int             i;
+	char*   		view_infos = NULL;
+	char*   		view_name = NULL;
+	time_t  		creation_date;
+	char*			creation_date_str = NULL;
+	index_t 		line_count;
+	char    		line_count_str[256];
+	OBIDMS_column_p column;
+	char*			column_alias = NULL;
+	char*   		column_infos = NULL;
+	char*			comments = NULL;
+
+	// View name
+	view_name = (view->infos)->name;
+	view_infos = (char*) malloc((strlen("# View name:\n")+strlen(view_name)+1) * sizeof(char));
+	strcpy(view_infos, "# View name:\n");
+	strcat(view_infos, view_name);
+
+	// Date created
+	if (view->read_only) // Date not saved until view is finished writing
+	{
+		creation_date = (view->infos)->creation_date;
+		creation_date_str = ctime(&creation_date);
+		view_infos = realloc(view_infos, (strlen(view_infos)+strlen("\n# Date created:\n")+strlen(creation_date_str)+1) * sizeof(char));
+		strcat(view_infos, "\n# Date created:\n");
+		strcat(view_infos, creation_date_str);
+	}
+
+	// Line count
+	line_count = (view->infos)->line_count;
+	snprintf(line_count_str, sizeof line_count_str, "%lld", line_count);
+	view_infos = realloc(view_infos, (strlen(view_infos)+strlen("\n# Line count:\n")+strlen(line_count_str)+1) * sizeof(char));
+	strcat(view_infos, "# Line count:\n");
+	strcat(view_infos, line_count_str);
+
+	// Columns: go through each, print their alias then their infos
+	view_infos = realloc(view_infos, (strlen(view_infos)+strlen("\n# Columns:")+1) * sizeof(char));
+	strcat(view_infos, "\n# Columns:");
+	for (i=0; i<((view->infos)->column_count); i++)
+	{
+		column = *((OBIDMS_column_p*)ll_get(view->columns, i));
+		if (column == NULL)
+		{
+			obidebug(1, "\nError getting a column from the linked list of column pointers of a view to format view infos");
+			return NULL;
+		}
+
+		// Column alias
+		column_alias = (((view->infos)->column_references)[i]).alias;
+		view_infos = realloc(view_infos, (strlen(view_infos)+strlen("\n")+strlen(column_alias)+strlen(", ")+1) * sizeof(char));
+		strcat(view_infos, "\n");
+		strcat(view_infos, column_alias);
+		strcat(view_infos, ", ");
+
+		// Column infos
+		column_infos = obi_column_formatted_infos(column, detailed);
+		if (column_infos == NULL)
+		{
+			obidebug(1, "\nError getting column infos to format view infos");
+			return NULL;
+		}
+
+		view_infos = realloc(view_infos, (strlen(view_infos)+strlen(column_infos)+1) * sizeof(char));
+		strcat(view_infos, column_infos);
+		free(column_infos);
+	}
+
+	// Get commments if detailed informations required
+	if (detailed)
+	{
+		comments = (view->infos)->comments;
+		if (strlen(comments)>2) // Add all comments if not empty
+		{
+			view_infos = realloc(view_infos, (strlen(view_infos)+strlen("\n# Comments:\n")+strlen(comments)+1) * sizeof(char));
+			if (view_infos == NULL)
+			{
+				obi_set_errno(OBI_MALLOC_ERROR);
+				obidebug(1, "\nError allocating memory for formatted view infos");
+				return NULL;
+			}
+
+			strcat(view_infos, "\n# Comments:\n");
+			strcat(view_infos, comments);
+		}
+	}
+
+	view_infos = realloc(view_infos, (strlen(view_infos)+2) * sizeof(char));
+	strcat(view_infos, "\n");
+
+	return view_infos;
+}
+
+
+char* obi_view_formatted_infos_one_line(Obiview_p view)
+{
+	int             i;
+	char*   		view_infos = NULL;
+	char*   		view_name = NULL;
+	time_t  		creation_date;
+	char*			creation_date_str = NULL;
+	index_t 		line_count;
+	char    		line_count_str[256];
+
+	// View name
+	view_name = (view->infos)->name;
+	view_infos = (char*) malloc((strlen("  # ")+strlen(view_name)+2) * sizeof(char));
+	strcpy(view_infos, "  # ");
+	strcat(view_infos, view_name);
+	strcat(view_infos, ":");
+
+	// Date created
+	if (view->read_only) // Date not saved until view is finished writing
+	{
+		creation_date = (view->infos)->creation_date;
+		creation_date_str = ctime(&creation_date);
+		// Delete \n added by ctime
+		creation_date_str[strlen(creation_date_str)-1] = '\0';
+		view_infos = realloc(view_infos, (strlen(view_infos)+strlen(" Date created: ")+strlen(creation_date_str)+1) * sizeof(char));
+		strcat(view_infos, " Date created: ");
+		strcat(view_infos, creation_date_str);
+	}
+
+	// Line count
+	line_count = (view->infos)->line_count;
+	snprintf(line_count_str, sizeof line_count_str, "%lld", line_count);
+	view_infos = realloc(view_infos, (strlen(view_infos)+strlen(" ; Line count: ")+strlen(line_count_str)+1) * sizeof(char));
+	strcat(view_infos, " ; Line count: ");
+	strcat(view_infos, line_count_str);
+
+	view_infos = realloc(view_infos, (strlen(view_infos)+2) * sizeof(char));
+	strcat(view_infos, "\n");
+
+	return view_infos;
+}
+
+
 int obi_view_write_comments(Obiview_p view, const char* comments)
 {
 	size_t new_size;

