switch to version 3.0.1b1

size (#68)

MANIFEST.in

@@ -6,7 +6,7 @@ recursive-include doc/sphinx/source *.txt *.rst *.py
 recursive-include doc/sphinx/sphinxext *.py
 include doc/sphinx/Makefile
 include doc/sphinx/Doxyfile
-include README.txt
+include README.md
 include requirements.txt
 include scripts/obi
 

obi_completion_script.bash

@@ -1,4 +1,3 @@
-#/usr/bin/env bash
 
 _obi_comp ()
 {

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/apps/progress.pyx

@@ -30,12 +30,12 @@ cdef class ProgressBar:
                  off_t maxi,
                  dict  config={},
                  str head="",
-                 double seconde=0.1,
+                 double seconds=5,
                  cut=False):
         
         self.starttime = self.clock()
         self.lasttime  = self.starttime
-        self.tickcount = <clock_t> (seconde * CLOCKS_PER_SEC)
+        self.tickcount = <clock_t> (seconds * CLOCKS_PER_SEC)
         self.freq      = 1
         self.cycle     = 0
         self.arrow     = 0

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/align.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 addMinimalInputOption, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.dms.view import RollbackException
 from obitools3.apps.config import logger
 from obitools3.utils cimport tobytes, str2bytes
@@ -12,6 +12,9 @@ from obitools3.utils cimport tobytes, str2bytes
 from obitools3.dms.capi.obilcsalign cimport obi_lcs_align_one_column, \
                                             obi_lcs_align_two_columns
 
+from io import BufferedWriter
+from cpython.exc cimport PyErr_CheckSignals
+
 import time
 import sys
 
@@ -23,6 +26,7 @@ def addOptions(parser):
    
    addMinimalInputOption(parser)
    addMinimalOutputOption(parser)
+   addNoProgressBarOption(parser)
 
    group=parser.add_argument_group('obi align specific options')
 
@@ -201,20 +205,20 @@ def run(config):
     if output is None:
         raise Exception("Could not create output")
     o_dms = output[0]
+    output_0 = output[0]
     o_dms_name = o_dms.name 
     final_o_view_name = output[1]
+    o_view_name = final_o_view_name
 
-    # If the input and output DMS are not the same, align creating a temporary view in the input dms that will be exported to 
+    # If stdout output or the input and output DMS are not the same, align creating a temporary view in the input dms that will be exported to 
     # the right DMS and deleted in the other afterwards.
-    if i_dms != o_dms:
-        temporary_view_name = final_o_view_name
-        i=0
-        while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
-            temporary_view_name = final_o_view_name+b"_"+str2bytes(str(i))
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        if type(output_0)==BufferedWriter:
+            o_dms = i_dms
+        o_view_name = b"temp"
+        while o_view_name in i_dms:   # Making sure view name is unique in input DMS
+            o_view_name = o_view_name+b"_"+str2bytes(str(i))
             i+=1
-        o_view_name = temporary_view_name
-    else:
-        o_view_name = final_o_view_name
         
     # Save command config in View comments
     command_line = " ".join(sys.argv[1:])
@@ -263,12 +267,19 @@ def run(config):
         View.delete_view(i_dms, i_view_name_2)
         i_dms_2.close()
 
+    # stdout output: write to buffer
+    if type(output_0)==BufferedWriter:
+        logger("info", "Printing to output...")
+        o_view = o_dms[o_view_name]
+        o_view.print_to_output(output_0, noprogressbar=config['obi']['noprogressbar'])
+        o_view.close()
+
     # If the input and the output DMS are different, delete the temporary result view in the input DMS
-    if i_dms != o_dms:
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
         View.delete_view(i_dms, o_view_name)
-        o_dms.close()
+        o_dms.close(force=True)
     
-    i_dms.close()
+    i_dms.close(force=True)
     
     logger("info", "Done.")
   

python/obitools3/commands/alignpairedend.pyx

@@ -6,7 +6,7 @@ from obitools3.dms.view.typed_view.view_NUC_SEQS cimport View_NUC_SEQS
 from obitools3.dms.column.column cimport Column
 from obitools3.dms.capi.obiview cimport QUALITY_COLUMN
 from obitools3.dms.capi.obitypes cimport OBI_QUAL
-from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.uri.decode import open_uri
 from obitools3.apps.config import logger
 from obitools3.libalign._qsassemble import QSolexaReverseAssemble
@@ -15,7 +15,9 @@ from obitools3.libalign._solexapairend import buildConsensus, buildJoinedSequenc
 from obitools3.dms.obiseq cimport Nuc_Seq
 from obitools3.libalign.shifted_ali cimport Kmer_similarity, Ali_shifted
 from obitools3.dms.capi.obiview cimport REVERSE_SEQUENCE_COLUMN, REVERSE_QUALITY_COLUMN
+from obitools3.utils cimport str2bytes
 
+from io import BufferedWriter
 import sys
 import os
 
@@ -29,6 +31,7 @@ def addOptions(parser):
 
     addMinimalInputOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
  
     group = parser.add_argument_group('obi alignpairedend specific options')
 
@@ -39,12 +42,13 @@ def addOptions(parser):
                      type=str,
                      help="URI to the reverse reads if they are in a different view than the forward reads")
 
-    group.add_argument('--score-min',
-                     action="store", dest="alignpairedend:smin",
-                     metavar="#.###",
-                     default=None,
-                     type=float,
-                     help="Minimum score for keeping alignments")
+#     group.add_argument('--score-min',
+#                      action="store", dest="alignpairedend:smin",
+#                      metavar="#.###",
+#                      default=None,
+#                      type=float,
+#                      help="Minimum score for keeping alignments. " 
+#                           "(for kmer alignment) The score is an approximation of the number of nucleotides matching in the overlap of the alignment.")
 
 #     group.add_argument('-A', '--true-ali',
 #                        action="store_true", dest="alignpairedend:trueali",
@@ -170,18 +174,29 @@ def run(config):
     if output is None:
         raise Exception("Could not create output view")
 
-    view = output[1]
+    output_0 = output[0]
+    o_dms = output[0]
     
-    Column.new_column(view, QUALITY_COLUMN, OBI_QUAL)   #TODO output URI quality option?
-
-    if 'smin' in config['alignpairedend']:
-        smin = config['alignpairedend']['smin']
+    # stdout output: create temporary view
+    if type(output_0)==BufferedWriter:
+        i_dms = forward.dms   # using any dms
+        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 input DMS
+            o_view_name = o_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        o_view = View_NUC_SEQS.new(o_dms, o_view_name, quality=True)
     else:
-        smin = 0
+        o_view = output[1]
+        Column.new_column(o_view, QUALITY_COLUMN, OBI_QUAL)
 
     # Initialize the progress bar
-    pb = ProgressBar(entries_len, config, seconde=5)
-    
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(entries_len, config)
+    else:
+        pb = None
+         
     #if config['alignpairedend']['trueali']:
     #    kmer_ali = False
     #    aligner = buildAlignment
@@ -206,18 +221,19 @@ def run(config):
     i = 0
     for ali in ba:
         
-        pb(i)
+        if pb is not None:
+            pb(i)
         
         PyErr_CheckSignals()
 
-        consensus = view[i]
+        consensus = o_view[i]
                
         if not two_views:
             seqF = entries[i]
         else:
             seqF = forward[i]
                         
-        if ali.score > smin and ali.consensus_len > 0 :
+        if ali.overlap_len > 0 :
             buildConsensus(ali, consensus, seqF)
         else:
             if not two_views:
@@ -225,32 +241,43 @@ def run(config):
             else:
                 seqR = reverse[i]
             buildJoinedSequence(ali, seqR, consensus, forward=seqF)
-        
-        consensus[b"smin"] = smin
-           
+                   
         if kmer_ali :
             ali.free()
     
         i+=1
 
-    pb(i, force=True)
-    print("", file=sys.stderr)
+    if pb is not None:
+        pb(i, force=True)
+        print("", file=sys.stderr)
 
     if kmer_ali :
         aligner.free()
 
     # Save command config in View and DMS comments
     command_line = " ".join(sys.argv[1:])
-    view.write_config(config, "alignpairedend", command_line, input_dms_name=input_dms_name, input_view_name=input_view_name)
-    output[0].record_command_line(command_line)
+    o_view.write_config(config, "alignpairedend", 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(view), file=sys.stderr)
-    
-    input[0].close()
+
+    # 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 stdout output, delete the temporary imported view used to create the final file
+    if type(output_0)==BufferedWriter:
+        View_NUC_SEQS.delete_view(o_dms, o_view_name)
+        output_0.close()
+
+    # Close all DMS
+    input[0].close(force=True)
     if two_views:
-        rinput[0].close()
-    output[0].close()
+        rinput[0].close(force=True)
+    o_dms.close(force=True)
 
     logger("info", "Done.")
     

python/obitools3/commands/annotate.pyx

@@ -4,16 +4,18 @@ 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
+from obitools3.apps.optiongroups import addMinimalInputOption, addTaxonomyOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.dms.view import RollbackException
 from functools import reduce
 from obitools3.apps.config import logger
 from obitools3.utils cimport tobytes, str2bytes
+from io import BufferedWriter
 from obitools3.dms.capi.obiview cimport NUC_SEQUENCE_COLUMN, \
                                         ID_COLUMN, \
                                         DEFINITION_COLUMN, \
                                         QUALITY_COLUMN, \
-                                        COUNT_COLUMN
+                                        COUNT_COLUMN, \
+                                        TAXID_COLUMN
 
 import time
 import math 
@@ -33,6 +35,7 @@ def addOptions(parser):
     addMinimalInputOption(parser)
     addTaxonomyOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
 
     group=parser.add_argument_group('obi annotate specific options')
  
@@ -175,8 +178,8 @@ def sequenceTaggerGenerator(config, taxo=None):
             counter[0]+=1
                             
         for rank in annoteRank:
-            if 'taxid' in seq:
-                taxid = seq['taxid']
+            if TAXID_COLUMN in seq:
+                taxid = seq[TAXID_COLUMN]
                 if taxid is not None:
                     rtaxid = taxo.get_taxon_at_rank(taxid, rank)
                     if rtaxid is not None:
@@ -190,58 +193,50 @@ def sequenceTaggerGenerator(config, taxo=None):
             seq['seq_rank']=counter[0]
     
         for i,v in toSet:
-            #try:
-            if taxo is not None:
-                environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
-            else:
-                environ = {'sequence':seq, 'counter':counter[0], 'math':math}
-            val = eval(v, environ, seq)
-            #except Exception,e:       # TODO discuss usefulness of this
-            #    if options.onlyValid:
-            #        raise e
-            #    val = v
+            try:
+                if taxo is not None:
+                    environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
+                else:
+                    environ = {'sequence':seq, 'counter':counter[0], 'math':math}
+                val = eval(v, environ, seq)
+            except Exception:       # set string if not a valid expression
+                val = v
             seq[i]=val
 
         if length:
             seq['seq_length']=len(seq)
 
         if newId is not None:
-#            try:
-            if taxo is not None:
-                environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
-            else:
-                environ = {'sequence':seq, 'counter':counter[0], 'math':math}     
-            val = eval(newId, environ, seq)
-#            except Exception,e:
-#                if options.onlyValid:
-#                    raise e
-#                val = newId
+            try:
+                if taxo is not None:
+                    environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
+                else:
+                    environ = {'sequence':seq, 'counter':counter[0], 'math':math}     
+                val = eval(newId, environ, seq)
+            except Exception:      # set string if not a valid expression
+                val = newId
             seq.id=val
 
         if newDef is not None:
-#            try:
-            if taxo is not None:
-                environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
-            else:
-                environ = {'sequence':seq, 'counter':counter[0], 'math':math}     
-            val = eval(newDef, environ, seq)
-#            except Exception,e:
-#                if options.onlyValid:
-#                    raise e
-#                val = newDef
+            try:
+                if taxo is not None:
+                    environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
+                else:
+                    environ = {'sequence':seq, 'counter':counter[0], 'math':math}     
+                val = eval(newDef, environ, seq)
+            except Exception:    # set string if not a valid expression
+                val = newDef
             seq.definition=val
-#             
+             
         if newSeq is not None:
-#            try:
-            if taxo is not None:
-                environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
-            else:
-                environ = {'sequence':seq, 'counter':counter[0], 'math':math}     
-            val = eval(newSeq, environ, seq)
-#            except Exception,e:
-#                if options.onlyValid:
-#                    raise e
-#                val = newSeq
+            try:
+                if taxo is not None:
+                    environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
+                else:
+                    environ = {'sequence':seq, 'counter':counter[0], 'math':math}     
+                val = eval(newSeq, environ, seq)
+            except Exception:    # set string if not a valid expression
+                val = newSeq
             seq.seq=val
             if 'seq_length' in seq:
                 seq['seq_length']=len(seq)
@@ -251,15 +246,14 @@ def sequenceTaggerGenerator(config, taxo=None):
                 seq.view.delete_column(QUALITY_COLUMN)
                     
         if run is not None:
-#            try:
-            if taxo is not None:
-                environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
-            else:
-                environ = {'sequence':seq, 'counter':counter[0], 'math':math}     
-            eval(run, environ, seq)
-#            except Exception,e:
-#                if options.onlyValid:
-#                    raise e
+            try:
+                if taxo is not None:
+                    environ = {'taxonomy' : taxo, 'sequence':seq, 'counter':counter[0], 'math':math}
+                else:
+                    environ = {'sequence':seq, 'counter':counter[0], 'math':math}     
+                eval(run, environ, seq)
+            except Exception,e:
+                raise e
 
     return sequenceTagger
 
@@ -286,8 +280,19 @@ def run(config):
     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:
@@ -298,7 +303,7 @@ def run(config):
             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:
@@ -315,7 +320,10 @@ def run(config):
         taxo = None
 
