Switch to version 3.0.1b2

size (#68)

python/obitools3/commands/alignpairedend.pyx

@@ -205,19 +205,25 @@ def run(config):
     if type(entries) == list:
         forward = entries[0]
         reverse = entries[1]
-        aligner = Kmer_similarity(forward, \
-                                  view2=reverse, \
-                                  kmer_size=config['alignpairedend']['kmersize'], \
-                                  reversed_column=None)
+        if len(forward) == 0 or len(reverse) == 0:
+            aligner = None
+        else:
+            aligner = Kmer_similarity(forward, \
+                                      view2=reverse, \
+                                      kmer_size=config['alignpairedend']['kmersize'], \
+                                      reversed_column=None)
     else:
-        aligner = Kmer_similarity(entries, \
-                                  column2=entries[REVERSE_SEQUENCE_COLUMN], \
-                                  qual_column2=entries[REVERSE_QUALITY_COLUMN], \
-                                  kmer_size=config['alignpairedend']['kmersize'], \
-                                  reversed_column=entries[b'reversed'])  # column created by the ngsfilter tool
+        if len(entries) == 0:
+            aligner = None
+        else:
+            aligner = Kmer_similarity(entries, \
+                                      column2=entries[REVERSE_SEQUENCE_COLUMN], \
+                                      qual_column2=entries[REVERSE_QUALITY_COLUMN], \
+                                      kmer_size=config['alignpairedend']['kmersize'], \
+                                      reversed_column=entries[b'reversed'])  # column created by the ngsfilter tool
         
     ba = alignmentIterator(entries, aligner)
-
+    
     i = 0
     for ali in ba:
         
@@ -251,7 +257,7 @@ def run(config):
         pb(i, force=True)
         print("", file=sys.stderr)
 
-    if kmer_ali :
+    if kmer_ali and aligner is not None:
         aligner.free()
 
     # Save command config in View and DMS comments

python/obitools3/commands/cat.pyx

@@ -115,7 +115,7 @@ def run(config):
             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']))
+                              nb_elements_per_line=dict_cols[coln]['nbelts'], elements_names=list(dict_cols[coln]['eltnames']), dict_column=True)
     
     # Initialize the progress bar
     if not config['obi']['noprogressbar']:

python/obitools3/commands/import.pyx

@@ -257,7 +257,8 @@ def run(config):
         for tag in dict_dict:
             dcols[tag] = (Column.new_column(view, tag, dict_dict[tag][1], \
                               nb_elements_per_line=len(dict_dict[tag][0]), \
-                              elements_names=list(dict_dict[tag][0])), \
+                              elements_names=list(dict_dict[tag][0]), \
+                              dict_column=True), \
                           dict_dict[tag][1])
     
         
@@ -328,10 +329,13 @@ def run(config):
                         value_type = type(value)
                         nb_elts = 1
                         value_obitype = OBI_VOID
+                        dict_col = False
                          
                         if value_type == dict or value_type == list :
                             nb_elts = len(value)
                             elt_names = list(value)
+                            if value_type == dict :
+                                dict_col = True
                         else :
                             nb_elts = 1
                             elt_names = None
@@ -339,11 +343,9 @@ def run(config):
                         value_obitype = get_obitype(value)
                          
                         if value_obitype != OBI_VOID :
-                            dcols[tag] = (Column.new_column(view, tag, value_obitype, nb_elements_per_line=nb_elts, elements_names=elt_names), value_obitype)
+                            dcols[tag] = (Column.new_column(view, tag, value_obitype, nb_elements_per_line=nb_elts, elements_names=elt_names, dict_column=dict_col), value_obitype)
                                                      
                             # Fill value
-                            if value_type == dict and nb_elts == 1:  # special case that makes the OBI3 create a 1 elt/line column which won't read a dict value
-                                value = value[list(value.keys())[0]]       # The solution is to transform the value in a simple atomic one acceptable by the column
                             dcols[tag][0][i] = value
                          
                         # TODO else log error?

python/obitools3/commands/ngsfilter.pyx

@@ -645,6 +645,7 @@ def run(config):
     
     g = 0
     u = 0
+    i = 0
     no_tags = config['ngsfilter']['notags']
     try:
         for i in range(entries_len):

python/obitools3/commands/test.pyx

@@ -301,8 +301,11 @@ def fill_column(config, infos, col) :
 def create_random_column(config, infos) :
     alias = random.choice([b'', random_unique_name(infos)])
     tuples = random.choice([True, False])
+    dict_column = False
     if not tuples :
         nb_elements_per_line=random.randint(1, config['test']['maxelts'])
+        if nb_elements_per_line > 1:
+            dict_column = True
         elements_names = []
         for i in range(nb_elements_per_line) :
             elements_names.append(random_unique_element_name(config, infos))
@@ -318,6 +321,7 @@ def create_random_column(config, infos) :
                                data_type, 
                                nb_elements_per_line=nb_elements_per_line,
                                elements_names=elements_names,
+                               dict_column=dict_column,
                                tuples=tuples,
                                comments=random_comments(config),
                                alias=alias
@@ -442,7 +446,7 @@ def addOptions(parser):
                        default=20,
                        type=int,
                        help="Maximum length of tuples. "
-                            "Default: 50")
+                            "Default: 20")
   
     group.add_argument('--max_ini_col_count','-o',
                        action="store", dest="test:maxinicolcount",
@@ -455,7 +459,7 @@ def addOptions(parser):
     group.add_argument('--max_line_nb','-l',
                        action="store", dest="test:maxlinenb",
                        metavar='<MAX_LINE_NB>',
-                       default=10000,
+                       default=1000,
                        type=int,
                        help="Maximum number of lines in a column. "
                             "Default: 1000")

python/obitools3/commands/uniq.pyx

@@ -378,6 +378,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, di
                               OBI_INT,
                               nb_elements_per_line=merged_infos[merged_col_name]['nb_elts'],
                               elements_names=list(merged_infos[merged_col_name]['elt_names']),
+                              dict_column=True,
                               comments=i_col.comments,
                               alias=merged_col_name
                              )
@@ -400,6 +401,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, di
                               OBI_INT,
                               nb_elements_per_line=len(view),
                               elements_names=[id for id in i_id_col],
+                              dict_column=True,
                               alias=TAXID_DIST_COLUMN
                              )
 

python/obitools3/dms/capi/obidmscolumn.pxd

@@ -31,6 +31,7 @@ cdef extern from "obidmscolumn.h" nogil:
         const_char_p        elements_names
         OBIType_t           returned_data_type
         OBIType_t           stored_data_type
+        bint                dict_column
         bint                tuples
         bint                to_eval
         time_t              creation_date

python/obitools3/dms/capi/obiview.pxd

@@ -95,6 +95,7 @@ cdef extern from "obiview.h" nogil:
                             index_t nb_elements_per_line,
                             char* elements_names,
                             bint  elt_names_formatted,
+                            bint dict_column,
                             bint tuples,
                             bint to_eval,
                             const_char_p indexer_name,

python/obitools3/dms/column/column.pyx

@@ -90,6 +90,7 @@ cdef class Column(OBIWrapper) :
                    obitype_t data_type,
                    index_t   nb_elements_per_line=1,
                    list      elements_names=None,
+                   bint      dict_column=False,
                    bint      tuples=False,
                    bint      to_eval=False,
                    object    associated_column_name=b"",
@@ -152,6 +153,7 @@ cdef class Column(OBIWrapper) :
                                 nb_elements_per_line      = nb_elements_per_line,
                                 elements_names            = elements_names_p,
                                 elt_names_formatted       = False,
+                                dict_column               = dict_column,
                                 tuples                    = tuples,
                                 to_eval                   = to_eval,
                                 indexer_name              = NULL,
@@ -200,7 +202,7 @@ cdef class Column(OBIWrapper) :
         
         column_p = column_pp[0]
         column_type = column_p.header.returned_data_type
-        column_class = Column.get_column_class(column_type, (column_p.header.nb_elements_per_line > 1), column_p.header.tuples)
+        column_class = Column.get_column_class(column_type, (column_p.header.nb_elements_per_line > 1 or column_p.header.dict_column == True), column_p.header.tuples)
         column = OBIWrapper.new_wrapper(column_class, column_pp)
         
         column._view = view
@@ -236,6 +238,7 @@ cdef class Column(OBIWrapper) :
                                 nb_elements_per_line      = -1,
                                 elements_names            = NULL,
                                 elt_names_formatted       = False,
+                                dict_column               = False,
                                 tuples                    = False,
                                 to_eval                   = False,
                                 indexer_name              = NULL,
@@ -374,6 +377,13 @@ cdef class Column(OBIWrapper) :
             raise OBIDeactivatedInstanceError()
         return self.pointer().header.nb_elements_per_line
  