src/obiview.h

@@ -30,54 +30,56 @@
 #include "obiblob.h"
 
 
-#define OBIVIEW_NAME_MAX_LENGTH (249)   		/**< The maximum length of an OBIDMS view name, without the extension.
+#define OBIVIEW_NAME_MAX_LENGTH (249)   		   /**< The maximum length of an OBIDMS view name, without the extension.
-                                	 	  	  	 */
+                                	 	  	  	    */
-#define VIEW_TYPE_MAX_LENGTH (1024)   			/**< The maximum length of the type name of a view.
+#define VIEW_TYPE_MAX_LENGTH (1024)   			   /**< The maximum length of the type name of a view.
-                                	 	  	  	 */
+                                	 	  	  	    */
-#define LINES_COLUMN_NAME "LINES"				/**< The name of the column containing the line selections
+#define LINES_COLUMN_NAME "LINES"				   /**< The name of the column containing the line selections
- 	 	 	 	 	 	 	 	 	 	 	 	 *   in all views.
+ 	 	 	 	 	 	 	 	 	 	 	 	    *   in all views.
-                                	 	  	  	 */
+                                	 	  	  	    */
-#define VIEW_TYPE_NUC_SEQS "NUC_SEQS_VIEW"   	/**< The type name of views based on nucleotide sequences
+#define VIEW_TYPE_NUC_SEQS "NUC_SEQS_VIEW"   	   /**< The type name of views based on nucleotide sequences
-												 *   and their metadata.
+												    *   and their metadata.
-                                	 	  	  	 */
+                                	 	  	  	    */
-#define NUC_SEQUENCE_COLUMN "NUC_SEQ"			/**< The name of the column containing the nucleotide sequences
+#define NUC_SEQUENCE_COLUMN "NUC_SEQ"			   /**< The name of the column containing the nucleotide sequences
- 	 	 	 	 	 	 	 	 	 	 	 	 *   in NUC_SEQS_VIEW views.
+ 	 	 	 	 	 	 	 	 	 	 	 	    *   in NUC_SEQS_VIEW views.
-                                	 	  	  	 */
+                                	 	  	  	    */
-#define ID_COLUMN "ID"							/**< The name of the column containing the sequence identifiers
+#define ID_COLUMN "ID"							   /**< The name of the column containing the sequence identifiers
- 	 	 	 	 	 	 	 	 	 	 	 	 *   in NUC_SEQS_VIEW views.
+ 	 	 	 	 	 	 	 	 	 	 	 	    *   in NUC_SEQS_VIEW views.
-                                	 	  	  	 */
+                                	 	  	  	    */
-#define DEFINITION_COLUMN "DEFINITION"			/**< The name of the column containing the sequence definitions
+#define DEFINITION_COLUMN "DEFINITION"			   /**< The name of the column containing the sequence definitions
- 	 	 	 	 	 	 	 	 	 	 	 	 *   in NUC_SEQS_VIEW views.
+ 	 	 	 	 	 	 	 	 	 	 	 	    *   in NUC_SEQS_VIEW views.
-                                	 	  	  	 */
+                                	 	  	  	    */
-#define QUALITY_COLUMN "QUALITY"				/**< The name of the column containing the sequence qualities
+#define QUALITY_COLUMN "QUALITY"				   /**< The name of the column containing the sequence qualities
- 	 	 	 	 	 	 	 	 	 	 	 	 *   in NUC_SEQS_VIEW views.
+ 	 	 	 	 	 	 	 	 	 	 	 	    *   in NUC_SEQS_VIEW views.
-                                	 	  	   	 */
+                                	 	  	   	    */
-#define REVERSE_QUALITY_COLUMN "REVERSE_QUALITY" /**< The name of the column containing the sequence qualities
+#define REVERSE_QUALITY_COLUMN "REVERSE_QUALITY"   /**< The name of the column containing the sequence qualities
- 	 	 	 	 	 	 	 	 	 	 	 	 *    of the reverse read (generated by ngsfilter, used by alignpairedend).
+ 	 	 	 	 	 	 	 	 	 	 	 	    *    of the reverse read (generated by ngsfilter, used by alignpairedend).
-                                	 	  	   	 */
+                                	 	  	   	    */
 #define REVERSE_SEQUENCE_COLUMN "REVERSE_SEQUENCE" /**< The name of the column containing the sequence
- 	 	 	 	 	 	 	 	 	 	 	 	 *    of the reverse read (generated by ngsfilter, used by alignpairedend).
+ 	 	 	 	 	 	 	 	 	 	 	 	    *    of the reverse read (generated by ngsfilter, used by alignpairedend).
-                                	 	  	   	 */
+                                	 	  	   	    */
-#define QUALITY_COLUMN "QUALITY"				/**< The name of the column containing the sequence qualities
+#define QUALITY_COLUMN "QUALITY"				   /**< The name of the column containing the sequence qualities
- 	 	 	 	 	 	 	 	 	 	 	 	 *   in NUC_SEQS_VIEW views.
+ 	 	 	 	 	 	 	 	 	 	 	 	    *   in NUC_SEQS_VIEW views.
-                                	 	  	   	 */
+                                	 	  	   	    */
-#define COUNT_COLUMN "COUNT"				    /**< The name of the column containing the sequence counts
+#define COUNT_COLUMN "COUNT"				       /**< The name of the column containing the sequence counts
- 	 	 	 	 	 	 	 	 	 	 	 	 *   in NUC_SEQS_VIEW views.
+ 	 	 	 	 	 	 	 	 	 	 	 	    *   in NUC_SEQS_VIEW views.
-                                	 	  	  	 */
+                                	 	  	  	    */
-#define TAXID_COLUMN "TAXID"				    /**< The name of the column containing the taxids.       TODO subtype of INT column?
+#define SCIENTIFIC_NAME_COLUMN "SCIENTIFIC_NAME"   /**< The name of the column containing the taxon scientific name.
-                                	             */
+                                	                */
-#define MERGED_TAXID_COLUMN "MERGED_TAXID"		/**< The name of the column containing the merged taxids information.
+#define TAXID_COLUMN "TAXID"				       /**< The name of the column containing the taxids.       TODO subtype of INT column?
-                                	             */
+                                	                */
-#define MERGED_PREFIX "MERGED_"		            /**< The prefix to prepend to column names when merging informations during obi uniq.
+#define MERGED_TAXID_COLUMN "MERGED_TAXID"		   /**< The name of the column containing the merged taxids information.
-                                	             */
+                                	                */
-#define TAXID_DIST_COLUMN "TAXID_DIST"			/**< The name of the column containing a dictionary of taxid:[list of ids] when merging informations during obi uniq.
+#define MERGED_PREFIX "MERGED_"		               /**< The prefix to prepend to column names when merging informations during obi uniq.
-                                	             */
+                                	                */
-#define MERGED_COLUMN "MERGED"					/**< The name of the column containing a list of ids when merging informations during obi uniq.
+#define TAXID_DIST_COLUMN "TAXID_DIST"			   /**< The name of the column containing a dictionary of taxid:[list of ids] when merging informations during obi uniq.
-                                	             */
+                                	                */
-#define ID_PREFIX "seq"						    /**< The default prefix of sequence identifiers in automatic ID columns.
+#define MERGED_COLUMN "MERGED"					   /**< The name of the column containing a list of ids when merging informations during obi uniq.
-                                	 	  	  	 */
+                                	                */
-#define PREDICATE_KEY "predicates"		        /**< The key used in the json-formatted view comments to store predicates.
+#define ID_PREFIX "seq"						       /**< The default prefix of sequence identifiers in automatic ID columns.
-                                	 	  	  	 */
+                                	 	  	  	    */
+#define PREDICATE_KEY "predicates"		           /**< The key used in the json-formatted view comments to store predicates.
+                                	 	  	  	    */
 