     # Initialize the progress bar
-    pb = ProgressBar(len(o_view), config, seconde=5)
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(len(o_view), config)
+    else:
+        pb = None
 
     try:
 
@@ -354,14 +362,16 @@ def run(config):
         sequenceTagger = sequenceTaggerGenerator(config, taxo=taxo)
         for i in range(len(o_view)):
             PyErr_CheckSignals()
-            pb(i)
+            if pb is not None:
+                pb(i)
             sequenceTagger(o_view[i])
 
     except Exception, e:
         raise RollbackException("obi annotate error, rollbacking view: "+str(e), o_view)
 
-    pb(i, force=True)
-    print("", file=sys.stderr)
+    if pb is not None:
+        pb(i, force=True)
+        print("", file=sys.stderr)
     
     # Save command config in View and DMS comments
     command_line = " ".join(sys.argv[1:])
@@ -371,15 +381,21 @@ def run(config):
         input_dms_name.append(config['obi']['taxoURI'].split("/")[-3])
         input_view_name.append("taxonomy/"+config['obi']['taxoURI'].split("/")[-1])
     o_view.write_config(config, "annotate", command_line, input_dms_name=input_dms_name, input_view_name=input_view_name)
-    output[0].record_command_line(command_line)
+    o_dms.record_command_line(command_line)
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_view), file=sys.stderr)
 
-    # If the input and the output DMS are different, delete the temporary imported view used to create the final view
-    if i_dms != o_dms:
+    # 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()
-    i_dms.close()
+        o_dms.close(force=True)
+    i_dms.close(force=True)
 
     logger("info", "Done.")

python/obitools3/commands/build_ref_db.pyx

@@ -4,14 +4,16 @@ from obitools3.apps.progress cimport ProgressBar  # @UnresolvedImport
 from obitools3.dms.dms cimport DMS
 from obitools3.dms.view import RollbackException
 from obitools3.dms.capi.build_reference_db cimport build_reference_db
-from obitools3.apps.optiongroups import addMinimalInputOption, addTaxonomyOption, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addTaxonomyOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.uri.decode import open_uri
 from obitools3.apps.config import logger
 from obitools3.utils cimport tobytes, str2bytes
 from obitools3.dms.view.view cimport View
 from obitools3.dms.view.typed_view.view_NUC_SEQS cimport View_NUC_SEQS
 
+from io import BufferedWriter
 import sys
+from cpython.exc cimport PyErr_CheckSignals
 
 
 __title__="Tag a set of sequences for PCR and sequencing errors identification"
@@ -22,6 +24,7 @@ def addOptions(parser):
     addMinimalInputOption(parser)
     addTaxonomyOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
  
     group = parser.add_argument_group('obi build_ref_db specific options')
 
@@ -56,17 +59,20 @@ def run(config):
     if output is None:
         raise Exception("Could not create output")
     o_dms = output[0]
+    output_0 = output[0]
     final_o_view_name = output[1]
     
-    # If the input and output DMS are not the same, build the database creating a temporary view that will be exported to 
+    # If stdout output or the input and output DMS are not the same, build the database creating a temporary view that will be exported to 
     # the right DMS and deleted in the other afterwards.
-    if i_dms != o_dms:
-        temporary_view_name = final_o_view_name
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        temporary_view_name = b"temp"
         i=0
         while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
-            temporary_view_name = final_o_view_name+b"_"+str2bytes(str(i))
+            temporary_view_name = temporary_view_name+b"_"+str2bytes(str(i))
             i+=1
         o_view_name = temporary_view_name
+        if type(output_0)==BufferedWriter:
+            o_dms = i_dms
     else:
         o_view_name = final_o_view_name
 
@@ -80,26 +86,33 @@ def run(config):
     input_dms_name.append(config['obi']['taxoURI'].split("/")[-3])
     input_view_name.append("taxonomy/"+config['obi']['taxoURI'].split("/")[-1])
     comments = View.print_config(config, "build_ref_db", command_line, input_dms_name=input_dms_name, input_view_name=input_view_name)
-
+    
     if build_reference_db(i_dms.name_with_full_path, tobytes(i_view_name), tobytes(taxonomy_name), tobytes(o_view_name), comments, config['build_ref_db']['threshold']) < 0:
         raise Exception("Error building a reference database")
     
     # If the input and output DMS are not the same, export result view to output DMS
     if i_dms != o_dms:
         View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, final_o_view_name)
-    
+
+    # stdout output: write to buffer
+    if type(output_0)==BufferedWriter:
+        logger("info", "Printing to output...")
+        o_view = o_dms[o_view_name]
+        o_view.print_to_output(output_0, noprogressbar=config['obi']['noprogressbar'])
+        o_view.close()
+
     # Save command config in DMS comments
     o_dms.record_command_line(command_line)
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_dms[final_o_view_name]), file=sys.stderr)
 
-    # If the input and the output DMS are different, delete the temporary result view in the input DMS
-    if i_dms != o_dms:
+    # 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(i_dms, o_view_name)
-        o_dms.close()
+        o_dms.close(force=True)
     
-    i_dms.close()
+    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
@@ -15,6 +15,7 @@ from obitools3.dms.capi.obiview cimport NUC_SEQUENCE_COLUMN, REVERSE_SEQUENCE_CO
 from obitools3.dms.capi.obitypes cimport OBI_SEQ, OBI_QUAL
 from obitools3.dms.column.column cimport Column
 
+from io import BufferedWriter
 import time
 import sys
  
@@ -27,6 +28,7 @@ __title__="Concatenate views."
 def addOptions(parser):
     
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
 
     group=parser.add_argument_group('obi cat specific options')
 
@@ -46,9 +48,9 @@ def run(config):
     
     logger("info", "obi cat")
 
-    # Open the views to concatenate
-    iview_list = []
+    # Check the views to concatenate
     idms_list = []
+    iview_list = []
     total_len = 0
     remove_qual = False
     remove_rev_qual = False
@@ -66,8 +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'],
@@ -76,47 +79,86 @@ def run(config):
     if output is None:
         raise Exception("Could not create output view")
     o_dms = output[0]
+    output_0 = output[0]
     o_view = output[1]
     
+    # stdout output
+    if type(output_0)==BufferedWriter: 
+        o_dms = i_dms
+    
     # Initialize quality columns and their associated sequence columns if needed
-    if not remove_qual:
-        if NUC_SEQUENCE_COLUMN not in o_view:
-            Column.new_column(o_view, NUC_SEQUENCE_COLUMN, OBI_SEQ)
-        Column.new_column(o_view, QUALITY_COLUMN, OBI_QUAL, associated_column_name=NUC_SEQUENCE_COLUMN, associated_column_version=o_view[NUC_SEQUENCE_COLUMN].version)    
-    if not remove_rev_qual:
-        Column.new_column(o_view, REVERSE_SEQUENCE_COLUMN, OBI_SEQ)
-        Column.new_column(o_view, REVERSE_QUALITY_COLUMN, OBI_QUAL, associated_column_name=REVERSE_SEQUENCE_COLUMN, associated_column_version=o_view[REVERSE_SEQUENCE_COLUMN].version)
-
+    if type(output_0) != BufferedWriter: 
+        if not remove_qual:
+            if NUC_SEQUENCE_COLUMN not in o_view:
+                Column.new_column(o_view, NUC_SEQUENCE_COLUMN, OBI_SEQ)
+            Column.new_column(o_view, QUALITY_COLUMN, OBI_QUAL, associated_column_name=NUC_SEQUENCE_COLUMN, associated_column_version=o_view[NUC_SEQUENCE_COLUMN].version)    
+        if not remove_rev_qual:
+            Column.new_column(o_view, REVERSE_SEQUENCE_COLUMN, OBI_SEQ)
+            Column.new_column(o_view, REVERSE_QUALITY_COLUMN, OBI_QUAL, associated_column_name=REVERSE_SEQUENCE_COLUMN, associated_column_version=o_view[REVERSE_SEQUENCE_COLUMN].version)
+        
+    # Initialize multiple elements columns
+    if type(output_0)==BufferedWriter: 
+        dict_cols = {}
+        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] = {}
+                        dict_cols[coln]['eltnames'] = set(v[coln].elements_names)
+                        dict_cols[coln]['nbelts'] = v[coln].nb_elements_per_line
+                        dict_cols[coln]['obitype'] = v[coln].data_type_int
+                    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']), dict_column=True)
+    
     # Initialize the progress bar
-    pb = ProgressBar(total_len, config, seconde=5)
+    if not config['obi']['noprogressbar']:
+        pb = ProgressBar(total_len, config)
+    else:
+        pb = None
     
     i = 0
-    for v in iview_list:
-        for l in v:
+    for v_uri in config["cat"]["views_to_cat"]:
+        v = open_uri(v_uri)[1]
+        for entry in v:
             PyErr_CheckSignals()
-            pb(i)
-            o_view[i] = l
+            if pb is not None:
+                pb(i)
+            if type(output_0)==BufferedWriter: 
+                rep = repr(entry)
+                output_0.write(str2bytes(rep)+b"\n")
+            else:
+                o_view[i] = entry
             i+=1
+        v.close()
 
     # Deletes quality columns if needed
-    if QUALITY_COLUMN in o_view and remove_qual :
-        o_view.delete_column(QUALITY_COLUMN)
-    if REVERSE_QUALITY_COLUMN in o_view and remove_rev_qual :
-        o_view.delete_column(REVERSE_QUALITY_COLUMN)
-
-    pb(i, force=True)
-    print("", file=sys.stderr)
+    if type(output_0)!=BufferedWriter: 
+        if QUALITY_COLUMN in o_view and remove_qual :
+            o_view.delete_column(QUALITY_COLUMN)
+        if REVERSE_QUALITY_COLUMN in o_view and remove_rev_qual :
+            o_view.delete_column(REVERSE_QUALITY_COLUMN)
+    
+    if pb is not None:
+        pb(i, force=True)
+        print("", file=sys.stderr)
     
     # 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)
     #print(repr(view), file=sys.stderr)
 
     for d in idms_list:
-        d.close()
-    o_dms.close()
+        d.close(force=True)
+    o_dms.close(force=True)
     
     logger("info", "Done.")

python/obitools3/commands/clean.pyx

@@ -4,13 +4,14 @@ from obitools3.apps.progress cimport ProgressBar  # @UnresolvedImport
 from obitools3.dms.dms cimport DMS
 from obitools3.dms.view import RollbackException
 from obitools3.dms.capi.obiclean cimport obi_clean
-from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.uri.decode import open_uri
 from obitools3.apps.config import logger
 from obitools3.utils cimport tobytes, str2bytes
 from obitools3.dms.view.view cimport View
 from obitools3.dms.view.typed_view.view_NUC_SEQS cimport View_NUC_SEQS
 
+from io import BufferedWriter
 import sys
 
 
@@ -21,7 +22,8 @@ def addOptions(parser):
 
     addMinimalInputOption(parser)
     addMinimalOutputOption(parser)
- 
+    addNoProgressBarOption(parser)
+    
     group = parser.add_argument_group('obi clean specific options')
 
     group.add_argument('--distance', '-d',
@@ -36,8 +38,7 @@ def addOptions(parser):
                        dest="clean:sample-tag-name",
                        metavar="<SAMPLE TAG NAME>",
                        type=str,
-                       default="merged_sample",
-                       help="Name of the tag where sample counts are kept.")
+                       help="Name of the tag where merged sample count informations are kept (typically generated by obi uniq, usually MERGED_sample, default: None).")
     
     group.add_argument('--ratio', '-r',
                        action="store", dest="clean:ratio",
@@ -89,17 +90,20 @@ def run(config):
     if output is None:
         raise Exception("Could not create output")
     o_dms = output[0]
+    output_0 = output[0]
     final_o_view_name = output[1]
     
-    # If the input and output DMS are not the same, run obiclean creating a temporary view that will be exported to 
+    # If stdout output or the input and output DMS are not the same, create a temporary view that will be exported to 
     # the right DMS and deleted in the other afterwards.
-    if i_dms != o_dms:
-        temporary_view_name = final_o_view_name
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        temporary_view_name = b"temp"
         i=0
         while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
-            temporary_view_name = final_o_view_name+b"_"+str2bytes(str(i))
+            temporary_view_name = temporary_view_name+b"_"+str2bytes(str(i))
             i+=1
         o_view_name = temporary_view_name
+        if type(output_0)==BufferedWriter:
+            o_dms = i_dms
     else:
         o_view_name = final_o_view_name
         
@@ -107,6 +111,9 @@ def run(config):
     command_line = " ".join(sys.argv[1:])
     comments = View.print_config(config, "clean", command_line, input_dms_name=[i_dms_name], input_view_name=[i_view_name])
 
+    if 'sample-tag-name' not in config['clean']:
+        config['clean']['sample-tag-name'] = ""
+        
     if obi_clean(i_dms.name_with_full_path, tobytes(i_view_name), tobytes(config['clean']['sample-tag-name']), tobytes(o_view_name), comments, \
               config['clean']['distance'], config['clean']['ratio'], config['clean']['heads-only'], config['clean']['thread-count']) < 0:
         raise Exception("Error running obiclean")
@@ -114,18 +121,25 @@ def run(config):
     # If the input and output DMS are not the same, export result view to output DMS
     if i_dms != o_dms:
         View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, final_o_view_name)
-    
+
+    # stdout output: write to buffer
+    if type(output_0)==BufferedWriter:
+        logger("info", "Printing to output...")
+        o_view = o_dms[o_view_name]
+        o_view.print_to_output(output_0, noprogressbar=config['obi']['noprogressbar'])
+        o_view.close()
+
     # Save command config in DMS comments
     o_dms.record_command_line(command_line)
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_dms[final_o_view_name]), file=sys.stderr)
 