+    # dict_column property getter
+    @property
+    def dict_column(self):
+        if not self.active() :
+            raise OBIDeactivatedInstanceError()
+        return self.pointer().header.dict_column
+ 
     # data_type property getter
     @property
     def data_type(self):

python/obitools3/dms/column/typed_column/bool.pyx

@@ -38,11 +38,13 @@ cdef class Column_bool(Column):
             object  column_name,
             index_t nb_elements_per_line=1,
             object  elements_names=None,
+            bint    dict_column=False,
             bint    tuples=False,
             object  comments={}):
         return Column.new_column(view, column_name, OBI_BOOL,
                           nb_elements_per_line=nb_elements_per_line, 
-                          elements_names=elements_names, 
+                          elements_names=elements_names,
+                          dict_column=dict_column, 
                           tuples=tuples,
                           comments=comments)
                 

python/obitools3/dms/column/typed_column/char.pyx

@@ -36,12 +36,14 @@ cdef class Column_char(Column):
             object  column_name,
             index_t nb_elements_per_line=1,
             object  elements_names=None,
+            bint    dict_column=False,
             bint    tuples=False,
             object  comments={}):
         
         return Column.new_column(view, column_name, OBI_CHAR,
                                  nb_elements_per_line=nb_elements_per_line, 
                                  elements_names=elements_names, 
+                                 dict_column=dict_column, 
                                  tuples=tuples,
                                  comments=comments)
 

python/obitools3/dms/column/typed_column/float.pyx

@@ -36,12 +36,14 @@ cdef class Column_float(Column):
             object  column_name,
             index_t nb_elements_per_line=1,
             object  elements_names=None,
+            bint    dict_column=False,
             bint    tuples=False,
             object  comments={}):
         
         return Column.new_column(view, column_name, OBI_FLOAT,
                                  nb_elements_per_line=nb_elements_per_line, 
                                  elements_names=elements_names, 
+                                 dict_column=dict_column, 
                                  tuples=tuples,
                                  comments=comments)
 

python/obitools3/dms/column/typed_column/int.pyx

@@ -38,12 +38,14 @@ cdef class Column_int(Column):
             object  column_name,
             index_t nb_elements_per_line=1,
             object  elements_names=None,
+            bint    dict_column=False,
             bint    tuples=False,
             object  comments={}):
         
         return Column.new_column(view, column_name, OBI_INT,
                                  nb_elements_per_line=nb_elements_per_line, 
                                  elements_names=elements_names, 
+                                 dict_column=dict_column, 
                                  tuples=tuples,
                                  comments=comments)
 

python/obitools3/dms/column/typed_column/qual.pyx

@@ -38,6 +38,7 @@ cdef class Column_qual(Column_idx):
             object  column_name,
             index_t nb_elements_per_line=1,
             object  elements_names=None,
+            bint    dict_column=False,
             object  associated_column_name=b"",
             int     associated_column_version=-1,
             object  comments={}):
@@ -45,6 +46,7 @@ cdef class Column_qual(Column_idx):
         return Column.new_column(view, column_name, OBI_QUAL,
                                  nb_elements_per_line=nb_elements_per_line, 
                                  elements_names=elements_names, 
+                                 dict_column=dict_column, 
                                  tuples=False,
                                  associated_column_name=associated_column_name,
                                  associated_column_version=associated_column_name,

python/obitools3/dms/column/typed_column/seq.pyx

@@ -39,12 +39,14 @@ cdef class Column_seq(Column_idx):
             object  column_name,
             index_t nb_elements_per_line=1,
             object  elements_names=None,
+            bint    dict_column=False,
             bint    tuples=False,
             object  comments={}):
         
         return Column.new_column(view, column_name, OBI_SEQ,
                                  nb_elements_per_line=nb_elements_per_line, 
                                  elements_names=elements_names, 
+                                 dict_column=dict_column, 
                                  tuples=tuples,
                                  comments=comments)
 

python/obitools3/dms/column/typed_column/str.pyx

@@ -38,12 +38,14 @@ cdef class Column_str(Column_idx):
             object  column_name,
             index_t nb_elements_per_line=1,
             object  elements_names=None,
+            bint    dict_column=False,
             bint    tuples=False,
             object  comments={}):
         
         return Column.new_column(view, column_name, OBI_STR,
                                  nb_elements_per_line=nb_elements_per_line, 
                                  elements_names=elements_names, 
+                                 dict_column=dict_column, 
                                  tuples=tuples,
                                  comments=comments)
 

python/obitools3/dms/view/view.pyx

@@ -343,7 +343,7 @@ cdef class View(OBIWrapper) :
 
         new_column = Column.new_column(self, old_column.pointer().header.name, new_data_type, 
                                        nb_elements_per_line=new_nb_elements_per_line, elements_names=new_elements_names, 
-                                       comments=old_column.comments, alias=column_name_b+tobytes('___new___'))
+                                       dict_column=(new_nb_elements_per_line>1), comments=old_column.comments, alias=column_name_b+tobytes('___new___'))
         
         switch_to_dict = old_column.nb_elements_per_line == 1 and new_nb_elements_per_line > 1
         ori_key = old_column._elements_names[0]
@@ -404,6 +404,7 @@ cdef class View(OBIWrapper) :
                                   col.data_type_int, 
                                   nb_elements_per_line = col.nb_elements_per_line,
                                   elements_names = col._elements_names,
+                                  dict_column = col.dict_column,
                                   tuples = col.tuples,
                                   to_eval = col.to_eval,
                                   comments = col.comments,
@@ -805,7 +806,7 @@ cdef class Line :
 
 
     def __repr__(self):
-        return bytes2str(self).repr_bytes()
+        return bytes2str(self.repr_bytes())
 
 
     cpdef repr_bytes(self):

python/obitools3/parsers/genbank.pyx

@@ -22,10 +22,10 @@ from libc.stdlib  cimport free, malloc, realloc
 from libc.string cimport strcpy, strlen
 
 
-_featureMatcher = re.compile(b'^FEATURES.+\n(?=ORIGIN)',re.DOTALL + re.M)
+_featureMatcher = re.compile(b'^FEATURES.+\n(?=ORIGIN )',re.DOTALL + re.M)
 
 _headerMatcher = re.compile(b'^LOCUS.+(?=\nFEATURES)', re.DOTALL + re.M)
-_seqMatcher    = re.compile(b'ORIGIN.+(?=//\n)', re.DOTALL + re.M)
+_seqMatcher    = re.compile(b'ORIGIN .+(?=//\n)', re.DOTALL + re.M)
 _cleanSeq1     = re.compile(b'ORIGIN.+\n')
 _cleanSeq2     = re.compile(b'[ \n0-9]+')
 _acMatcher     = re.compile(b'(?<=^ACCESSION   ).+',re.M)

python/obitools3/parsers/universal.pyx

@@ -53,7 +53,11 @@ def entryIteratorFactory(lineiterator,
 
     i = iterator
         
-    first=next(i)    
+    try:
+        first=next(i)
+    except StopIteration:
+        first=""
+        pass
 
     format=b"tabular"
     

python/obitools3/version.py

@@ -1,5 +1,5 @@
 major = 3
 minor = 0
-serial= '0b40'
+serial= '1b2'
 
 version ="%d.%d.%s" % (major,minor,serial)

src/build_reference_db.c

@@ -243,6 +243,7 @@ int build_reference_db(const char* dms_name,
 							false,
 							false,
 							false,
+							false,
 							"",
 							"",
 							-1,
@@ -392,6 +393,7 @@ int build_reference_db(const char* dms_name,
 							1,
 							"",
 							false,
+							false,
 							true,
 							false,
 							"",
@@ -415,6 +417,7 @@ int build_reference_db(const char* dms_name,
 							1,
 							"",
 							false,
+							false,
 							true,
 							false,
 							"",

src/obi_clean.c

@@ -88,42 +88,42 @@ static int create_output_columns(Obiview_p o_view,
 								 int sample_count)
 {
 	// Status column
-	if (obi_view_add_column(o_view, CLEAN_STATUS_COLUMN_NAME, -1, NULL, OBI_CHAR, 0, sample_count, (sample_column->header)->elements_names, true, false, false, NULL, NULL, -1, CLEAN_STATUS_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, CLEAN_STATUS_COLUMN_NAME, -1, NULL, OBI_CHAR, 0, sample_count, (sample_column->header)->elements_names, true, true, false, false, NULL, NULL, -1, CLEAN_STATUS_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", CLEAN_STATUS_COLUMN_NAME);
 		return -1;
 	}
 