 
 /**
@@ -519,6 +521,39 @@ OBIDMS_column_p* obi_view_get_pointer_on_column_in_view(Obiview_p view, const ch
 int obi_view_create_column_alias(Obiview_p view, const char* current_name, const char* alias);
 
 
+/**
+ * @brief Returns the informations of a view with a human readable format (view name, date created, line count, column informations, comments).
+ *
+ * @warning The returned pointer has to be freed by the caller.
+ *
+ * @param column A pointer on a view.
+ * @param detailed Whether the informations should contain view comments.
+ *
+ * @returns A pointer on a character array where the formatted view informations are stored.
+ * @retval NULL if an error occurred.
+ *
+ * @since September 2020
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+char* obi_view_formatted_infos(Obiview_p view, bool detailed);
+
+
+/**
+ * @brief Returns the informations of a view with a human readable format on one line (view name, date created, line count).
+ *
+ * @warning The returned pointer has to be freed by the caller.
+ *
+ * @param column A pointer on a view.
+ *
+ * @returns A pointer on a character array where the formatted view informations are stored.
+ * @retval NULL if an error occurred.
+ *
+ * @since September 2020
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+char* obi_view_formatted_infos_one_line(Obiview_p view);
+
+
 /**
  * @brief Internal function writing new comments in a view file.
  *