-    # If the input and the output DMS are different, delete the temporary result view in the input DMS
-    if i_dms != o_dms:
+    # 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(i_dms, o_view_name)
-        o_dms.close()
+        o_dms.close(force=True)
     
-    i_dms.close()
+    i_dms.close(force=True)
 
     logger("info", "Done.")

python/obitools3/commands/count.pyx

@@ -22,7 +22,7 @@ def addOptions(parser):
     group.add_argument('-s','--sequence',
                         action="store_true", dest="count:sequence",
                         default=False,
-                        help="Prints only the number of sequence records.")
+                        help="Prints only the number of sequence records (much faster, default: False).")
  
     group.add_argument('-a','--all',
                         action="store_true", dest="count:all",
@@ -56,3 +56,5 @@ def run(config):
         print(count2)
     else:
         print(count1)
+    
+    input[0].close(force=True)

python/obitools3/commands/ecopcr.pyx

@@ -5,10 +5,10 @@ from obitools3.dms.dms cimport DMS
 from obitools3.dms.capi.obidms cimport OBIDMS_p
 from obitools3.dms.view import RollbackException
 from obitools3.dms.capi.obiecopcr cimport obi_ecopcr
-from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addTaxonomyOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addTaxonomyOption, addNoProgressBarOption
 from obitools3.uri.decode import open_uri
 from obitools3.apps.config import logger
-from obitools3.utils cimport tobytes
+from obitools3.utils cimport tobytes, str2bytes
 from obitools3.dms.view.typed_view.view_NUC_SEQS cimport View_NUC_SEQS
 from obitools3.dms.view import View
 
@@ -16,6 +16,7 @@ from libc.stdlib  cimport malloc, free
 from libc.stdint  cimport int32_t
 
 import sys
+from io import BufferedWriter
 
 
 __title__="in silico PCR"
@@ -27,6 +28,7 @@ def addOptions(parser):
     addMinimalInputOption(parser)
     addTaxonomyOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
 
  
     group = parser.add_argument_group('obi ecopcr specific options')
@@ -35,13 +37,15 @@ def addOptions(parser):
                        action="store", dest="ecopcr:primer1",
                        metavar='<PRIMER>',
                        type=str,
-                       help="Forward primer.")
+                       required=True,
+                       help="Forward primer, length must be less than or equal to 32")
 
     group.add_argument('--primer2', '-R',
                        action="store", dest="ecopcr:primer2",
                        metavar='<PRIMER>',
                        type=str,
-                       help="Reverse primer.")
+                       required=True,
+                       help="Reverse primer, length must be less than or equal to 32")
 
     group.add_argument('--error', '-e',
                        action="store", dest="ecopcr:error",
@@ -167,11 +171,20 @@ def run(config):
     if output is None:
         raise Exception("Could not create output")
     o_dms = output[0]
+    output_0 = output[0]
     o_dms_name = output[0].name
     o_view_name = output[1]
 
     # Read taxonomy name    
     taxonomy_name = config['obi']['taxoURI'].split("/")[-1]   # Robust in theory
+
+    # If stdout output create a temporary view in the input dms that will be deleted afterwards.
+    if type(output_0)==BufferedWriter:
+        o_dms = i_dms
+        o_view_name = b"temp"
+        while o_view_name in i_dms:   # Making sure view name is unique in input DMS
+            o_view_name = o_view_name+b"_"+str2bytes(str(i))
+            i+=1
     
     # Save command config in View comments
     command_line = " ".join(sys.argv[1:])
@@ -199,10 +212,22 @@ def run(config):
 
     free(restrict_to_taxids_p)
     free(ignore_taxids_p)
-    
+
+    # stdout output: write to buffer
+    if type(output_0)==BufferedWriter:
+        logger("info", "Printing to output...")
+        o_view = o_dms[o_view_name]
+        o_view.print_to_output(output_0, noprogressbar=config['obi']['noprogressbar'])
+        o_view.close()
+
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_dms[o_view_name]), file=sys.stderr)
 
-    o_dms.close()
+    # If stdout output, delete the temporary result view in the input DMS
+    if type(output_0)==BufferedWriter:
+        View.delete_view(i_dms, o_view_name)
+
+    i_dms.close(force=True)
+    o_dms.close(force=True)
 
     logger("info", "Done.")

python/obitools3/commands/ecotag.pyx

@@ -4,7 +4,7 @@ from obitools3.apps.progress cimport ProgressBar  # @UnresolvedImport
 from obitools3.dms.dms cimport DMS
 from obitools3.dms.view import RollbackException
 from obitools3.dms.capi.obiecotag cimport obi_ecotag
-from obitools3.apps.optiongroups import addMinimalInputOption, addTaxonomyOption, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addTaxonomyOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.uri.decode import open_uri
 from obitools3.apps.config import logger
 from obitools3.utils cimport tobytes, str2bytes
@@ -12,6 +12,7 @@ from obitools3.dms.view.view cimport View
 from obitools3.dms.view.typed_view.view_NUC_SEQS cimport View_NUC_SEQS
 
 import sys
+from io import BufferedWriter
 
 
 __title__="Taxonomic assignment of sequences"
@@ -22,6 +23,7 @@ def addOptions(parser):
     addMinimalInputOption(parser)
     addTaxonomyOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
  
     group = parser.add_argument_group('obi ecotag specific options')
 
@@ -39,6 +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()
@@ -64,10 +77,9 @@ 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(i_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).", 
-              config['ecotag']['threshold'], i_dms[ref_view_name].comments["ref_db_threshold"])
-
+    if config['ecotag']['bubble_threshold'] < eval(ref_dms[ref_view_name].comments["ref_db_threshold"]) :
+        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'])}).")
+    
     # Open the output: only the DMS
     output = open_uri(config['obi']['outputURI'],
                       input=False,
@@ -75,17 +87,19 @@ def run(config):
     if output is None:
         raise Exception("Could not create output")
     o_dms = output[0]
+    output_0 = output[0]
     final_o_view_name = output[1]
     
-    # If the input and output DMS are not the same, run ecotag creating a temporary view that will be exported to 
-    # the right DMS and deleted in the other afterwards.
-    if i_dms != o_dms:
-        temporary_view_name = final_o_view_name
+    # If stdout output or the input and output DMS are not the same, create a temporary view that will be exported and deleted.
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        temporary_view_name = b"temp"
         i=0
         while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
-            temporary_view_name = final_o_view_name+b"_"+str2bytes(str(i))
+            temporary_view_name = temporary_view_name+b"_"+str2bytes(str(i))
             i+=1
         o_view_name = temporary_view_name
+        if type(output_0)==BufferedWriter:
+            o_dms = i_dms
     else:
         o_view_name = final_o_view_name
 
@@ -107,10 +121,11 @@ def run(config):
     comments = View.print_config(config, "ecotag", command_line, input_dms_name=input_dms_name, input_view_name=input_view_name)
 
     if obi_ecotag(i_dms.name_with_full_path, tobytes(i_view_name), \
-                  tobytes(ref_dms_name), tobytes(ref_view_name), \
-                  tobytes(taxo_dms_name), tobytes(taxonomy_name), \
-                  tobytes(o_view_name), comments, 
-                  config['ecotag']['threshold']) < 0:
+                  ref_dms.name_with_full_path, tobytes(ref_view_name), \
+                  taxo_dms.name_with_full_path, tobytes(taxonomy_name), \
+                  tobytes(o_view_name), comments, \
+                  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
@@ -120,15 +135,24 @@ def run(config):
     # Save command config in DMS comments
     o_dms.record_command_line(command_line)
 
+    # stdout output: write to buffer
+    if type(output_0)==BufferedWriter:
+        logger("info", "Printing to output...")
+        o_view = o_dms[o_view_name]
+        o_view.print_to_output(output_0, noprogressbar=config['obi']['noprogressbar'])
+        o_view.close()
+
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_dms[final_o_view_name]), file=sys.stderr)
 
-    # If the input and the output DMS are different, delete the temporary result view in the input DMS
-    if i_dms != o_dms:
+    # 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(i_dms, o_view_name)
-        o_dms.close()
+        o_dms.close(force=True)
     
-    i_dms.close()
+    taxo_dms.close(force=True)
+    ref_dms.close(force=True)
+    i_dms.close(force=True)
     
     logger("info", "Done.")
 

python/obitools3/commands/export.pyx

@@ -59,13 +59,23 @@ def run(config):
      # Check that the input view has the type NUC_SEQS if needed    # TODO discuss, maybe bool property
     if (output[2] == Nuc_Seq) and (iview.type != b"NUC_SEQS_VIEW") :  # Nuc_Seq_Stored? TODO
         raise Exception("Error: the view to export in fasta or fastq format is not a NUC_SEQS view")
-   
+    
+    if config['obi']['only'] is not None:
+        withoutskip = min(input[4], config['obi']['only'])
+    else:
+        withoutskip = input[4]
+     
+    if config['obi']['skip'] is not None:
+        skip = min(input[4], config['obi']['skip'])
+    else:
+        skip = 0
+    
     # Initialize the progress bar
     if config['obi']['noprogressbar']:
         pb = None
     else:
-        pb = ProgressBar(len(iview), config, seconde=5)
-    
+        pb = ProgressBar(withoutskip - skip, config)
+
     i=0
     for seq in iview :
         PyErr_CheckSignals()
@@ -79,14 +89,14 @@ def run(config):
 
     if pb is not None:
         pb(i, force=True)
-    print("", file=sys.stderr)
+        print("", file=sys.stderr)
 
     # TODO save command in input dms?
     
     if not BrokenPipeError and not IOError:
         output_object.close()
     iview.close()
-    input[0].close()
+    input[0].close(force=True)
 
     logger("info", "Done.")
 

python/obitools3/commands/grep.pyx

@@ -4,7 +4,7 @@ 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
+from obitools3.apps.optiongroups import addMinimalInputOption, addTaxonomyOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.dms.view import RollbackException
 from obitools3.apps.config import logger
 from obitools3.utils cimport tobytes, str2bytes
@@ -14,6 +14,7 @@ import time
 import re
 import sys
 import ast
+from io import BufferedWriter
 from cpython.exc cimport PyErr_CheckSignals
 
  
@@ -28,6 +29,7 @@ def addOptions(parser):
     addMinimalInputOption(parser)
     addTaxonomyOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
 
     group=parser.add_argument_group("obi grep specific options")
  
@@ -161,8 +163,7 @@ def obi_eval(compiled_expr, loc_env, line):
     return obi_eval_result
     
 
-def Filter_generator(options, tax_filter):
-    #taxfilter = taxonomyFilterGenerator(options)
+def Filter_generator(options, tax_filter, i_view):
 
     # Initialize conditions
     predicates = None
@@ -171,6 +172,9 @@ def Filter_generator(options, tax_filter):
     attributes = None
     if "attributes" in options and len(options["attributes"]) > 0:
         attributes = options["attributes"]
+        for attribute in attributes:
+            if attribute not in i_view:
+                return None
     lmax = None
     if "lmax" in options:
         lmax = options["lmax"]
@@ -180,7 +184,7 @@ def Filter_generator(options, tax_filter):
     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
@@ -196,6 +200,8 @@ def Filter_generator(options, tax_filter):
     if "attribute_patterns" in options and len(options["attribute_patterns"]) > 0:
         for p in options["attribute_patterns"]:
             attribute, pattern = p.split(":", 1)
+            if attribute not in i_view:
+                return None
             attribute_patterns[tobytes(attribute)] = re.compile(tobytes(pattern))
     
     def filter(line, loc_env):
@@ -300,16 +306,21 @@ def run(config):
     if output is None:
         raise Exception("Could not create output view")
     o_dms = output[0]
-    o_view_name_final = output[1]
-    o_view_name = o_view_name_final
-    
-    # If the input and output DMS are not the same, create output view in input DMS first, then export it
-    # to output DMS, making sure the temporary view name is unique in the input DMS 
-    if i_dms != o_dms:
+    output_0 = output[0]
+    final_o_view_name = output[1]
+
+    # If stdout output or the input and output DMS are not the same, create a temporary view that will be exported and deleted afterwards.
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        temporary_view_name = b"temp"
         i=0
-        while o_view_name in i_dms:
-            o_view_name = o_view_name_final+b"_"+str2bytes(str(i))
-            i+=1        
+        while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
+            temporary_view_name = temporary_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        o_view_name = temporary_view_name
+        if type(output_0)==BufferedWriter:
+            o_dms = i_dms
+    else:
+        o_view_name = final_o_view_name
 
     if 'taxoURI' in config['obi'] and config['obi']['taxoURI'] is not None:
         taxo_uri = open_uri(config["obi"]["taxoURI"])
@@ -320,33 +331,49 @@ def run(config):
         taxo = None
 
     # Initialize the progress bar
-    pb = ProgressBar(len(i_view), config, seconde=5)
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(len(i_view), config)
+    else:
+        pb = None
  