 	// Head column
-	if (obi_view_add_column(o_view, CLEAN_HEAD_COLUMN_NAME, -1, NULL, OBI_BOOL, 0, 1, NULL, false, false, false, NULL, NULL, -1, CLEAN_HEAD_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, CLEAN_HEAD_COLUMN_NAME, -1, NULL, OBI_BOOL, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, CLEAN_HEAD_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", CLEAN_HEAD_COLUMN_NAME);
 		return -1;
 	}
 
 	// Sample count column
-	if (obi_view_add_column(o_view, CLEAN_SAMPLECOUNT_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, CLEAN_SAMPLECOUNT_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, CLEAN_SAMPLECOUNT_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, CLEAN_SAMPLECOUNT_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", CLEAN_SAMPLECOUNT_COLUMN_NAME);
 		return -1;
 	}
 
 	// Head count column
-	if (obi_view_add_column(o_view, CLEAN_HEADCOUNT_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, CLEAN_HEADCOUNT_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, CLEAN_HEADCOUNT_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, CLEAN_HEADCOUNT_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", CLEAN_HEADCOUNT_COLUMN_NAME);
 		return -1;
 	}
 
 	// Internal count column
-	if (obi_view_add_column(o_view, CLEAN_INTERNALCOUNT_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, CLEAN_INTERNALCOUNT_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, CLEAN_INTERNALCOUNT_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, CLEAN_INTERNALCOUNT_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", CLEAN_INTERNALCOUNT_COLUMN_NAME);
 		return -1;
 	}
 
 	// Singleton count column
-	if (obi_view_add_column(o_view, CLEAN_SINGLETONCOUNT_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, CLEAN_SINGLETONCOUNT_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, CLEAN_SINGLETONCOUNT_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, CLEAN_SINGLETONCOUNT_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", CLEAN_SINGLETONCOUNT_COLUMN_NAME);
 		return -1;
@@ -229,6 +229,8 @@ int obi_clean(const char* dms_name,
 		return -1;
 	}
 
+	seq_count = (i_view->infos)->line_count;
+
 	// Open the sequence column
 	if (strcmp((i_view->infos)->view_type, VIEW_TYPE_NUC_SEQS) == 0)
 		iseq_column = obi_view_get_column(i_view, NUC_SEQUENCE_COLUMN);
@@ -245,7 +247,7 @@ int obi_clean(const char* dms_name,
 	}
 
 	// Open the sample column if there is one
-	if ((strcmp(sample_column_name, "") == 0) || (sample_column_name == NULL))
+	if ((strcmp(sample_column_name, "") == 0) || (sample_column_name == NULL) || (seq_count == 0))
 	{
 		fprintf(stderr, "Info: No sample information provided, assuming one sample.\n");
 		sample_column = obi_view_get_column(i_view, COUNT_COLUMN);
@@ -340,66 +342,67 @@ int obi_clean(const char* dms_name,
 		return -1;
 	}
 
-	// Build kmer tables
-	ktable = hash_seq_column(i_view, iseq_column, 0);
-	if (ktable == NULL)
+	if (seq_count > 0)
 	{
-		obi_set_errno(OBI_CLEAN_ERROR);
-		obidebug(1, "\nError building kmer tables before aligning");
-		return -1;
-	}
+		// Build kmer tables
+		ktable = hash_seq_column(i_view, iseq_column, 0);
+		if (ktable == NULL)
+		{
+			obi_set_errno(OBI_CLEAN_ERROR);
+			obidebug(1, "\nError building kmer tables before aligning");
+			return -1;
+		}
 
-	seq_count = (i_view->infos)->line_count;
-
-	// Allocate arrays for sample counts otherwise reading in mapped files takes longer
-	complete_sample_count_array = (int*) malloc(seq_count * sample_count * sizeof(int));
-	if (complete_sample_count_array == NULL)
-	{
-		obi_set_errno(OBI_MALLOC_ERROR);
-		obidebug(1, "\nError allocating memory for the array of sample counts, size: %lld", seq_count * sample_count * sizeof(int));
-		return -1;
-	}
-	for (samp=0; samp < sample_count; samp++)
-	{
-		for (k=0; k<seq_count; k++)
-			complete_sample_count_array[k+(samp*seq_count)] = obi_get_int_with_elt_idx_and_col_p_in_view(i_view, sample_column, k, samp);
-	}
-
-	// Allocate arrays for blobs otherwise reading in mapped files takes longer
-	blob_array = (Obi_blob_p*) malloc(seq_count * sizeof(Obi_blob_p));
-	if (blob_array == NULL)
-	{
-		obi_set_errno(OBI_MALLOC_ERROR);
-		obidebug(1, "\nError allocating memory for the array of blobs");
-		return -1;
-	}
-	for (k=0; k<seq_count; k++)
-	{
-		blob_array[k] = obi_get_blob_with_elt_idx_and_col_p_in_view(i_view, iseq_column, k, 0);
-	}
-
-	// Allocate alignment result array (byte at 0 if not aligned yet,
-	//											1 if sequence at index has a similarity above the threshold with the current sequence,
-	//											2 if sequence at index has a similarity below the threshold with the current sequence)
-	alignment_result_array = (byte_t*) calloc(seq_count, sizeof(byte_t));
-	if (alignment_result_array == NULL)
-	{
-		obi_set_errno(OBI_MALLOC_ERROR);
-		obidebug(1, "\nError allocating memory for alignment result array");
-		return -1;
-	}
-
-	// Initialize all sequences to singletons or NA if no sequences in that sample
-	for (k=0; k<seq_count; k++)
-	{
+		// Allocate arrays for sample counts otherwise reading in mapped files takes longer
+		complete_sample_count_array = (int*) malloc(seq_count * sample_count * sizeof(int));
+		if (complete_sample_count_array == NULL)
+		{
+			obi_set_errno(OBI_MALLOC_ERROR);
+			obidebug(1, "\nError allocating memory for the array of sample counts, size: %lld", seq_count * sample_count * sizeof(int));
+			return -1;
+		}
 		for (samp=0; samp < sample_count; samp++)
 		{
-			if (obi_get_int_with_elt_idx_and_col_p_in_view(i_view, sample_column, k, samp) != OBIInt_NA)  // Only initialize samples where there are some sequences
+			for (k=0; k<seq_count; k++)
+				complete_sample_count_array[k+(samp*seq_count)] = obi_get_int_with_elt_idx_and_col_p_in_view(i_view, sample_column, k, samp);
+		}
+
+		// Allocate arrays for blobs otherwise reading in mapped files takes longer
+		blob_array = (Obi_blob_p*) malloc(seq_count * sizeof(Obi_blob_p));
+		if (blob_array == NULL)
+		{
+			obi_set_errno(OBI_MALLOC_ERROR);
+			obidebug(1, "\nError allocating memory for the array of blobs");
+			return -1;
+		}
+		for (k=0; k<seq_count; k++)
+		{
+			blob_array[k] = obi_get_blob_with_elt_idx_and_col_p_in_view(i_view, iseq_column, k, 0);
+		}
+
+		// Allocate alignment result array (byte at 0 if not aligned yet,
+		//											1 if sequence at index has a similarity above the threshold with the current sequence,
+		//											2 if sequence at index has a similarity below the threshold with the current sequence)
+		alignment_result_array = (byte_t*) calloc(seq_count, sizeof(byte_t));
+		if (alignment_result_array == NULL)
+		{
+			obi_set_errno(OBI_MALLOC_ERROR);
+			obidebug(1, "\nError allocating memory for alignment result array");
+			return -1;
+		}
+
+		// Initialize all sequences to singletons or NA if no sequences in that sample
+		for (k=0; k<seq_count; k++)
+		{
+			for (samp=0; samp < sample_count; samp++)
 			{
-				if (obi_set_char_with_elt_idx_and_col_p_in_view(o_view, status_column, k, samp, 's') < 0)
+				if (obi_get_int_with_elt_idx_and_col_p_in_view(i_view, sample_column, k, samp) != OBIInt_NA)  // Only initialize samples where there are some sequences
 				{
-					obidebug(1, "\nError initializing all sequences to singletons");
-					return -1;
+					if (obi_set_char_with_elt_idx_and_col_p_in_view(o_view, status_column, k, samp, 's') < 0)
+					{
+						obidebug(1, "\nError initializing all sequences to singletons");
+						return -1;
+					}
 				}
 			}
 		}
@@ -551,17 +554,20 @@ int obi_clean(const char* dms_name,
 		}
 	}
 
-	free_kmer_tables(ktable, seq_count);
-	free(complete_sample_count_array);
-	free(blob_array);
-	free(alignment_result_array);
+	if (seq_count > 0)
+	{
+		free_kmer_tables(ktable, seq_count);
+		free(complete_sample_count_array);
+		free(blob_array);
+		free(alignment_result_array);
+	}
 
 	fprintf(stderr, "\n");
 
 	if (stop)
 		return -1;
 
-	if (heads_only)
+	if (heads_only && (seq_count > 0))
 	{
 		line_selection = malloc((((o_view->infos)->line_count) + 1) * sizeof(index_t));
 		if (line_selection == NULL)
@@ -635,7 +641,7 @@ int obi_clean(const char* dms_name,
 	}
 
 	// Flag the end of the line selection
-	if (heads_only)
+	if (heads_only && (seq_count > 0))
 		line_selection[l] = -1;
 
 	// Create new view with line selection if heads only

src/obi_ecopcr.c

@@ -150,49 +150,49 @@ static int print_seq(Obiview_p i_view, Obiview_p o_view,
 static int create_output_columns(Obiview_p o_view, bool kingdom_mode)
 {
 	// Original length column
-	if (obi_view_add_column(o_view, ECOPCR_SEQLEN_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_SEQLEN_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_SEQLEN_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_SEQLEN_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_SEQLEN_COLUMN_NAME);
 		return -1;
 	}
 
 	// Amplicon length column
-	if (obi_view_add_column(o_view, ECOPCR_AMPLICONLEN_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_AMPLICONLEN_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_AMPLICONLEN_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_AMPLICONLEN_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_AMPLICONLEN_COLUMN_NAME);
 		return -1;
 	}
 
 	// Taxid column
-	if (obi_view_add_column(o_view, TAXID_COLUMN, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, TAXID_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, TAXID_COLUMN, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, TAXID_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", TAXID_COLUMN);
 		return -1;
 	}
 
 	// Taxonomic rank column
-	if (obi_view_add_column(o_view, ECOPCR_RANK_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_RANK_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_RANK_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_RANK_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_RANK_COLUMN_NAME);
 		return -1;
 	}
 
 	// Species taxid column
-	if (obi_view_add_column(o_view, ECOPCR_SPECIES_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_SPECIES_TAXID_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_SPECIES_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_SPECIES_TAXID_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_SPECIES_TAXID_COLUMN_NAME);
 		return -1;
 	}
 
 	// Genus taxid column
-	if (obi_view_add_column(o_view, ECOPCR_GENUS_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_GENUS_TAXID_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_GENUS_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_GENUS_TAXID_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_GENUS_TAXID_COLUMN_NAME);
 		return -1;
 	}
 
 	// Family taxid column
-	if (obi_view_add_column(o_view, ECOPCR_FAMILY_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_FAMILY_TAXID_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_FAMILY_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_FAMILY_TAXID_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_FAMILY_TAXID_COLUMN_NAME);
 		return -1;
@@ -201,7 +201,7 @@ static int create_output_columns(Obiview_p o_view, bool kingdom_mode)
 	if (kingdom_mode)
 	{
 		// Kingdom taxid column
-		if (obi_view_add_column(o_view, ECOPCR_KINGDOM_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_KINGDOM_TAXID_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(o_view, ECOPCR_KINGDOM_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_KINGDOM_TAXID_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the %s column", ECOPCR_KINGDOM_TAXID_COLUMN_NAME);
 			return -1;
@@ -210,7 +210,7 @@ static int create_output_columns(Obiview_p o_view, bool kingdom_mode)
 	else
 	{
 		// Superkingdom taxid column
-		if (obi_view_add_column(o_view, ECOPCR_SUPERKINGDOM_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_SUPERKINGDOM_TAXID_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(o_view, ECOPCR_SUPERKINGDOM_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_SUPERKINGDOM_TAXID_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the %s column", ECOPCR_SUPERKINGDOM_TAXID_COLUMN_NAME);
 			return -1;
@@ -218,28 +218,28 @@ static int create_output_columns(Obiview_p o_view, bool kingdom_mode)
 	}
 
 	// Scientific name column
-	if (obi_view_add_column(o_view, ECOPCR_SCIENTIFIC_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_SCIENTIFIC_NAME_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_SCIENTIFIC_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_SCIENTIFIC_NAME_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_SCIENTIFIC_NAME_COLUMN_NAME);
 		return -1;
 	}
 
 	// Species name column
-	if (obi_view_add_column(o_view, ECOPCR_SPECIES_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_SPECIES_NAME_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_SPECIES_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_SPECIES_NAME_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_SPECIES_NAME_COLUMN_NAME);
 		return -1;
 	}
 
 	// Genus name column
-	if (obi_view_add_column(o_view, ECOPCR_GENUS_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_GENUS_NAME_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_GENUS_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_GENUS_NAME_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_GENUS_NAME_COLUMN_NAME);
 		return -1;
 	}
 
 	// Family name column
-	if (obi_view_add_column(o_view, ECOPCR_FAMILY_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_FAMILY_NAME_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_FAMILY_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_FAMILY_NAME_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_FAMILY_NAME_COLUMN_NAME);
 		return -1;
@@ -248,7 +248,7 @@ static int create_output_columns(Obiview_p o_view, bool kingdom_mode)
 	if (kingdom_mode)
 	{
 		// Kingdom name column
-		if (obi_view_add_column(o_view, ECOPCR_KINGDOM_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_KINGDOM_NAME_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(o_view, ECOPCR_KINGDOM_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_KINGDOM_NAME_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the %s column", ECOPCR_KINGDOM_NAME_COLUMN_NAME);
 			return -1;
@@ -257,7 +257,7 @@ static int create_output_columns(Obiview_p o_view, bool kingdom_mode)
 	else
 	{
 		// Superkingdom name column
-		if (obi_view_add_column(o_view, ECOPCR_SUPERKINGDOM_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_SUPERKINGDOM_NAME_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(o_view, ECOPCR_SUPERKINGDOM_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_SUPERKINGDOM_NAME_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the %s column", ECOPCR_SUPERKINGDOM_NAME_COLUMN_NAME);
 			return -1;
@@ -265,49 +265,49 @@ static int create_output_columns(Obiview_p o_view, bool kingdom_mode)
 	}
 
 	// Strand column
-	if (obi_view_add_column(o_view, ECOPCR_STRAND_COLUMN_NAME, -1, NULL, OBI_CHAR, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_STRAND_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_STRAND_COLUMN_NAME, -1, NULL, OBI_CHAR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_STRAND_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_STRAND_COLUMN_NAME);
 		return -1;
 	}
 
 	// Primer 1 column
-	if (obi_view_add_column(o_view, ECOPCR_PRIMER1_COLUMN_NAME, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_PRIMER1_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_PRIMER1_COLUMN_NAME, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_PRIMER1_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_PRIMER1_COLUMN_NAME);
 		return -1;
 	}
 
 	// Primer 2 column
-	if (obi_view_add_column(o_view, ECOPCR_PRIMER2_COLUMN_NAME, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_PRIMER2_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_PRIMER2_COLUMN_NAME, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_PRIMER2_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_PRIMER2_COLUMN_NAME);
 		return -1;
 	}
 
 	// Error 1 column
-	if (obi_view_add_column(o_view, ECOPCR_ERROR1_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_ERROR1_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_ERROR1_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_ERROR1_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_ERROR1_COLUMN_NAME);
 		return -1;
 	}
 
 	// Error 2 column
-	if (obi_view_add_column(o_view, ECOPCR_ERROR2_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_ERROR2_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_ERROR2_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_ERROR2_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_ERROR2_COLUMN_NAME);
 		return -1;
 	}
 