     # Apply filter
     tax_filter = Taxonomy_filter_generator(taxo, config["grep"])
-    filter = Filter_generator(config["grep"], tax_filter)
+    filter = Filter_generator(config["grep"], tax_filter, i_view)        
     selection = Line_selection(i_view)
-    for i in range(len(i_view)):
-        PyErr_CheckSignals()
-        pb(i)
-        line = i_view[i]
-                 
-        loc_env = {"sequence": line, "line": line, "taxonomy": taxo, "obi_eval_result": False}
-        
-        good = filter(line, loc_env)
- 
-        if good :
+    
+    if filter is None and config["grep"]["invert_selection"]: # all sequences are selected: filter is None if no line will be selected because some columns don't exist
+        for i in range(len(i_view)):
+            PyErr_CheckSignals()
+            if pb is not None:
+                pb(i)
             selection.append(i)
-
-    pb(i, force=True)
-    print("", file=sys.stderr)
+    
+    elif filter is not None :   # filter is None if no line will be selected because some columns don't exist
+        for i in range(len(i_view)):
+            PyErr_CheckSignals()
+            if pb is not None:
+                pb(i)
+            line = i_view[i]
+            
+            loc_env = {"sequence": line, "line": line, "taxonomy": taxo, "obi_eval_result": False}
+            
+            good = filter(line, loc_env)
+     
+            if good :
+                selection.append(i)
+   
+    if pb is not None:
+        pb(len(i_view), force=True)
+        print("", file=sys.stderr)
 
     # Create output view with the line selection
     try:
         o_view = selection.materialize(o_view_name)
     except Exception, e:
         raise RollbackException("obi grep error, rollbacking view: "+str(e), o_view)
-    
+
+    logger("info", "Grepped %d entries" % len(o_view))
+
     # Save command config in View and DMS comments
     command_line = " ".join(sys.argv[1:])
     input_dms_name=[input[0].name]
@@ -361,16 +388,22 @@ def run(config):
     # and delete the temporary view in the input DMS
     if i_dms != o_dms:
         o_view.close()
-        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, o_view_name_final)
-        o_view = o_dms[o_view_name_final]
+        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, final_o_view_name)
+        o_view = o_dms[final_o_view_name]
+    
+    # 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()
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_view), file=sys.stderr)
 
-    # If the input and the output DMS are different, delete the temporary imported view used to create the final view
-    if i_dms != o_dms:
+    # 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(i_dms, o_view_name)
-        o_dms.close()
-    i_dms.close()
+        o_dms.close(force=True)
+    i_dms.close(force=True)
 
     logger("info", "Done.")

python/obitools3/commands/head.pyx

@@ -4,14 +4,15 @@ 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, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.dms.view import RollbackException
 from obitools3.apps.config import logger
 from obitools3.utils cimport str2bytes
 
 import time
 import sys
- 
+from io import BufferedWriter
+
 from cpython.exc cimport PyErr_CheckSignals
 
 
@@ -22,6 +23,7 @@ def addOptions(parser):
     
     addMinimalInputOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
 
     group=parser.add_argument_group('obi head specific options')
 
@@ -53,31 +55,41 @@ def run(config):
     if output is None:
         raise Exception("Could not create output view")
     o_dms = output[0]
-    o_view_name_final = output[1]
-    o_view_name = o_view_name_final
+    output_0 = output[0]
+    final_o_view_name = output[1]
     
-    # If the input and output DMS are not the same, create output view in input DMS first, then export it
-    # to output DMS, making sure the temporary view name is unique in the input DMS 
-    if i_dms != o_dms:
+    # If stdout output or the input and output DMS are not the same, create a temporary view that will be exported and deleted.
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        temporary_view_name = b"temp"
         i=0
-        while o_view_name in i_dms:
-            o_view_name = o_view_name_final+b"_"+str2bytes(str(i))
-            i+=1        
+        while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
+            temporary_view_name = temporary_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        o_view_name = temporary_view_name
+        if type(output_0)==BufferedWriter:
+            o_dms = i_dms
+    else:
+        o_view_name = final_o_view_name
     
     n = min(config['head']['count'], len(i_view))
 
     # Initialize the progress bar
-    pb = ProgressBar(n, config, seconde=5)
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(n, config)
+    else:
+        pb = None
         
     selection = Line_selection(i_view)
     
     for i in range(n):
         PyErr_CheckSignals()
-        pb(i)
+        if pb is not None:
+            pb(i)
         selection.append(i)
 
-    pb(i, force=True)
-    print("", file=sys.stderr)
+    if pb is not None:
+        pb(i, force=True)
+        print("", file=sys.stderr)
 
     # Create output view with the line selection
     try:
@@ -94,16 +106,22 @@ def run(config):
     # and delete the temporary view in the input DMS
     if i_dms != o_dms:
         o_view.close()
-        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, o_view_name_final)
-        o_view = o_dms[o_view_name_final]
+        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, final_o_view_name)
+        o_view = o_dms[final_o_view_name]
+
+    # 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()
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(view), file=sys.stderr)
 
-    # If the input and the output DMS are different, delete the temporary imported view used to create the final view
-    if i_dms != o_dms:
+    # 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(i_dms, o_view_name)
-        o_dms.close()
-    i_dms.close()
+        o_dms.close(force=True)
+    i_dms.close(force=True)
     
     logger("info", "Done.")

python/obitools3/commands/history.pyx

@@ -54,4 +54,5 @@ def run(config):
             print(bytes2str(entries.ascii_history))
         else:
             raise Exception("ASCII history only available for views")
-            
+    
+    input[0].close(force=True)

python/obitools3/commands/import.pyx

@@ -25,13 +25,16 @@ from obitools3.dms.capi.obiview cimport VIEW_TYPE_NUC_SEQS, \
                                         DEFINITION_COLUMN, \
                                         QUALITY_COLUMN, \
                                         COUNT_COLUMN, \
-                                        TAXID_COLUMN
+                                        TAXID_COLUMN, \
+                                        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
 
@@ -46,6 +49,8 @@ from obitools3.apps.config import logger
 
 from cpython.exc cimport PyErr_CheckSignals
 
+from io import BufferedWriter
+
 
 __title__="Imports sequences from different formats into a DMS"
  
@@ -72,8 +77,13 @@ def addOptions(parser):
                      action="store_true", dest="import:preread",
                      default=False,
                      help="Do a first readthrough of the dataset if it contains huge dictionaries (more than 100 keys) for "
-                          "a much faster import.")
+                          "a much faster import. This option is not recommended and will slow down the import in any other case.")
 
+    group.add_argument('--space-priority',
+                     action="store_true", dest="import:space_priority",
+                     default=False,
+                     help="If importing a view into another DMS, do it by importing each line, saving disk space if the original view "
+                          "has a line selection associated.")
 
 def run(config):
     
@@ -86,6 +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
@@ -96,6 +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
@@ -129,7 +142,7 @@ def run(config):
         if entry_count > 0:
             logger("info", "Importing %d entries", entry_count)
         else:
-            logger("info", "Importing an unknow number of entries")
+            logger("info", "Importing an unknown number of entries")
         
         # TODO a bit dirty?
         if input[2]==Nuc_Seq or input[2]==View_NUC_SEQS:
@@ -139,7 +152,7 @@ def run(config):
     else:
         v = None
     
-    if config['obi']['taxdump'] or isinstance(input[1], View): 
+    if config['obi']['taxdump'] or (isinstance(input[1], View) and not config['import']['space_priority']): 
         dms_only=True
     else:
         dms_only=False
@@ -163,21 +176,24 @@ def run(config):
         taxo.write(taxo_name)
         taxo.close()
         o_dms.record_command_line(" ".join(sys.argv[1:]))
-        o_dms.close()
+        o_dms.close(force=True)
         logger("info", "Done.")
         return
 
-    # If importing a view between two DMS, use C API
-    if isinstance(input[1], View):
+    # If importing a view between two DMS and not wanting to save space if line selection in original view, use C API
+    if isinstance(input[1], View) and not config['import']['space_priority']:
         if obi_import_view(input[0].name_with_full_path, o_dms.name_with_full_path, input[1].name, tobytes((config['obi']['outputURI'].split('/'))[-1])) < 0 :
+            input[0].close(force=True)
+            output[0].close(force=True)
             raise Exception("Error importing a view in a DMS")
         o_dms.record_command_line(" ".join(sys.argv[1:]))
-        o_dms.close()
+        input[0].close(force=True)
+        output[0].close(force=True)
         logger("info", "Done.")
         return
 
     if entry_count >= 0:
-        pb = ProgressBar(entry_count, config, seconde=5)
+        pb = ProgressBar(entry_count, config)
         
     NUC_SEQS_view = False
     if isinstance(output[1], View) :
@@ -192,9 +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...")
@@ -217,11 +240,14 @@ def run(config):
                 logger("info", "Read %d entries", i)
     
             for tag in entry :
+                newtag = tag
+                if tag[:7] == b"merged_":
+                    newtag = MERGED_PREFIX+tag[7:]
                 if type(entry[tag]) == dict :
                     if tag in dict_dict:
-                        dict_dict[tag][0].update(entry[tag].keys())
+                        dict_dict[newtag][0].update(entry[tag].keys())
                     else:
-                        dict_dict[tag] = [set(entry[tag].keys()), get_obitype(entry[tag])]
+                        dict_dict[newtag] = [set(entry[tag].keys()), get_obitype(entry[tag])]
             i+=1
         
         if pb is not None:
@@ -231,15 +257,16 @@ 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])), \
-                          value_obitype)
+                              elements_names=list(dict_dict[tag][0]), \
+                              dict_column=True), \
+                          dict_dict[tag][1])
     
         
         # Reinitialize the input
         if isinstance(input[0], CompressedFile):
             input_is_file = True
         if entry_count >= 0:
-            pb = ProgressBar(entry_count, config, seconde=5)
+            pb = ProgressBar(entry_count, config)
         try:
             input[0].close()
         except AttributeError:
@@ -256,7 +283,6 @@ def run(config):
 
         if entry is None:  # error or exception handled at lower level, not raised because Python generators can't resume after any exception is raised
             if config['obi']['skiperror']:
-                i-=1
                 continue
             else:
                 raise RollbackException("obi import error, rollbacking view", view)
@@ -265,123 +291,140 @@ def run(config):
             pb(i)
         elif not i%50000:
             logger("info", "Imported %d entries", i)
-              
-        if NUC_SEQS_view: 
-            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
-            if i == 0:
-                get_quality = QUALITY_COLUMN in entry
+        
+        try:
+             
+            if NUC_SEQS_view: 
+                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
+                if i == 0:
+                    get_quality = QUALITY_COLUMN in entry
+                    if get_quality:
+                        Column.new_column(view, QUALITY_COLUMN, OBI_QUAL)
+                        qual_col = view[QUALITY_COLUMN]
                 if get_quality:
-                    Column.new_column(view, QUALITY_COLUMN, OBI_QUAL)
-                    qual_col = view[QUALITY_COLUMN]
-            if get_quality:
-                qual_col[i] = entry.quality
-         
-        for tag in entry :
-            
-            if tag != ID_COLUMN and tag != DEFINITION_COLUMN and tag != NUC_SEQUENCE_COLUMN and tag != QUALITY_COLUMN :  # TODO dirty 
-                                
-                value = entry[tag]
-                if tag == b"taxid":
-                    tag = TAXID_COLUMN
-                if tag == b"count":
-                    tag = COUNT_COLUMN
+                    qual_col[i] = entry.quality
                     
-                if tag not in dcols :
-                     
-                    value_type = type(value)
-                    nb_elts = 1
-                    value_obitype = OBI_VOID
-                     
-                    if value_type == dict or value_type == list :
-                        nb_elts = len(value)
-                        elt_names = list(value)
-                    else :
-                        nb_elts = 1
-                        elt_names = None
-                     
-                    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)
-                                                 
-                        # 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?
- 
-                else :
-         
-                    rewrite = False
- 
-                    # Check type adequation
-                    old_type = dcols[tag][1]
-                    new_type = OBI_VOID
-                    new_type = update_obitype(old_type, value)
-                    if old_type != new_type :
-                        rewrite = True
- 
-                    try:
-                        # Check that it's not the case where the first entry contained a dict of length 1 and now there is a new key                        
-                        if type(value) == dict and \
-                            dcols[tag][0].nb_elements_per_line == 1 and len(value.keys()) == 1 \
-                            and dcols[tag][0].elements_names[0] != list(value.keys())[0] :
-                            raise IndexError  # trigger column rewrite
+                # 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 :
+                
+                if tag != ID_COLUMN and tag != DEFINITION_COLUMN and tag != NUC_SEQUENCE_COLUMN and tag != QUALITY_COLUMN :  # TODO dirty 
+                                    
+                    value = entry[tag]
+                    if tag == b"taxid":
+                        tag = TAXID_COLUMN
+                    if tag == b"count":
+                        tag = COUNT_COLUMN
+                    if tag[:7] == b"merged_":
+                        tag = MERGED_PREFIX+tag[7:]
                         
-                        # Fill value
-                        dcols[tag][0][i] = value
-                     
-                    except IndexError :
-                                                 
+                    if tag not in dcols :
+                         
                         value_type = type(value)
-                        old_column = dcols[tag][0]
-                        old_nb_elements_per_line = old_column.nb_elements_per_line
-                        new_nb_elements_per_line = 0
-                        old_elements_names = old_column.elements_names
-                        new_elements_names = None
-     
-                        #####################################################################
+                        nb_elts = 1
+                        value_obitype = OBI_VOID
+                        dict_col = False
                          
-                        # Check the length and keys of column lines if needed
-                        if value_type == dict :    # Check dictionary keys
-                            for k in value :
-                                if k not in old_elements_names :
-                                    new_elements_names = list(set(old_elements_names+[tobytes(k) for k in value]))
-                                    rewrite = True
-                                    break
+                        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
                          
-                        elif value_type == list or value_type == tuple :  # Check vector length
-                            if old_nb_elements_per_line < len(value) :
-                                new_nb_elements_per_line = len(value)
-                                rewrite = True
+                        value_obitype = get_obitype(value)
                          
-                        #####################################################################
-                         
-                        if rewrite :
-                            if new_nb_elements_per_line == 0 and new_elements_names is not None :
-                                new_nb_elements_per_line = len(new_elements_names)
-                             
-                            # Reset obierrno 
-                            obi_errno = 0
- 
-                            dcols[tag] = (view.rewrite_column_with_diff_attributes(old_column.name, 
-                                                                                   new_data_type=new_type, 
-                                                                                   new_nb_elements_per_line=new_nb_elements_per_line,
-                                                                                   new_elements_names=new_elements_names,
-                                                                                   rewrite_last_line=False), 
-                                          new_type)
-                             
-                            # Update the dictionary:
-                            for t in dcols :
-                                dcols[t] = (view[t], dcols[t][1])
-                             
+                        if value_obitype != OBI_VOID :
+                            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
                             dcols[tag][0][i] = value
-                                    
+                         
+                        # TODO else log error?
+     
+                    else :
+             
+                        rewrite = False
+     
+                        # Check type adequation
+                        old_type = dcols[tag][1]
+                        new_type = OBI_VOID
+                        new_type = update_obitype(old_type, value)
+                        if old_type != new_type :
+                            rewrite = True
+     
+                        try:
+                            # Check that it's not the case where the first entry contained a dict of length 1 and now there is a new key                        
+                            if type(value) == dict and \
+                                dcols[tag][0].nb_elements_per_line == 1 \
+                                and set(dcols[tag][0].elements_names) != set(value.keys()) :
+                                raise IndexError  # trigger column rewrite
+                            
+                            # Fill value
+                            dcols[tag][0][i] = value
+                         
+                        except IndexError :
+                                                     
+                            value_type = type(value)
+                            old_column = dcols[tag][0]
+                            old_nb_elements_per_line = old_column.nb_elements_per_line
+                            new_nb_elements_per_line = 0
+                            old_elements_names = old_column.elements_names
+                            new_elements_names = None
+         
+                            #####################################################################
+                             
+                            # Check the length and keys of column lines if needed
+                            if value_type == dict :    # Check dictionary keys
+                                for k in value :
+                                    if k not in old_elements_names :
+                                        new_elements_names = list(set(old_elements_names+[tobytes(k) for k in value]))
+                                        rewrite = True
+                                        break
+                             
+                            elif value_type == list or value_type == tuple :  # Check vector length
+                                if old_nb_elements_per_line < len(value) :
+                                    new_nb_elements_per_line = len(value)
+                                    rewrite = True
+                             
+                            #####################################################################
+                             
+                            if rewrite :
+                                if new_nb_elements_per_line == 0 and new_elements_names is not None :
+                                    new_nb_elements_per_line = len(new_elements_names)
+                                 
+                                # Reset obierrno 
+                                obi_errno = 0
+     
+                                dcols[tag] = (view.rewrite_column_with_diff_attributes(old_column.name, 
+                                                                                       new_data_type=new_type, 
+                                                                                       new_nb_elements_per_line=new_nb_elements_per_line,
+                                                                                       new_elements_names=new_elements_names,
+                                                                                       rewrite_last_line=False), 
+                                              new_type)
+                                 
+                                # Update the dictionary:
+                                for t in dcols :
+                                    dcols[t] = (view[t], dcols[t][1])
+                                 
+                                # Fill value
+                                dcols[tag][0][i] = value
+        
+        except Exception as e:
+            print("\nCould not import sequence id:", entry.id, "(error raised:", e, ")")
+            if 'skiperror' in config['obi'] and not config['obi']['skiperror']:
+                raise e
+            else:
+                pass
+        
         i+=1
  
     if pb is not None:
@@ -402,7 +445,7 @@ def run(config):
     except AttributeError:
         pass
     try:
-        output[0].close()
+        output[0].close(force=True)
     except AttributeError:
         pass
 

python/obitools3/commands/less.pyx

@@ -46,5 +46,5 @@ def run(config):
     process.wait()
 
     iview.close()
-    input[0].close()
+    input[0].close(force=True)
 

python/obitools3/commands/ls.pyx

@@ -31,24 +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]
-        l = []
-        for view in input[0]:
-            l.append(tostr(view) + "\t(Date created: " + str(bytes2str_object(dms[view].comments["Date created"]))+")")
-        l.sort()
-        for v in l:
-            print(v)
+    
+    # Print representation
+    if config['ls']['longformat']:
+        print(input[1].repr_longformat())
     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

@@ -2,10 +2,10 @@
 
 from obitools3.apps.progress cimport ProgressBar  # @UnresolvedImport
 from obitools3.dms import DMS
-from obitools3.dms.view import RollbackException
+from obitools3.dms.view import RollbackException, View
 from obitools3.dms.view.typed_view.view_NUC_SEQS cimport View_NUC_SEQS
 from obitools3.dms.column.column cimport Column, Column_line
-from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.uri.decode import open_uri
 from obitools3.apps.config import logger
 from obitools3.libalign._freeendgapfm import FreeEndGapFullMatch
@@ -14,17 +14,14 @@ from obitools3.dms.obiseq cimport Nuc_Seq
 from obitools3.dms.capi.obitypes cimport OBI_SEQ, OBI_QUAL
 from obitools3.dms.capi.apat cimport MAX_PATTERN
 from obitools3.dms.capi.obiview cimport REVERSE_SEQUENCE_COLUMN, REVERSE_QUALITY_COLUMN
-from obitools3.utils cimport tobytes
+from obitools3.utils cimport tobytes, str2bytes
 
 from libc.stdint cimport INT32_MAX
 from functools import reduce
 import math
 import sys
 from cpython.exc cimport PyErr_CheckSignals
-
-
-#REVERSE_SEQ_COLUMN_NAME = b"REVERSE_SEQUENCE"      # used by alignpairedend tool
-#REVERSE_QUALITY_COLUMN_NAME = b"REVERSE_QUALITY"   # used by alignpairedend tool
+from io import BufferedWriter
 
 
 __title__="Assigns sequence records to the corresponding experiment/sample based on DNA tags and primers"
@@ -34,7 +31,8 @@ def addOptions(parser):
 
     addMinimalInputOption(parser)
     addMinimalOutputOption(parser)
- 
+    addNoProgressBarOption(parser)
+    
     group = parser.add_argument_group('obi ngsfilter specific options')
 
     group.add_argument('-t','--info-view',
@@ -42,7 +40,9 @@ def addOptions(parser):
                      metavar="<URI>",
                      type=str,
                      default=None,
-                     help="URI to the view containing the samples definition (with tags, primers, sample names,...)")
+                     required=True,
+                     help="URI to the view containing the samples definition (with tags, primers, sample names,...).\n"
+                          "\nWarning: primer lengths must be less than or equal to 32")
 
     group.add_argument('-R', '--reverse-reads',
                      action="store", dest="ngsfilter:reverse",
@@ -56,7 +56,7 @@ def addOptions(parser):
                      metavar="<URI>",
                      type=str,
                      default=None,
-                     help="URI to the view used to store the sequences unassigned to any sample")
+                     help="URI to the view used to store the sequences unassigned to any sample. Those sequences are untrimmed.")
 
     group.add_argument('--no-tags',
                      action="store_true", dest="ngsfilter:notags",
@@ -172,6 +172,13 @@ cdef read_info_view(info_view, max_errors=2, verbose=False, not_aligned=False):
     primer_list = []
     i=0
     for p in info_view:
+        
+        # Check primer length: should not be longer than 32, the max allowed by the apat lib
+        if len(p[b'forward_primer']) > 32:
+            raise RollbackException("Error: primers can not be longer than 32bp, rollbacking views")
+        if len(p[b'reverse_primer']) > 32:
+            raise RollbackException("Error: primers can not be longer than 32bp, rollbacking views")
+        
         forward=Primer(p[b'forward_primer'],
                        len(p[b'forward_tag']) if (b'forward_tag' in p and p[b'forward_tag']!=None) else None,
                        True,
@@ -470,6 +477,8 @@ cdef tuple annotate(sequences, infos, no_tags, verbose=False):
     if not directmatch[0].forward:
         sequences[0] = sequences[0].reverse_complement
         sequences[0][b'reversed'] = True   # used by the alignpairedend tool (in kmer_similarity.c)
+    else:
+        sequences[0][b'reversed'] = False   # used by the alignpairedend tool (in kmer_similarity.c)
 
     sample=None
     if not no_tags:
@@ -497,7 +506,7 @@ cdef tuple annotate(sequences, infos, no_tags, verbose=False):
                     sample=None
         
         if sample is None:
-            sequences[0][b'error']=b"No tags found"
+            sequences[0][b'error']=b"No sample with that tag combination"
             return False, sequences[0]
     
         sequences[0].update(sample)
@@ -528,7 +537,8 @@ def run(config):
         raise Exception("Could not open input reads")
     if input[2] != View_NUC_SEQS:
         raise NotImplementedError('obi ngsfilter only works on NUC_SEQS views')
-
+    i_dms = input[0]
+    
     if "reverse" in config["ngsfilter"]:
         
         forward = input[1]        
@@ -569,8 +579,19 @@ def run(config):
     if output is None:
         raise Exception("Could not create output view")
     
+    o_dms = output[0]
+    output_0 = output[0]
     o_view = output[1]
     
+    # If stdout output, create a temporary view in the input dms that will be deleted afterwards.
+    if type(output_0)==BufferedWriter:
+        o_dms = i_dms
+        o_view_name = b"temp"
+        while o_view_name in i_dms:   # Making sure view name is unique in input DMS
+            o_view_name = o_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        o_view = View_NUC_SEQS.new(i_dms, o_view_name)
+
     # Open the view containing the informations about the tags and the primers
     info_input = open_uri(config['ngsfilter']['info_view'])
     if info_input is None:
@@ -591,10 +612,19 @@ def run(config):
         unidentified = None
         
     # Initialize the progress bar
-    pb = ProgressBar(entries_len, config, seconde=5)
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(entries_len, config)
+    else:
+        pb = None
 
     # Check and store primers and tags
-    infos, primer_list = read_info_view(info_view, max_errors=config['ngsfilter']['error'], verbose=False, not_aligned=not_aligned)   # TODO obi verbose option
+    try:
+        infos, primer_list = read_info_view(info_view, max_errors=config['ngsfilter']['error'], verbose=False, not_aligned=not_aligned)   # TODO obi verbose option
+    except RollbackException, e:
+        if unidentified is not None:
+            raise RollbackException("obi ngsfilter error, rollbacking views: "+str(e), o_view, unidentified)
+        else:
+            raise RollbackException("obi ngsfilter error, rollbacking view: "+str(e), o_view)
     
     aligner = Primer_search(primer_list, config['ngsfilter']['error'])
     
@@ -619,7 +649,8 @@ def run(config):
     try:
         for i in range(entries_len):
             PyErr_CheckSignals()
-            pb(i)
+            if pb is not None:
+                pb(i)
             if not_aligned:
                 modseq = [Nuc_Seq.new_from_stored(forward[i]), Nuc_Seq.new_from_stored(reverse[i])]
             else:
@@ -629,7 +660,13 @@ def run(config):
                 o_view[g].set(oseq.id, oseq.seq, definition=oseq.definition, quality=oseq.quality, tags=oseq)
                 g+=1
             elif unidentified is not None:
-                unidentified[u].set(oseq.id, oseq.seq, definition=oseq.definition, quality=oseq.quality, tags=oseq)
+                # Untrim sequences (put original back)
+                if len(modseq) > 1:
+                    oseq[REVERSE_SEQUENCE_COLUMN] = reverse[i].seq
+                    oseq[REVERSE_QUALITY_COLUMN] = reverse[i].quality
+                    unidentified[u].set(oseq.id, forward[i].seq, definition=oseq.definition, quality=forward[i].quality, tags=oseq)
+                else:
+                    unidentified[u].set(oseq.id, entries[i].seq, definition=oseq.definition, quality=entries[i].quality, tags=oseq)
                 u+=1
     except Exception, e:
         if unidentified is not None:
@@ -637,8 +674,9 @@ def run(config):
         else:
             raise RollbackException("obi ngsfilter error, rollbacking view: "+str(e), o_view)
     
-    pb(i, force=True)
-    print("", file=sys.stderr)
+    if pb is not None:
+        pb(i, force=True)
+        print("", file=sys.stderr)
 
     # Save command config in View and DMS comments
     command_line = " ".join(sys.argv[1:])
@@ -647,16 +685,26 @@ def run(config):
         unidentified.write_config(config, "ngsfilter", command_line, input_dms_name=input_dms_name, input_view_name=input_view_name)
         # Add comment about unidentified seqs
         unidentified.comments["info"] = "View containing sequences categorized as unidentified by the ngsfilter command"
-    output[0].record_command_line(command_line)
+    o_dms.record_command_line(command_line)
     
+    # 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()
+
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_view), file=sys.stderr)
 
-    input[0].close()
-    output[0].close()
-    info_input[0].close()
+    # If stdout output, delete the temporary result view in the input DMS
+    if type(output_0)==BufferedWriter:
+        View.delete_view(i_dms, o_view_name)
+
+    i_dms.close(force=True)
+    o_dms.close(force=True)
+    info_input[0].close(force=True)
     if unidentified is not None:
-        unidentified_input[0].close()
+        unidentified_input[0].close(force=True)
     aligner.free()
 
     logger("info", "Done.")