 	// Temperature 1 column
-	if (obi_view_add_column(o_view, ECOPCR_TEMP1_COLUMN_NAME, -1, NULL, OBI_FLOAT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_TEMP1_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_TEMP1_COLUMN_NAME, -1, NULL, OBI_FLOAT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_TEMP1_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_TEMP1_COLUMN_NAME);
 		return -1;
 	}
 
 	// Temperature 2 column
-	if (obi_view_add_column(o_view, ECOPCR_TEMP2_COLUMN_NAME, -1, NULL, OBI_FLOAT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ECOPCR_TEMP2_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(o_view, ECOPCR_TEMP2_COLUMN_NAME, -1, NULL, OBI_FLOAT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ECOPCR_TEMP2_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the %s column", ECOPCR_TEMP2_COLUMN_NAME);
 		return -1;

src/obi_ecotag.c

@@ -104,42 +104,42 @@ int print_assignment_result(Obiview_p output_view, index_t line,
 static int create_output_columns(Obiview_p o_view)
 {
 	// Score column
-	if (obi_view_add_column(o_view, ECOTAG_SCORE_COLUMN_NAME, -1, NULL, OBI_FLOAT, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)
+	if (obi_view_add_column(o_view, ECOTAG_SCORE_COLUMN_NAME, -1, NULL, OBI_FLOAT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)
 	{
 		obidebug(1, "\nError creating the column for the score in ecotag");
 		return -1;
 	}
 
 	// Assigned taxid column
-	if (obi_view_add_column(o_view, ECOTAG_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)
+	if (obi_view_add_column(o_view, ECOTAG_TAXID_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)
 	{
 		obidebug(1, "\nError creating the column for the assigned taxid in ecotag");
 		return -1;
 	}
 
 	// Assigned scientific name column
-	if (obi_view_add_column(o_view, ECOTAG_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)
+	if (obi_view_add_column(o_view, ECOTAG_NAME_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)
 	{
 		obidebug(1, "\nError creating the column for the assigned scientific name in ecotag");
 		return -1;
 	}
 
 	// Assignement status column
-	if (obi_view_add_column(o_view, ECOTAG_STATUS_COLUMN_NAME, -1, NULL, OBI_BOOL, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)
+	if (obi_view_add_column(o_view, ECOTAG_STATUS_COLUMN_NAME, -1, NULL, OBI_BOOL, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)
 	{
 		obidebug(1, "\nError creating the column for the assignment status in ecotag");
 		return -1;
 	}
 
 	// Column for array of best match ids
-	if (obi_view_add_column(o_view, ECOTAG_BEST_MATCH_IDS_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, true, false, NULL, NULL, -1, "{}", true) < 0)
+	if (obi_view_add_column(o_view, ECOTAG_BEST_MATCH_IDS_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, true, false, NULL, NULL, -1, "{}", true) < 0)
 	{
 		obidebug(1, "\nError creating the column for the array of ids of best matches in ecotag");
 		return -1;
 	}
 
 	// Column for array of best match taxids
-	if (obi_view_add_column(o_view, ECOTAG_BEST_MATCH_TAXIDS_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, true, false, NULL, NULL, -1, "{}", true) < 0)
+	if (obi_view_add_column(o_view, ECOTAG_BEST_MATCH_TAXIDS_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, true, false, NULL, NULL, -1, "{}", true) < 0)
 	{
 		obidebug(1, "\nError creating the column for the array of taxids of best matches in ecotag");
 		return -1;

src/obi_lcs.c

@@ -155,35 +155,35 @@ static int create_alignment_output_columns(Obiview_p output_view,
 										   bool normalize, int reference, bool similarity_mode)
 {
 	// Create the column for the ids of the 1st sequence aligned
-	if (obi_view_add_column(output_view, ID1_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, id1_indexer_name, NULL, -1, ID1_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(output_view, ID1_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, id1_indexer_name, NULL, -1, ID1_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the first column for the sequence ids when aligning");
 		return -1;
 	}
 
 	// Create the column for the ids of the 2nd sequence aligned
-	if (obi_view_add_column(output_view, ID2_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, id2_indexer_name, NULL, -1, ID2_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(output_view, ID2_COLUMN_NAME, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, id2_indexer_name, NULL, -1, ID2_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the second column for the sequence ids when aligning");
 		return -1;
 	}
 
 	// Create the column for the index (in the input view) of the first sequences aligned
-	if (obi_view_add_column(output_view, IDX1_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, IDX1_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(output_view, IDX1_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, IDX1_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the first column for the sequence indices when aligning");
 		return -1;
 	}
 
 	// Create the column for the index (in the input view) of the second sequences aligned
-	if (obi_view_add_column(output_view, IDX2_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, IDX2_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(output_view, IDX2_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, IDX2_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the second column for the sequence indices when aligning");
 		return -1;
 	}
 
 	// Create the column for the LCS length
-	if (obi_view_add_column(output_view, LCS_LENGTH_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, LCS_LENGTH_COLUMN_COMMENTS, true) < 0)
+	if (obi_view_add_column(output_view, LCS_LENGTH_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, LCS_LENGTH_COLUMN_COMMENTS, true) < 0)
 	{
 		obidebug(1, "\nError creating the column for the LCS length when aligning");
 		return -1;
@@ -192,7 +192,7 @@ static int create_alignment_output_columns(Obiview_p output_view,
 	// Create the column for the alignment length if it is computed
 	if ((reference == ALILEN) && (normalize || !similarity_mode))
 	{
-		if (obi_view_add_column(output_view, ALI_LENGTH_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, ALI_LENGTH_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(output_view, ALI_LENGTH_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, ALI_LENGTH_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the column for the alignment length when aligning");
 			return -1;
@@ -201,7 +201,7 @@ static int create_alignment_output_columns(Obiview_p output_view,
 	// Create the column for the alignment score
 	if (normalize)
 	{
-		if (obi_view_add_column(output_view, SCORE_COLUMN_NAME, -1, NULL, OBI_FLOAT, 0, 1, NULL, false, false, false, NULL, NULL, -1, SCORE_COLUMN_NAME, true) < 0)
+		if (obi_view_add_column(output_view, SCORE_COLUMN_NAME, -1, NULL, OBI_FLOAT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, SCORE_COLUMN_NAME, true) < 0)
 		{
 			obidebug(1, "\nError creating the column for the score when aligning");
 			return -1;
@@ -209,7 +209,7 @@ static int create_alignment_output_columns(Obiview_p output_view,
 	}
 	else
 	{
-		if (obi_view_add_column(output_view, SCORE_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, SCORE_COLUMN_NAME, true) < 0)
+		if (obi_view_add_column(output_view, SCORE_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, SCORE_COLUMN_NAME, true) < 0)
 		{
 			obidebug(1, "\nError creating the column for the score when aligning");
 			return -1;
@@ -219,14 +219,14 @@ static int create_alignment_output_columns(Obiview_p output_view,
 	if (print_seq)
 	{
 		// Create the column for the first sequences aligned
-		if (obi_view_add_column(output_view, SEQ1_COLUMN_NAME, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, seq1_indexer_name, NULL, -1, SEQ1_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(output_view, SEQ1_COLUMN_NAME, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, false, seq1_indexer_name, NULL, -1, SEQ1_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the first column for the sequences when aligning");
 			return -1;
 		}
 
 		// Create the column for the second sequences aligned
-		if (obi_view_add_column(output_view, SEQ2_COLUMN_NAME, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, seq2_indexer_name, NULL, -1, SEQ2_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(output_view, SEQ2_COLUMN_NAME, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, false, seq2_indexer_name, NULL, -1, SEQ2_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the second column for the sequences when aligning");
 			return -1;
@@ -235,14 +235,14 @@ static int create_alignment_output_columns(Obiview_p output_view,
 	if (print_count)
 	{
 		// Create the column for the count of the first sequences aligned
-		if (obi_view_add_column(output_view, COUNT1_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, COUNT1_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(output_view, COUNT1_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, COUNT1_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the first column for the sequence counts when aligning");
 			return -1;
 		}
 