python/obitools3/commands/sort.pyx

@@ -4,7 +4,7 @@ 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, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.dms.view import RollbackException
 from obitools3.apps.config import logger
 from obitools3.utils cimport str2bytes
@@ -24,6 +24,7 @@ from obitools3.dms.capi.obitypes cimport OBI_BOOL, \
 import time
 import sys
 from cpython.exc cimport PyErr_CheckSignals
+from io import BufferedWriter
 
  
 NULL_VALUE = {OBI_BOOL: OBIBool_NA, 
@@ -42,6 +43,7 @@ def addOptions(parser):
     
     addMinimalInputOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
 
     group=parser.add_argument_group('obi sort specific options')
 
@@ -59,7 +61,7 @@ def addOptions(parser):
 
 
 def line_cmp(line, key, pb): 
-    pb   
+    pb
     if line[key] is None:
         return NULL_VALUE[line.view[key].data_type_int]
     else:
@@ -86,20 +88,28 @@ def run(config):
     if output is None:
         raise Exception("Could not create output view")
     o_dms = output[0]
-    o_view_name_final = output[1]
-    o_view_name = o_view_name_final
+    output_0 = output[0]
+    final_o_view_name = output[1]
     
-    # If the input and output DMS are not the same, create output view in input DMS first, then export it
-    # to output DMS, making sure the temporary view name is unique in the input DMS 
-    if i_dms != o_dms:
+    # If stdout output or the input and output DMS are not the same, create a temporary view that will be exported and deleted.
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        temporary_view_name = b"temp"
         i=0
-        while o_view_name in i_dms:
-            o_view_name = o_view_name_final+b"_"+str2bytes(str(i))
-            i+=1        
+        while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
+            temporary_view_name = temporary_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        o_view_name = temporary_view_name
+        if type(output_0)==BufferedWriter:
+            o_dms = i_dms
+    else:
+        o_view_name = final_o_view_name
 
     # Initialize the progress bar
-    pb = ProgressBar(len(i_view), config, seconde=5)
-    
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(len(i_view), config)
+    else:
+        pb = None
+
     keys = config['sort']['keys']
         
     selection = Line_selection(i_view)
@@ -110,10 +120,14 @@ def run(config):
     
     for k in keys:  # TODO order?
         PyErr_CheckSignals()
-        selection.sort(key=lambda line_idx: line_cmp(i_view[line_idx], k, pb(line_idx)), reverse=config['sort']['reverse'])
-
-    pb(len(i_view), force=True)
-    print("", file=sys.stderr)
+        if pb is not None:
+            selection.sort(key=lambda line_idx: line_cmp(i_view[line_idx], k, pb(line_idx)), reverse=config['sort']['reverse'])
+        else:
+            selection.sort(key=lambda line_idx: line_cmp(i_view[line_idx], k, None), reverse=config['sort']['reverse'])
+    
+    if pb is not None:
+        pb(len(i_view), force=True)
+        print("", file=sys.stderr)
 
     # Create output view with the sorted line selection
     try:
@@ -132,16 +146,23 @@ def run(config):
     # and delete the temporary view in the input DMS
     if i_dms != o_dms:
         o_view.close()
-        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, o_view_name_final)
-        o_view = o_dms[o_view_name_final]
+        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, final_o_view_name)
+        o_view = o_dms[final_o_view_name]
+
+    # 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()
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_view), file=sys.stderr)
 
-    # If the input and the output DMS are different, delete the temporary imported view used to create the final view
-    if i_dms != o_dms:
+    # 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(i_dms, o_view_name)
-        o_dms.close()
-    i_dms.close()
+        o_dms.close(force=True)
+
+    i_dms.close(force=True)
 
     logger("info", "Done.")

python/obitools3/commands/stats.pyx

@@ -162,7 +162,7 @@ def run(config):
     lcat=0
     
     # Initialize the progress bar
-    pb = ProgressBar(len(i_view), config, seconde=5)
+    pb = ProgressBar(len(i_view), config)
     
     for i in range(len(i_view)):
         PyErr_CheckSignals()
@@ -251,7 +251,7 @@ def run(config):
     for i in range(len(sorted_stats)):
         c = sorted_stats[i][0]
         for v in c:
-            if v is not None:
+            if type(v) == bytes:
                 print(pcat % tostr(v)+"\t", end="")
             else:
                 print(pcat % str(v)+"\t", end="")
@@ -268,6 +268,6 @@ def run(config):
         print("%7d" %catcount[c], end="")
         print("%9d" %totcount[c])
 
-    input[0].close()
+    input[0].close(force=True)
     
     logger("info", "Done.")

python/obitools3/commands/tail.pyx

@@ -4,7 +4,7 @@ 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, addMinimalOutputOption
+from obitools3.apps.optiongroups import addMinimalInputOption, addMinimalOutputOption, addNoProgressBarOption
 from obitools3.dms.view import RollbackException
 from obitools3.apps.config import logger
 from obitools3.utils cimport str2bytes
@@ -12,6 +12,7 @@ from obitools3.utils cimport str2bytes
 import time
 import sys
 from cpython.exc cimport PyErr_CheckSignals
+from io import BufferedWriter
 
 
 __title__="Keep the N last lines of a view."
@@ -21,6 +22,7 @@ def addOptions(parser):
     
     addMinimalInputOption(parser)
     addMinimalOutputOption(parser)
+    addNoProgressBarOption(parser)
 
     group=parser.add_argument_group('obi tail specific options')
 
@@ -52,31 +54,41 @@ def run(config):
     if output is None:
         raise Exception("Could not create output view")
     o_dms = output[0]
-    o_view_name_final = output[1]
-    o_view_name = o_view_name_final
+    output_0 = output[0]
+    final_o_view_name = output[1]
     
-    # If the input and output DMS are not the same, create output view in input DMS first, then export it
-    # to output DMS, making sure the temporary view name is unique in the input DMS 
-    if i_dms != o_dms:
+    # If stdout output or the input and output DMS are not the same, create a temporary view that will be exported and deleted.
+    if i_dms != o_dms or type(output_0)==BufferedWriter:
+        temporary_view_name = b"temp"
         i=0
-        while o_view_name in i_dms:
-            o_view_name = o_view_name_final+b"_"+str2bytes(str(i))
-            i+=1        
+        while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
+            temporary_view_name = temporary_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        o_view_name = temporary_view_name
+        if type(output_0)==BufferedWriter:
+            o_dms = i_dms
+    else:
+        o_view_name = final_o_view_name
     
     start = max(len(i_view) - config['tail']['count'], 0)
 
     # Initialize the progress bar
-    pb = ProgressBar(len(i_view) - start, config, seconde=5)
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(len(i_view) - start, config)
+    else:
+        pb = None
         
     selection = Line_selection(i_view)
     
     for i in range(start, len(i_view)):
         PyErr_CheckSignals()
-        pb(i)
+        if pb is not None:
+            pb(i)
         selection.append(i)
 
-    pb(i, force=True)
-    print("", file=sys.stderr)
+    if pb is not None:
+        pb(i, force=True)
+        print("", file=sys.stderr)
 
     # Save command config in View comments
     command_line = " ".join(sys.argv[1:])
@@ -97,16 +109,22 @@ def run(config):
     # and delete the temporary view in the input DMS
     if i_dms != o_dms:
         o_view.close()
-        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, o_view_name_final)
-        o_view = o_dms[o_view_name_final]
+        View.import_view(i_dms.full_path[:-7], o_dms.full_path[:-7], o_view_name, final_o_view_name)
+        o_view = o_dms[final_o_view_name]
+
+    # 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()
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_view), file=sys.stderr)
 
-    # If the input and the output DMS are different, delete the temporary imported view used to create the final view
-    if i_dms != o_dms:
+    # 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(i_dms, o_view_name)
-        o_dms.close()
-    i_dms.close()
+        o_dms.close(force=True)
+    i_dms.close(force=True)
     
     logger("info", "Done.")

python/obitools3/commands/test.pyx

@@ -23,6 +23,7 @@ from obitools3.dms.capi.obiview cimport NUC_SEQUENCE_COLUMN, \
 import shutil
 import string
 import random
+import sys
 from cpython.exc cimport PyErr_CheckSignals
 
 
@@ -300,8 +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))
@@ -317,6 +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
@@ -366,7 +371,7 @@ def random_new_view(config, infos, first=False):
             infos['view'] = View_NUC_SEQS.new(infos['dms'], random_unique_name(infos), comments=random_comments(config))   # TODO quality column
         else :
             infos['view'] = View.new(infos['dms'], random_unique_name(infos), comments=random_comments(config))   # TODO quality column
-        
+        infos['view'].write_config(config, "test", infos["command_line"], input_dms_name=[infos['dms'].name], input_view_name=["random"])
     print_test(config, repr(infos['view']))
     if v_to_clone is not None :
         if line_selection is None:
@@ -441,7 +446,7 @@ def addOptions(parser):
                        default=20,
                        type=int,
                        help="Maximum length of tuples. "
-                            "Default: 200")
+                            "Default: 20")
   
     group.add_argument('--max_ini_col_count','-o',
                        action="store", dest="test:maxinicolcount",
@@ -454,10 +459,10 @@ 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: 10000")
+                            "Default: 1000")
  
     group.add_argument('--max_elts_per_line','-e',
                        action="store", dest="test:maxelts",
@@ -497,7 +502,8 @@ def run(config):
                                     (b"OBI_SEQ", False): random_seq, (b"OBI_SEQ", True): random_seq_tuples,
                                     (b"OBI_STR", False): random_bytes, (b"OBI_STR", True): random_bytes_tuples
                                   },
-             'tests': [test_set_and_get, test_add_col, test_delete_col, test_col_alias, test_new_view]
+             'tests': [test_set_and_get, test_add_col, test_delete_col, test_col_alias, test_new_view],
+             'command_line': " ".join(sys.argv[1:])
             }
  
     # TODO ???
@@ -529,7 +535,7 @@ def run(config):
         test_taxo(config, infos)
     
     infos['view'].close()
-    infos['dms'].close()
+    infos['dms'].close(force=True)
     shutil.rmtree(config['obi']['defaultdms']+'.obidms', ignore_errors=True)
      
     print("Done.")

python/obitools3/commands/uniq.pxd

@@ -5,5 +5,5 @@ from obitools3.dms.taxo.taxo cimport Taxonomy
 from obitools3.dms.view.typed_view.view_NUC_SEQS cimport View_NUC_SEQS
 
 
-cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy)
-cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, list mergedKeys_list=*, Taxonomy taxonomy=*, bint mergeIds=*, list categories=*, int max_elts=*)
+cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy, dict config)
+cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, dict config, list mergedKeys_list=*, Taxonomy taxonomy=*, bint mergeIds=*, list categories=*, int max_elts=*)

python/obitools3/commands/uniq.pyx

@@ -14,13 +14,15 @@ from obitools3.dms.capi.obitypes cimport OBI_INT, OBI_STR, index_t
 from obitools3.apps.optiongroups import addMinimalInputOption, \
                                         addMinimalOutputOption, \
                                         addTaxonomyOption, \
-                                        addEltLimitOption
+                                        addEltLimitOption, \
+                                        addNoProgressBarOption
 from obitools3.uri.decode import open_uri
 from obitools3.apps.config import logger
-from obitools3.utils cimport tobytes, tostr
+from obitools3.utils cimport tobytes, tostr, str2bytes
 
 import sys
 from cpython.exc cimport PyErr_CheckSignals
+from io import BufferedWriter
 
 
 __title__="Group sequence records together"
@@ -32,6 +34,7 @@ def addOptions(parser):
     addTaxonomyOption(parser)
     addMinimalOutputOption(parser)
     addEltLimitOption(parser)
+    addNoProgressBarOption(parser)
  
     group = parser.add_argument_group('obi uniq specific options')
  
@@ -56,7 +59,7 @@ def addOptions(parser):
                              "(option can be used several times).")
 
 
-cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
+cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy, dict config) :
     
     cdef int             taxid
     cdef Nuc_Seq_Stored  seq
@@ -69,7 +72,7 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
     cdef object          gn_sn
     cdef object          fa_sn
 
-    # Create columns
+    # Create columns and save them for efficiency
     if b"species" in o_view and o_view[b"species"].data_type_int != OBI_INT :
         o_view.delete_column(b"species")
     if b"species" not in o_view:
@@ -77,6 +80,7 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
                           b"species", 
                           OBI_INT
                          )
+    species_column = o_view[b"species"]
 
     if b"genus" in o_view and o_view[b"genus"].data_type_int != OBI_INT :
         o_view.delete_column(b"genus")
@@ -85,6 +89,7 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
                           b"genus", 
                           OBI_INT
                          )
+    genus_column = o_view[b"genus"]
 
     if b"family" in o_view and o_view[b"family"].data_type_int != OBI_INT :
         o_view.delete_column(b"family")
@@ -93,6 +98,7 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
                           b"family", 
                           OBI_INT
                          )
+    family_column = o_view[b"family"]
 
     if b"species_name" in o_view and o_view[b"species_name"].data_type_int != OBI_STR :
         o_view.delete_column(b"species_name")
@@ -101,6 +107,7 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
                           b"species_name", 
                           OBI_STR
                          )
+    species_name_column = o_view[b"species_name"]
 
     if b"genus_name" in o_view and o_view[b"genus_name"].data_type_int != OBI_STR :
         o_view.delete_column(b"genus_name")
@@ -109,6 +116,7 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
                           b"genus_name", 
                           OBI_STR
                          )
+    genus_name_column = o_view[b"genus_name"]
 
     if b"family_name" in o_view and o_view[b"family_name"].data_type_int != OBI_STR :
         o_view.delete_column(b"family_name")
@@ -117,6 +125,7 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
                           b"family_name", 
                           OBI_STR
                          )
+    family_name_column = o_view[b"family_name"]
 
     if b"rank" in o_view and o_view[b"rank"].data_type_int != OBI_STR :
         o_view.delete_column(b"rank")
@@ -125,6 +134,7 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
                           b"rank", 
                           OBI_STR
                          )
+    rank_column = o_view[b"rank"]
 
     if b"scientific_name" in o_view and o_view[b"scientific_name"].data_type_int != OBI_STR :
         o_view.delete_column(b"scientific_name")
@@ -133,9 +143,19 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
                           b"scientific_name", 
                           OBI_STR
                          )
-        
-    for seq in o_view:     
-        PyErr_CheckSignals()   
+    scientific_name_column = o_view[b"scientific_name"]  
+    
+    # Initialize the progress bar
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(len(o_view), config)
+    else:
+        pb = None
+    
+    i=0
+    for seq in o_view:
+        PyErr_CheckSignals()
+        if pb is not None:
+            pb(i)
         if MERGED_TAXID_COLUMN in seq :
             m_taxids = []            
             m_taxids_dict = seq[MERGED_TAXID_COLUMN]
@@ -165,20 +185,24 @@ cdef merge_taxonomy_classification(View_NUC_SEQS o_view, Taxonomy taxonomy) :
             else:
                 fa_sn = None
                 tfa = None
-                
-            seq[b"species"] = tsp
-            seq[b"genus"] = tgn
-            seq[b"family"] = tfa
-                
-            seq[b"species_name"] = sp_sn
-            seq[b"genus_name"] = gn_sn
-            seq[b"family_name"] = fa_sn
             
-            seq[b"rank"] = taxonomy.get_rank(taxid)
-            seq[b"scientific_name"] = taxonomy.get_scientific_name(taxid)
+            species_column[i] = tsp
+            genus_column[i] = tgn
+            family_column[i] = tfa
+                
+            species_name_column[i] = sp_sn
+            genus_name_column[i] = gn_sn
+            family_name_column[i] = fa_sn
+            
+            rank_column[i] = taxonomy.get_rank(taxid)
+            scientific_name_column[i] = taxonomy.get_scientific_name(taxid)
+        i+=1
+    
+    if pb is not None:
+        pb(len(o_view), force=True)
+    
 
-
-cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, list mergedKeys_list=None, Taxonomy taxonomy=None, bint mergeIds=False, list categories=None, int max_elts=1000000) :
+cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, dict config, list mergedKeys_list=None, Taxonomy taxonomy=None, bint mergeIds=False, list categories=None, int max_elts=1000000) :
      
     cdef int            i
     cdef int            k
@@ -187,6 +211,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
     cdef int            u_idx
     cdef int            i_idx
     cdef int            i_count
+    cdef int            o_count
     cdef str            key_str
     cdef bytes          key
     cdef bytes          mkey
@@ -209,7 +234,6 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
     cdef Nuc_Seq_Stored i_seq
     cdef Nuc_Seq_Stored o_seq
     cdef Nuc_Seq_Stored u_seq
-    cdef Column         i_col
     cdef Column         i_seq_col
     cdef Column         i_id_col
     cdef Column         i_taxid_col
@@ -217,6 +241,8 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
     cdef Column         o_id_col
     cdef Column         o_taxid_dist_col
     cdef Column         o_merged_col
+    cdef Column         o_count_col
+    cdef Column         i_count_col
     cdef Column_line    i_mcol  
     cdef object         taxid_dist_dict
     cdef object         iter_view
@@ -252,7 +278,12 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
     mergedKeys_m = []
     for k in range(k_count):
         mergedKeys_m.append(MERGED_PREFIX + mergedKeys[k])
-     
+    
+    # Check that not trying to remerge without total count information
+    for key in mergedKeys_m:
+        if key in view and COUNT_COLUMN not in view:
+            raise Exception("\n>>>>\nError: trying to re-merge tags without total count tag. Run obi annotate to add the count tag from the relevant merged tag, i.e.: \nobi annotate --set-tag COUNT:'sum([value for key,value in sequence['MERGED_sample'].items()])' dms/input dms/output\n")
+    
     if categories is None:
         categories = []
  
@@ -274,7 +305,8 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
     iter_view = iter(view)
     for i_seq in iter_view :
         PyErr_CheckSignals()
-        pb(i)
+        if pb is not None:
+            pb(i)
          
         # This can't be done in the same line as the unique_id tuple creation because it generates a bug
         # where Cython (version 0.25.2) does not detect the reference to the categs_list variable and deallocates 
@@ -284,6 +316,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
         for x in categories :
             catl.append(i_seq[x])    
           
+        #unique_id = tuple(catl) + (i_seq_col[i],)
         unique_id = tuple(catl) + (i_seq_col.get_line_idx(i),)
         #unique_id = tuple(i_seq[x] for x in categories) + (seq_col.get_line_idx(i),)  # The line that cython can't read properly
          
@@ -320,7 +353,14 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
     for k in range(k_count):
         key = mergedKeys[k]
         merged_col_name = mergedKeys_m[k]
-        i_col = view[key]
+        
+        if merged_infos[merged_col_name]['nb_elts'] == 1:
+            raise Exception("Can't merge information from a tag with only one element (e.g. one sample ; don't use -m option)")
+        
+        if merged_col_name in view:
+            i_col = view[merged_col_name]
+        else:
+            i_col = view[key]
         
         if merged_infos[merged_col_name]['nb_elts'] > max_elts:
             str_merged_cols.append(merged_col_name)
@@ -338,6 +378,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
                               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
                              )
@@ -360,6 +401,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
                               OBI_INT,
                               nb_elements_per_line=len(view),
                               elements_names=[id for id in i_id_col],
+                              dict_column=True,
                               alias=TAXID_DIST_COLUMN
                              )
 
@@ -374,23 +416,35 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
                           alias=MERGED_COLUMN
                          )
 
-    # Keep columns that are going to be used a lot in variables 
+    # Keep columns in variables for efficiency
     o_id_col = o_view[ID_COLUMN]
     if TAXID_DIST_COLUMN in o_view:
         o_taxid_dist_col = o_view[TAXID_DIST_COLUMN]
     if MERGED_COLUMN in o_view:
         o_merged_col = o_view[MERGED_COLUMN]
+    if COUNT_COLUMN not in o_view:
+        Column.new_column(o_view,
+                          COUNT_COLUMN,
+                          OBI_INT)
+    o_count_col = o_view[COUNT_COLUMN]
+    if COUNT_COLUMN in view:
+        i_count_col = view[COUNT_COLUMN]
+    
+    if pb is not None:
+        pb(len(view), force=True)
+        print("")
         
-    pb(len(view), force=True)
-    print("")
     logger("info", "Second browsing through the input")
+    
     # Initialize the progress bar
-    pb = ProgressBar(len(uniques), seconde=5)
+    if pb is not None:
+        pb = ProgressBar(len(view))
+        
     o_idx = 0
+    total_treated = 0
     
     for unique_id in uniques :
         PyErr_CheckSignals()
-        pb(o_idx)
         
         merged_sequences = uniques[unique_id]
         
@@ -407,7 +461,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
             merged_list = list(set(merged_list)) # deduplicate the list
             o_merged_col[o_idx] = merged_list
 
-        o_seq[COUNT_COLUMN] = 0
+        o_count = 0
 
         if TAXID_DIST_COLUMN in u_seq and i_taxid_dist_col[u_idx] is not None:
             taxid_dist_dict = i_taxid_dist_col[u_idx]
@@ -419,16 +473,20 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
             merged_dict[mkey] = {}
 
         for i_idx in merged_sequences:
-                        
+            PyErr_CheckSignals()
+            
+            if pb is not None:
+                pb(total_treated)
+ 
             i_id = i_id_col[i_idx]
             i_seq = view[i_idx]
 
-            if COUNT_COLUMN not in i_seq or i_seq[COUNT_COLUMN] is None:
+            if COUNT_COLUMN not in i_seq or i_count_col[i_idx] is None:
                 i_count = 1
             else:
-                i_count = i_seq[COUNT_COLUMN]
+                i_count = i_count_col[i_idx]
  
-            o_seq[COUNT_COLUMN] += i_count
+            o_count += i_count
         
             for k in range(k_count):
                                 
@@ -463,44 +521,55 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, li
                                 mcol[key2] = i_mcol[key2]
                             else:
                                 mcol[key2] = mcol[key2] + i_mcol[key2]
-            
-            # Write taxid_dist
-            if mergeIds and TAXID_COLUMN in mergedKeys:
-                if TAXID_DIST_COLUMN in str_merged_cols:
-                    o_taxid_dist_col[o_idx] = str(taxid_dist_dict)
-                else:
-                    o_taxid_dist_col[o_idx] = taxid_dist_dict
-            
-            # Write merged dicts
-            for mkey in merged_dict: 
-                if mkey in str_merged_cols:
-                    mkey_cols[mkey][o_idx] = str(merged_dict[mkey])
-                else:
-                    mkey_cols[mkey][o_idx] = merged_dict[mkey]
-                    # Sets NA values to 0  # TODO discuss, for now keep as None and test for None instead of testing for 0 in tools
-                    #for key in mkey_cols[mkey][o_idx]:
-                    #    if mkey_cols[mkey][o_idx][key] is None:
-                    #        mkey_cols[mkey][o_idx][key] = 0
-                            
+                                        
             for key in i_seq.keys():
                 # Delete informations that differ between the merged sequences
-                # TODO make special columns list?
+                # TODO make special columns list? // could be more efficient
                 if key != COUNT_COLUMN and key != ID_COLUMN and key != NUC_SEQUENCE_COLUMN and key in o_seq and o_seq[key] != i_seq[key] \
                     and key not in merged_dict :
                     o_seq[key] = None
+            
+            total_treated += 1
+            
+        # Write merged dicts
+        for mkey in merged_dict: 
+            if mkey in str_merged_cols:
+                mkey_cols[mkey][o_idx] = str(merged_dict[mkey])
+            else:
+                mkey_cols[mkey][o_idx] = merged_dict[mkey]
+                # Sets NA values to 0  # TODO discuss, for now keep as None and test for None instead of testing for 0 in tools
+                #for key in mkey_cols[mkey][o_idx]:
+                #    if mkey_cols[mkey][o_idx][key] is None:
+                #        mkey_cols[mkey][o_idx][key] = 0
 
+        # Write taxid_dist
+        if mergeIds and TAXID_COLUMN in mergedKeys:
+            if TAXID_DIST_COLUMN in str_merged_cols:
+                o_taxid_dist_col[o_idx] = str(taxid_dist_dict)
+            else:
+                o_taxid_dist_col[o_idx] = taxid_dist_dict
+                
+        o_count_col[o_idx] = o_count
         o_idx += 1
     
+    if pb is not None:
+        pb(len(view), force=True)
+    
     # Deletes quality columns if there is one because the matching between sequence and quality will be broken (quality set to NA when sequence not)
     if QUALITY_COLUMN in view:
         o_view.delete_column(QUALITY_COLUMN)
     if REVERSE_QUALITY_COLUMN in view:
         o_view.delete_column(REVERSE_QUALITY_COLUMN)
     
+    # Delete old columns that are now merged
+    for k in range(k_count):
+        if mergedKeys[k] in o_view:
+            o_view.delete_column(mergedKeys[k])
+    
     if taxonomy is not None:
         print("")  # TODO because in the middle of progress bar. Better solution?
         logger("info", "Merging taxonomy classification")
-        merge_taxonomy_classification(o_view, taxonomy)
+        merge_taxonomy_classification(o_view, taxonomy, config)
 
 
 
@@ -532,8 +601,23 @@ def run(config):
     if output is None:
         raise Exception("Could not create output view")
     
+    i_dms = input[0]
     entries = input[1]
-    o_view = output[1]
+    o_dms = output[0]
+    output_0 = output[0]
+    
+    # If stdout output create a temporary view that will be exported and deleted.
+    if type(output_0)==BufferedWriter:
+        temporary_view_name = b"temp"
+        i=0
+        while temporary_view_name in i_dms:  # Making sure view name is unique in input DMS
+            temporary_view_name = temporary_view_name+b"_"+str2bytes(str(i))
+            i+=1
+        o_view_name = temporary_view_name
+        o_dms = i_dms
+        o_view = View_NUC_SEQS.new(i_dms, o_view_name)
+    else:
+        o_view = output[1]
     
     if 'taxoURI' in config['obi'] and config['obi']['taxoURI'] is not None:
         taxo_uri = open_uri(config['obi']['taxoURI'])
@@ -544,15 +628,19 @@ def run(config):
         taxo = None
 
     # Initialize the progress bar
-    pb = ProgressBar(len(entries), config, seconde=5)
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(len(entries), config)
+    else:
+        pb = None
     
-    try:
-        uniq_sequences(entries, o_view, pb, mergedKeys_list=config['uniq']['merge'], taxonomy=taxo, mergeIds=config['uniq']['mergeids'], categories=config['uniq']['categories'], max_elts=config['obi']['maxelts'])       
-    except Exception, e:
-        raise RollbackException("obi uniq error, rollbacking view: "+str(e), o_view)
+    if len(entries) > 0:
+        try:
+            uniq_sequences(entries, o_view, pb, config, mergedKeys_list=config['uniq']['merge'], taxonomy=taxo, mergeIds=config['uniq']['mergeids'], categories=config['uniq']['categories'], max_elts=config['obi']['maxelts'])       
+        except Exception, e:
+            raise RollbackException("obi uniq error, rollbacking view: "+str(e), o_view)
     
-    pb(len(entries), force=True)
-    print("", file=sys.stderr)
+    if pb is not None:
+        print("", file=sys.stderr)
 
     # Save command config in View and DMS comments
     command_line = " ".join(sys.argv[1:])
@@ -562,13 +650,23 @@ def run(config):
         input_dms_name.append(config['obi']['taxoURI'].split("/")[-3])
         input_view_name.append("taxonomy/"+config['obi']['taxoURI'].split("/")[-1])
     o_view.write_config(config, "uniq", command_line, input_dms_name=input_dms_name, input_view_name=input_view_name)
-    output[0].record_command_line(command_line)
+    o_dms.record_command_line(command_line)
+
+    # 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()
 
     #print("\n\nOutput view:\n````````````", file=sys.stderr)
     #print(repr(o_view), file=sys.stderr)
-    
-    input[0].close()
-    output[0].close()
+
+    # If stdout output, delete the temporary result view in the input DMS
+    if type(output_0)==BufferedWriter:
+        View.delete_view(i_dms, o_view_name)
+
+    i_dms.close(force=True)
+    o_dms.close(force=True)
 
     logger("info", "Done.")
 