 		// Create the column for the count of the second sequences aligned
-		if (obi_view_add_column(output_view, COUNT2_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, COUNT2_COLUMN_COMMENTS, true) < 0)
+		if (obi_view_add_column(output_view, COUNT2_COLUMN_NAME, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, COUNT2_COLUMN_COMMENTS, true) < 0)
 		{
 			obidebug(1, "\nError creating the second column for the sequence counts when aligning");
 			return -1;

src/obiavl.c

@@ -582,6 +582,7 @@ int truncate_avl_to_size_used(OBIDMS_avl_p avl)	// TODO is it necessary to unmap
 {
 	size_t  			file_size;
 	size_t  			new_data_size;
+	size_t  			header_size;
 	double  			multiple;
 	int 				file_descriptor;
 
@@ -589,6 +590,8 @@ int truncate_avl_to_size_used(OBIDMS_avl_p avl)	// TODO is it necessary to unmap
 	multiple = ceil((double) (ONE_IF_ZERO((avl->header)->nb_items * sizeof(AVL_node_t))) / (double) getpagesize());
 	new_data_size = ((size_t) multiple) * getpagesize();
 
+	header_size = (avl->header)->header_size;
+
 	// Check that it is actually greater than the current size of the file, otherwise no need to truncate
 	if ((avl->header)->avl_size == new_data_size)
 		return 0;
@@ -596,16 +599,22 @@ int truncate_avl_to_size_used(OBIDMS_avl_p avl)	// TODO is it necessary to unmap
 	// Get the file descriptor
 	file_descriptor = avl->avl_fd;
 
-	// Unmap the tree before truncating the file
+	// Unmap the entire file before truncating it (WSL requirement)
 	if (munmap(avl->tree, (avl->header)->avl_size) < 0)
 	{
 		obi_set_errno(OBI_AVL_ERROR);
 		obidebug(1, "\nError munmapping the tree of an AVL before truncating");
 		return -1;
 	}
+	if (munmap(avl->header, header_size) < 0)
+	{
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nError munmapping the tree of an AVL before truncating");
+		return -1;
+	}
 
 	// Truncate the file
-	file_size = (avl->header)->header_size + new_data_size;
+	file_size = header_size + new_data_size;
 	if (ftruncate(file_descriptor, file_size) < 0)
 	{
 		obi_set_errno(OBI_AVL_ERROR);
@@ -613,7 +622,22 @@ int truncate_avl_to_size_used(OBIDMS_avl_p avl)	// TODO is it necessary to unmap
 		return -1;
 	}
 
-	// Remap the data
+	// Remap the header and the data
+
+	avl->header    = mmap(NULL,
+			         header_size,
+					 PROT_READ | PROT_WRITE,
+					 MAP_SHARED,
+					 file_descriptor,
+					 0
+					);
+	if (avl->header == MAP_FAILED)
+	{
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nError re-mmapping the header of an AVL after truncating");
+		return -1;
+	}
+
 	avl->tree = mmap(NULL,
 			         new_data_size,
 					 PROT_READ | PROT_WRITE,
@@ -640,6 +664,7 @@ int truncate_avl_data_to_size_used(OBIDMS_avl_data_p avl_data)	// TODO is it nec
 {
 	size_t  			file_size;
 	index_t  			new_data_size;
+	size_t  			header_size;
 	double  			multiple;
 	int 				file_descriptor;
 
@@ -647,6 +672,8 @@ int truncate_avl_data_to_size_used(OBIDMS_avl_data_p avl_data)	// TODO is it nec
 	multiple = ceil((double) (ONE_IF_ZERO((avl_data->header)->data_size_used)) / (double) getpagesize());
 	new_data_size = ((index_t) multiple) * getpagesize();
 
+	header_size = (avl_data->header)->header_size;
+
 	// Check that it is actually greater than the current size of the file, otherwise no need to truncate
 	if ((avl_data->header)->data_size_max >= new_data_size)
 		return 0;
@@ -654,7 +681,8 @@ int truncate_avl_data_to_size_used(OBIDMS_avl_data_p avl_data)	// TODO is it nec
 	// Get the file descriptor
 	file_descriptor = avl_data->data_fd;
 
-	// Unmap the data before truncating the file
+	// Unmap the entire file before truncating it (WSL requirement)
+
 	if (munmap(avl_data->data, (avl_data->header)->data_size_max) < 0)
 	{
 		obi_set_errno(OBI_AVL_ERROR);
@@ -662,8 +690,15 @@ int truncate_avl_data_to_size_used(OBIDMS_avl_data_p avl_data)	// TODO is it nec
 		return -1;
 	}
 
+	if (munmap(avl_data->header, header_size) < 0)
+	{
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nError munmapping the header of an AVL before truncating");
+		return -1;
+	}
+
 	// Truncate the file
-	file_size = (avl_data->header)->header_size + new_data_size;
+	file_size = header_size + new_data_size;
 	if (ftruncate(file_descriptor, file_size) < 0)
 	{
 		obi_set_errno(OBI_AVL_ERROR);
@@ -672,6 +707,22 @@ int truncate_avl_data_to_size_used(OBIDMS_avl_data_p avl_data)	// TODO is it nec
 	}
 
 	// Remap the data
+
+	avl_data->header = mmap(NULL,
+					        header_size,
+						    PROT_READ | PROT_WRITE,
+						    MAP_SHARED,
+						    file_descriptor,
+						    0
+					      );
+
+	if (avl_data->header == MAP_FAILED)
+	{
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nError re-mmapping the header of an AVL after truncating");
+		return -1;
+	}
+
 	avl_data->data = mmap(NULL,
 					      new_data_size,
 						  PROT_READ | PROT_WRITE,
@@ -710,6 +761,20 @@ int grow_avl(OBIDMS_avl_p avl)		// TODO Lock when needed
 	header_size = (avl->header)->header_size;
 	file_size = header_size + new_data_size;
 
+	// Unmap the entire file before truncating it (WSL requirement)
+	if (munmap(avl->tree, old_data_size) < 0)
+	{
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nError munmapping the tree of an AVL tree file before enlarging");
+		return -1;
+	}
+	if (munmap(avl->header, header_size) < 0)
+	{
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nError munmapping the header of an AVL tree file before enlarging");
+		return -1;
+	}
+
 	// Enlarge the file
 	if (ftruncate(avl_file_descriptor, file_size) < 0)
 	{
@@ -718,12 +783,20 @@ int grow_avl(OBIDMS_avl_p avl)		// TODO Lock when needed
 		return -1;
 	}
 
-	// Unmap and re-map the data
+	// Re-map
 
-	if (munmap(avl->tree, old_data_size) < 0)
+	avl->header = mmap(NULL,
+					   header_size,
+					   PROT_READ | PROT_WRITE,
+					   MAP_SHARED,
+					   avl_file_descriptor,
+					   0
+					  );
+
+	if (avl->header == MAP_FAILED)
 	{
 		obi_set_errno(OBI_AVL_ERROR);
-		obidebug(1, "\nError munmapping the tree of an AVL tree file before enlarging");
+		obidebug(1, "\nError re-mmapping the header of an AVL tree file after enlarging the file");
 		return -1;
 	}
 
@@ -768,6 +841,20 @@ int grow_avl_data(OBIDMS_avl_data_p avl_data)		// TODO Lock when needed
 	header_size = (avl_data->header)->header_size;
 	file_size = header_size + new_data_size;
 
+	// Unmap the entire file before truncating it (WSL requirement)
+	if (munmap(avl_data->data, old_data_size) < 0)
+	{
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nError munmapping the data of an AVL tree data file before enlarging");
+		return -1;
+	}
+	if (munmap(avl_data->header, header_size) < 0)
+	{
+		obi_set_errno(OBI_AVL_ERROR);
+		obidebug(1, "\nError munmapping the header of an AVL tree data file before enlarging");
+		return -1;
+	}
+
 	// Enlarge the file
 	if (ftruncate(avl_data_file_descriptor, file_size) < 0)
 	{
@@ -776,12 +863,19 @@ int grow_avl_data(OBIDMS_avl_data_p avl_data)		// TODO Lock when needed
 		return -1;
 	}
 
-	// Unmap and re-map the data
+	// Re-map
 
-	if (munmap(avl_data->data, old_data_size) < 0)
+	avl_data->header = mmap(NULL,
+						    header_size,
+						    PROT_READ | PROT_WRITE,
+						    MAP_SHARED,
+						    avl_data_file_descriptor,
+						    0
+						   );
+	if (avl_data->header == MAP_FAILED)
 	{
 		obi_set_errno(OBI_AVL_ERROR);
-		obidebug(1, "\nError munmapping the data of an AVL tree data file before enlarging");
+		obidebug(1, "\nError re-mmapping the header of an AVL tree data file after enlarging the file");
 		return -1;
 	}
 
@@ -792,7 +886,6 @@ int grow_avl_data(OBIDMS_avl_data_p avl_data)		// TODO Lock when needed
 						  avl_data_file_descriptor,
 						  header_size
 						 );
-
 	if (avl_data->data == MAP_FAILED)
 	{
 		obi_set_errno(OBI_AVL_ERROR);

src/obidms.c

@@ -316,6 +316,15 @@ static int enlarge_infos_file(OBIDMS_p dms, size_t new_size)
 	multiple = 	ceil((double) new_size / (double) getpagesize());
 	rounded_new_size = multiple * getpagesize();
 
+	// Unmap the entire file before truncating it (WSL requirement)
+	if (munmap(dms->infos, (dms->infos)->file_size) < 0)
+	{
+		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
+		obidebug(1, "\nError munmapping a DMS information file when enlarging");
+		close(infos_file_descriptor);
+		return -1;
+	}
+
 	// Enlarge the file
 	if (ftruncate(infos_file_descriptor, rounded_new_size) < 0)
 	{
@@ -325,15 +334,7 @@ static int enlarge_infos_file(OBIDMS_p dms, size_t new_size)
 		return -1;
 	}
 
-	// Unmap and remap the file
-	if (munmap(dms->infos, (dms->infos)->file_size) < 0)
-	{
-		obi_set_errno(OBIDMS_UNKNOWN_ERROR);
-		obidebug(1, "\nError munmapping a DMS information file when enlarging");
-		close(infos_file_descriptor);
-		return -1;
-	}
-
+	// Remap the file
 	dms->infos = mmap(NULL,
 					  rounded_new_size,
 					  PROT_READ | PROT_WRITE,
@@ -1575,8 +1576,8 @@ obiversion_t obi_import_column(const char* dms_path_1, const char* dms_path_2, c
 
 	// Create new column
 	column_2 = obi_create_column(dms_2, column_name, header_1->returned_data_type, header_1->line_count,
-								 header_1->nb_elements_per_line, header_1->elements_names, true, header_1->tuples,
-								 header_1->to_eval, new_avl_name, (header_1->associated_column).column_name,
+								 header_1->nb_elements_per_line, header_1->elements_names, true, header_1->dict_column,
+								 header_1->tuples, header_1->to_eval, new_avl_name, (header_1->associated_column).column_name,
 								 (header_1->associated_column).version, header_1->comments);
 
 	if (column_2 == NULL)
@@ -1813,6 +1814,7 @@ int obi_import_view(const char* dms_path_1, const char* dms_path_2, const char*
 								false,
 								false,
 								false,
+								false,
 								NULL,
 								NULL,
 								-1,

src/obidms.h

@@ -40,7 +40,7 @@
                                 	 	 	 	 */
 #define MAX_NB_OPENED_INDEXERS (1000) 			/**< The maximum number of indexers open at the same time.
                                 	 	 	 	 */
-#define MAX_PATH_LEN (1024)						/**< Maximum length for the character string defining a
+#define MAX_PATH_LEN (2048)						/**< Maximum length for the character string defining a
  	 	 	 	 	 	 	 	 	 	 	 	 *	 file or directory path.
  	 	 	 	 	 	 	 	 	 	 	 	 */
 

src/obidmscolumn.c

@@ -1024,6 +1024,7 @@ OBIDMS_column_p obi_create_column(OBIDMS_p     dms,
 								  index_t      nb_elements_per_line,
 								  char*        elements_names,
 								  bool		   elt_names_formatted,
+								  bool         dict_column,
 								  bool		   tuples,
 								  bool         to_eval,
 								  const char*  indexer_name,
@@ -1282,6 +1283,7 @@ OBIDMS_column_p obi_create_column(OBIDMS_p     dms,
 	header->nb_elements_per_line          = nb_elements_per_line;
 	header->stored_data_type     		  = stored_data_type;
 	header->returned_data_type   		  = returned_data_type;
+	header->dict_column     			  = dict_column;
 	header->tuples     					  = tuples;
 	header->to_eval                       = to_eval;
 	header->creation_date 		 		  = time(NULL);
@@ -1611,6 +1613,7 @@ OBIDMS_column_p obi_clone_column(OBIDMS_p         dms,
 								   nb_elements_per_line,
 								   (column_to_clone->header)->elements_names,
 								   true,
+								   (column_to_clone->header)->dict_column,
 								   (column_to_clone->header)->tuples,
 								   (column_to_clone->header)->to_eval,
 								   (column_to_clone->header)->indexer_name,
@@ -1766,6 +1769,7 @@ int obi_truncate_column(OBIDMS_column_p column)	// TODO is it necessary to unmap
 {
 	size_t  file_size;
 	size_t  data_size;
+	size_t  header_size;
 	index_t new_line_count;
 	double  multiple;
 	int 	column_file_descriptor;
@@ -1788,6 +1792,8 @@ int obi_truncate_column(OBIDMS_column_p column)	// TODO is it necessary to unmap
 
 	data_size = obi_array_sizeof((column->header)->stored_data_type, new_line_count, (column->header)->nb_elements_per_line);
 
+	header_size = (column->header)->header_size;
+
 	// Check that it is actually greater than the current data size, otherwise no need to truncate
 	if ((column->header)->data_size == data_size)
 		return 0;
@@ -1852,7 +1858,7 @@ int obi_truncate_column(OBIDMS_column_p column)	// TODO is it necessary to unmap
 		return -1;
 	}
 
-	// Unmap the data before truncating the file
+	// Unmap the entire file before truncating it (WSL requirement)
 	if (munmap(column->data, (column->header)->data_size) < 0)
 	{
 		obi_set_errno(OBICOL_UNKNOWN_ERROR);
@@ -1860,9 +1866,16 @@ int obi_truncate_column(OBIDMS_column_p column)	// TODO is it necessary to unmap
 		close(column_file_descriptor);
 		return -1;
 	}
+	if (munmap(column->header, header_size) < 0)
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError munmapping the header of a column before truncating");
+		close(column_file_descriptor);
+		return -1;
+	}
 
 	// Truncate the column file
-	file_size = (column->header)->header_size + data_size;
+	file_size = header_size + data_size;
 	if (ftruncate(column_file_descriptor, file_size) < 0)
 	{
 		obi_set_errno(OBICOL_UNKNOWN_ERROR);
@@ -1871,13 +1884,30 @@ int obi_truncate_column(OBIDMS_column_p column)	// TODO is it necessary to unmap
 		return -1;
 	}
 
-	// Remap the data
+	// Remap the header and the data
+
+	column->header = mmap(NULL,
+						  header_size,
+						  PROT_READ | PROT_WRITE,
+						  MAP_SHARED,
+						  column_file_descriptor,
+						  0
+					     );
+
+	if (column->header == MAP_FAILED)
+	{
+		obi_set_errno(OBICOL_UNKNOWN_ERROR);
+		obidebug(1, "\nError re-mmapping the header of a column after truncating");
+		close(column_file_descriptor);
+		return -1;
+	}
+
 	column->data = mmap(NULL,
 						data_size,
 						PROT_READ | PROT_WRITE,
 						MAP_SHARED,
 						column_file_descriptor,
-						(column->header)->header_size
+						header_size
 					   );
 
 	if (column->data == MAP_FAILED)

src/obidmscolumn.h

@@ -77,6 +77,8 @@ typedef struct OBIDMS_column_header {
 	OBIType_t			stored_data_type;		    						/**< Type of the data that is actually stored in the data
 															 	 	 	 	 *   part of the column.
 															 	 	 	 	 */
+	bool                dict_column;										/**< Whether the column contains dictionary-like values.
+	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 	 */
 	bool				tuples;												/**< A boolean indicating whether the column contains indices referring to indexed tuples.
 																			 */
 	bool				to_eval;											/**< A boolean indicating whether the column contains expressions that should be evaluated
@@ -249,6 +251,7 @@ size_t obi_calculate_header_size(index_t nb_elements_per_line, int64_t elts_name
  * @param elements_names The names of the elements with ';' as separator (no terminal ';'),
  *                       NULL or "" if the default names are to be used ("0\01\02\0...\0n").
  * @param elt_names_formatted Whether the separator for the elements names is ';' (false), or '\0' (true, as formatted by format_elements_names()).
+ * @param dict_column A boolean indicating whether the column should contain dictionary-like values.
  * @param tuples A boolean indicating whether the column should contain indices referring to indexed tuples.
  * @param to_eval A boolean indicating whether the column contains expressions that should be evaluated
  *                (typically OBI_STR columns containing character strings to be evaluated by Python).
@@ -271,6 +274,7 @@ OBIDMS_column_p obi_create_column(OBIDMS_p     dms,
 								  index_t      nb_elements_per_line,
 								  char*        elements_names,
 								  bool		   elt_names_formatted,
+								  bool         dict_column,
 								  bool         tuples,
 								  bool         to_eval,
 								  const char*  indexer_name,

src/obiview.c

@@ -638,6 +638,15 @@ static int enlarge_view_file(Obiview_p view, size_t new_size)
 	multiple = 	ceil((double) new_size / (double) getpagesize());
 	rounded_new_size = multiple * getpagesize();
 
+	// Unmap the entire file before truncating it (WSL requirement)
+	if (munmap(view->infos, (view->infos)->file_size) < 0)
+	{
+		obi_set_errno(OBIVIEW_ERROR);
+		obidebug(1, "\nError munmapping a view file when enlarging");
+		close(obiview_file_descriptor);
+		return -1;
+	}
+
 	// Enlarge the file
 	if (ftruncate(obiview_file_descriptor, rounded_new_size) < 0)
 	{
@@ -647,15 +656,7 @@ static int enlarge_view_file(Obiview_p view, size_t new_size)
 		return -1;
 	}
 
-	// Unmap and remap the file
-	if (munmap(view->infos, (view->infos)->file_size) < 0)
-	{
-		obi_set_errno(OBIVIEW_ERROR);
-		obidebug(1, "\nError munmapping a view file when enlarging");
-		close(obiview_file_descriptor);
-		return -1;
-	}
-
+	// Remap the file
 	view->infos = mmap(NULL,
 					   rounded_new_size,
 					   PROT_READ | PROT_WRITE,
@@ -1712,7 +1713,7 @@ Obiview_p obi_new_view(OBIDMS_p dms, const char* view_name, Obiview_p view_to_cl
 		// If there is a new line selection, build it by combining it with the one from the view to clone if there is one
 		else if (line_selection != NULL)
 		{
-			view->line_selection = obi_create_column(view->dms, LINES_COLUMN_NAME, OBI_IDX, 0, 1, NULL, false, false, false, NULL, NULL, -1, NULL);
+			view->line_selection = obi_create_column(view->dms, LINES_COLUMN_NAME, OBI_IDX, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, NULL);
 			if ((view->line_selection) == NULL)
 			{
 				obidebug(1, "\nError creating a column corresponding to a line selection");
@@ -1862,6 +1863,7 @@ Obiview_p obi_new_view(OBIDMS_p dms, const char* view_name, Obiview_p view_to_cl
 									false,
 									false,
 									false,
+									false,
 									NULL,
 									NULL,
 									-1,
@@ -1930,19 +1932,19 @@ Obiview_p obi_new_view_nuc_seqs(OBIDMS_p dms, const char* view_name, Obiview_p v
 	if ((view_to_clone == NULL) && create_default_columns)
 	{
 		// Adding sequence column
-		if (obi_view_add_column(view, NUC_SEQUENCE_COLUMN, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)	// discuss using same indexer "NUC_SEQ_INDEXER"
+		if (obi_view_add_column(view, NUC_SEQUENCE_COLUMN, -1, NULL, OBI_SEQ, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)	// discuss using same indexer "NUC_SEQ_INDEXER"
 		{
 			obidebug(1, "Error adding an obligatory column in a nucleotide sequences view");
 			return NULL;
 		}
 		// Adding id column
-		if (obi_view_add_column(view, ID_COLUMN, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)
+		if (obi_view_add_column(view, ID_COLUMN, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)
 		{
 			obidebug(1, "Error adding an obligatory column in a nucleotide sequences view");
 			return NULL;
 		}
 		// Adding definition column
-		if (obi_view_add_column(view, DEFINITION_COLUMN, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)
+		if (obi_view_add_column(view, DEFINITION_COLUMN, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)
 		{
 			obidebug(1, "Error adding an obligatory column in a nucleotide sequences view");
 			return NULL;
@@ -1951,7 +1953,7 @@ Obiview_p obi_new_view_nuc_seqs(OBIDMS_p dms, const char* view_name, Obiview_p v
 		if (quality_column)
 		{
 			associated_nuc_column = obi_view_get_column(view, NUC_SEQUENCE_COLUMN);
-			if (obi_view_add_column(view, QUALITY_COLUMN, -1, NULL, OBI_QUAL, 0, 1, NULL, false, false, false, NULL, (associated_nuc_column->header)->name, (associated_nuc_column->header)->version, "{}", true) < 0)		// TODO discuss automatic association
+			if (obi_view_add_column(view, QUALITY_COLUMN, -1, NULL, OBI_QUAL, 0, 1, NULL, false, false, false, false, NULL, (associated_nuc_column->header)->name, (associated_nuc_column->header)->version, "{}", true) < 0)		// TODO discuss automatic association
 			{
 				obidebug(1, "Error adding an obligatory column in a nucleotide sequences view");
 				return NULL;
@@ -2284,6 +2286,7 @@ int obi_view_add_column(Obiview_p    view,
 						index_t      nb_elements_per_line,
 						char*        elements_names,
 						bool         elt_names_formatted,
+						bool         dict_column,
 						bool		 tuples,
 						bool         to_eval,
 						const char*  indexer_name,
@@ -2366,7 +2369,7 @@ int obi_view_add_column(Obiview_p    view,
 	// Open or create the column
 	if (create)
 	{	// Create column
-		column = obi_create_column(view->dms, column_name, data_type, nb_lines, nb_elements_per_line, elements_names, elt_names_formatted, tuples, to_eval, indexer_name, associated_column_name, associated_column_version, comments);
+		column = obi_create_column(view->dms, column_name, data_type, nb_lines, nb_elements_per_line, elements_names, elt_names_formatted, dict_column, tuples, to_eval, indexer_name, associated_column_name, associated_column_version, comments);
 		if (column == NULL)
 		{
 			obidebug(1, "\nError creating a column to add to a view");
@@ -3230,7 +3233,7 @@ int obi_create_auto_count_column(Obiview_p view)
 		return -1;
 	}
 
-	if (obi_view_add_column(view, COUNT_COLUMN, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)
+	if (obi_view_add_column(view, COUNT_COLUMN, -1, NULL, OBI_INT, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)
 	{
 		obidebug(1, "Error adding an automatic count column in a view");
 		return -1;
@@ -3282,7 +3285,7 @@ int obi_create_auto_id_column(Obiview_p view, const char* prefix)
 	}
 
 	// Create the new ID column
-	if (obi_view_add_column(view, ID_COLUMN, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, NULL, NULL, -1, "{}", true) < 0)
+	if (obi_view_add_column(view, ID_COLUMN, -1, NULL, OBI_STR, 0, 1, NULL, false, false, false, false, NULL, NULL, -1, "{}", true) < 0)
 	{
 		obidebug(1, "Error adding an automatic ID column in a view");
 		return -1;

src/obiview.h

@@ -400,6 +400,7 @@ Obiview_p obi_open_view(OBIDMS_p dms, const char* view_name);
  * @param elements_names The names of the elements with ';' as separator (no terminal ';'),
  * 						 if the column is created; NULL or "" if the default names are to be used ("0\01\02\0...\0n").
  * @param elt_names_formatted Whether the separator for the elements names is ';' (false), or '\0' (true, as formatted by format_elements_names()).
+ * @param dict_column Whether the column contains dictionary-like values.
  * @param tuples A boolean indicating whether the column should contain indices referring to indexed tuples.
  * @param to_eval A boolean indicating whether the column contains expressions that should be evaluated
  *                (typically OBI_STR columns containing character strings to be evaluated by Python).
@@ -426,6 +427,7 @@ int obi_view_add_column(Obiview_p    view,
 						index_t      nb_elements_per_line,
 						char*        elements_names,
 						bool         elt_names_formatted,
+						bool         dict_column,
 						bool         tuples,
 						bool         to_eval,
 						const char*  indexer_name,