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

@@ -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
@@ -68,3 +69,6 @@ cdef extern from "obidmscolumn.h" nogil:
     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
@@ -94,6 +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,
@@ -103,13 +105,17 @@ 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.pxd

@@ -22,6 +22,7 @@ cdef class Column(OBIWrapper) :
     
     cdef inline OBIDMS_column_p pointer(self)
     cdef read_elements_names(self)
+    cpdef list keys(self)
     
     @staticmethod
     cdef type get_column_class(obitype_t obitype, bint multi_elts, bint tuples)

python/obitools3/dms/column/column.pyx

@@ -14,6 +14,7 @@ from ..capi.obidms cimport obi_import_column
 from ..capi.obidmscolumn cimport OBIDMS_column_header_p, \
                                  obi_close_column, \
                                  obi_get_elements_names, \
+                                 obi_column_formatted_infos, \
                                  obi_column_write_comments
                                      
 from ..capi.obiutils cimport obi_format_date
@@ -38,8 +39,9 @@ from obitools3.utils cimport tobytes, \
  
 from obitools3.dms.column import typed_column
  
-from libc.stdlib  cimport free
- 
+from libc.stdlib cimport free
+from libc.string cimport strcpy
+  
 import importlib
 import inspect
 import pkgutil
@@ -88,6 +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"",
@@ -96,6 +99,7 @@ cdef class Column(OBIWrapper) :
                    object    alias=b""):
         # TODO indexer_name?
         
+        cdef Column column
         cdef bytes  column_name_b = tobytes(column_name)
         cdef bytes  alias_b = tobytes(alias)
         cdef bytes  comments_b = str2bytes(json.dumps(bytes2str_object(comments)))
@@ -131,13 +135,14 @@ cdef class Column(OBIWrapper) :
                          raise RuntimeError("Cannot create column %s in view %s: trying to create quality column but no NUC_SEQ column to associate it with in the view" % (bytes2str(column_name_b),
                                                                                bytes2str(view.name)))
                     associated_column_name_b = NUC_SEQUENCE_COLUMN
-                    associated_column_version = view[NUC_SEQUENCE_COLUMN].version
+                    associated_column_version = view[NUC_SEQUENCE_COLUMN].version                    
                 elif column_name == REVERSE_QUALITY_COLUMN:
                     if REVERSE_SEQUENCE_COLUMN not in view:
                          raise RuntimeError("Cannot create column %s in view %s: trying to create reverse quality column but no REVERSE_SEQUENCE column to associate it with in the view" % (bytes2str(column_name_b),
                                                                                bytes2str(view.name)))
                     associated_column_name_b = REVERSE_SEQUENCE_COLUMN
                     associated_column_version = view[REVERSE_SEQUENCE_COLUMN].version
+                    
         
         if (obi_view_add_column(view                      = view.pointer(),
                                 column_name               = column_name_b,
@@ -148,6 +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,
@@ -157,8 +163,19 @@ cdef class Column(OBIWrapper) :
                                 create                    = True)<0):
             raise RuntimeError("Cannot create column %s in view %s" % (bytes2str(column_name_b),
                                                                        bytes2str(view.name)))
-
-        return Column.open(view, alias_b)
+        
+        column = Column.open(view, alias_b)
+        
+        # Automatically associate nuc sequence column to quality column if necessary
+        if data_type == OBI_QUAL:
+            if column_name == QUALITY_COLUMN:
+                view[NUC_SEQUENCE_COLUMN].associated_column_name = column.name
+                view[NUC_SEQUENCE_COLUMN].associated_column_version = column.version
+            elif column_name == REVERSE_QUALITY_COLUMN:
+                view[REVERSE_SEQUENCE_COLUMN].associated_column_name = column.name
+                view[REVERSE_SEQUENCE_COLUMN].associated_column_version = column.version
+        
+        return column
  
  
     @staticmethod
@@ -185,7 +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
@@ -221,6 +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,
@@ -287,10 +305,25 @@ cdef class Column(OBIWrapper) :
     
     @OBIWrapper.checkIsActive
     def __repr__(self) :
-        cdef bytes s
-        s = self._alias + b", data type: " + self.data_type
-        return bytes2str(s)
+        cdef str s
+        cdef char* sc
+        cdef OBIDMS_column_p pointer = self.pointer()
+        sc = obi_column_formatted_infos(pointer, False)
+        s = bytes2str(sc)
+        free(sc)
+        return 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
     
@@ -316,7 +349,10 @@ cdef class Column(OBIWrapper) :
         free(elts_names_b)
         return elts_names_list
     
-    
+    cpdef list keys(self):
+        return self._elements_names
+
+
     # Column alias property getter and setter
     @property
     def name(self):
@@ -333,7 +369,7 @@ cdef class Column(OBIWrapper) :
     @property
     def elements_names(self):
         return self._elements_names
- 
+       
     # nb_elements_per_line property getter
     @property
     def nb_elements_per_line(self):
@@ -341,6 +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):
@@ -397,6 +440,31 @@ cdef class Column(OBIWrapper) :
             raise OBIDeactivatedInstanceError()
         return obi_format_date(self.pointer().header.creation_date)
 
+
+    # associated_column name property getter and setter
+    @property
+    def associated_column_name(self):
+        if not self.active() :
+            raise OBIDeactivatedInstanceError()
+        return self.pointer().header.associated_column.column_name
+
+    @associated_column_name.setter
+    def associated_column_name(self, object new_name):
+        strcpy(self.pointer().header.associated_column.column_name, tobytes(new_name))
+
+
+    # associated_column version property getter and setter
+    @property
+    def associated_column_version(self):
+        if not self.active() :
+            raise OBIDeactivatedInstanceError()
+        return self.pointer().header.associated_column.version
+
+    @associated_column_version.setter
+    def associated_column_version(self, int new_version):
+        self.pointer().header.associated_column.version = new_version
+
+
     # comments property getter
     @property
     def comments(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/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):    
     
@@ -94,16 +97,16 @@ cdef class DMS(OBIWrapper):
         return dms
     
     
-    def close(self) :
+    def close(self, force=False) :
         '''
-        Closes the DMS instance and free the associated memory
+        Closes the DMS instance and free the associated memory (no counter, closing is final)
         
         The `close` method is automatically called by the object destructor.
         '''
         cdef OBIDMS_p pointer = self.pointer()
         if self.active() :
             OBIWrapper.close(self)
-            if (obi_close_dms(pointer, False)) < 0 :
+            if (obi_close_dms(pointer, force=force)) < 0 :
                 raise Exception("Problem closing an OBIDMS")
 
 
@@ -223,11 +226,24 @@ cdef class DMS(OBIWrapper):
 
 
     @OBIWrapper.checkIsActive
-    def __repr__(self):
-        cdef str s 
-        s=""
-        for view_name in self.keys():
-            s = s + repr(self.get_view(view_name)) + "\n"
+    def __repr__(self) :
+        cdef str s
+        cdef char* sc
+        cdef OBIDMS_p pointer = self.pointer()
+        sc = obi_dms_formatted_infos(pointer, False)
+        s = bytes2str(sc)
+        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
     
 
@@ -254,7 +270,8 @@ cdef class DMS(OBIWrapper):
     # bash command history property getter
     @property
     def bash_history(self):
-        s = b"#!/bin/bash\n\n"
+        #s = b"#!${bash}/bin/bash\n\n"
+        s = b""
         first = True
         for command in self.command_line_history:
             s+=b"#"

python/obitools3/dms/obiseq.pxd

@@ -39,4 +39,6 @@ cdef class Nuc_Seq_Stored(Seq_Stored) :
     cpdef set_quality_char(self, object new_qual, int offset=*)
     cpdef object build_quality_array(self, list quality)
     cpdef bytes build_reverse_complement(self)
-    cpdef str get_str(self)
+    cpdef str get_str(self)
+    cpdef repr_bytes(self)
+    

python/obitools3/dms/obiseq.pyx

@@ -431,9 +431,12 @@ cdef class Nuc_Seq_Stored(Seq_Stored) :
         return len(self._view.get_column(NUC_SEQUENCE_COLUMN).get_line(self.index))
 
     def __repr__(self):
+        return bytes2str(self.repr_bytes())
+    
+    cpdef repr_bytes(self):
         if self.quality is None:
             formatter = FastaFormat()
         else:
             formatter = FastqFormat()
-        return bytes2str(formatter(self))
+        return formatter(self)
 

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 :
-            raise NotImplementedError()
+        elif type(ref) == str or type(ref) == bytes :
+            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.pxd

@@ -20,6 +20,10 @@ cdef class View(OBIWrapper):
     cdef DMS _dms
     
     cdef inline Obiview_p pointer(self)   
+
+    cpdef print_to_output(self, 
+                          object output, 
+                          bint noprogressbar=*)
         
     cpdef delete_column(self, 
                         object column_name,
@@ -61,6 +65,8 @@ cdef class Line :
     cdef index_t _index
     cdef View    _view
     
+    cpdef repr_bytes(self)
+
 
 cdef register_view_class(bytes view_type_name,
                          type view_class)

python/obitools3/dms/view/view.pyx

@@ -6,6 +6,9 @@ cdef dict __VIEW_CLASS__= {}
 
 from libc.stdlib  cimport malloc
 
+from obitools3.apps.progress cimport ProgressBar  # @UnresolvedImport
+from obitools3.version import version
+
 from ..capi.obiview cimport Alias_column_pair_p, \
                             obi_new_view, \
                             obi_open_view, \
@@ -16,7 +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
@@ -48,10 +53,15 @@ from ..capi.obidms cimport obi_import_view
 
 from obitools3.format.tab import TabFormat
 
+from cpython.exc cimport PyErr_CheckSignals
+
 import importlib
 import inspect
 import pkgutil
 import json
+import sys
+
+from libc.stdlib cimport free
 
 
 cdef class View(OBIWrapper) :
@@ -178,13 +188,50 @@ cdef class View(OBIWrapper) :
     
     
     @OBIWrapper.checkIsActive
-    def __repr__(self) :         
-        cdef str s = "#View name:\n{name:s}\n#Line count:\n{line_count:d}\n#Columns:\n".format(name       = bytes2str(self.name),
-                                                                                                                         line_count = self.line_count)
-        for column_name in self.keys() :
-            s = s + repr(self[column_name]) + '\n'
+    def __repr__(self) :
+        cdef str s
+        cdef char* sc
+        cdef Obiview_p pointer = self.pointer()
+        sc = obi_view_formatted_infos(pointer, False)
+        s = bytes2str(sc)
+        free(sc)
         return s
-    
+
+
+    @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
+
+
+    cpdef print_to_output(self, object output, bint noprogressbar=False):      
+        
+        cdef int i
+        cdef Line entry
+        
+        self.checkIsActive(self)
+        
+        # Initialize the progress bar
+        if noprogressbar == False:
+            pb = ProgressBar(len(self))
+        else:
+            pb = None
+        i=0
+        for entry in self:
+            PyErr_CheckSignals()
+            if pb is not None:
+                pb(i)
+            output.write(entry.repr_bytes()+b"\n")
+            i+=1
+        if pb is not None:
+            pb(len(self), force=True)
+            print("", file=sys.stderr)
+
      
     def keys(self):
         
@@ -296,7 +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]
@@ -357,6 +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,
@@ -405,6 +453,7 @@ cdef class View(OBIWrapper) :
             for i in range(len(input_view_name)):
                 input_str.append(tostr(input_dms_name[i])+"/"+tostr(input_view_name[i]))
         comments["input_str"] = input_str
+        comments["version"] = version
         return bytes2str_object(comments)
 
     
@@ -526,11 +575,12 @@ cdef class View(OBIWrapper) :
     # bash command history property getter
     @property
     def bash_history(self):
-        s = b"#!/bin/bash\n\n"
+        s = b""
         first = True
         for level in self.view_history:
             command_list = [level[input][b"command_line"] for input in level.keys()]
             for command in command_list:
+                s+=b"obi "
                 s+=command
                 s+=b"\n"
         return s
@@ -756,8 +806,12 @@ cdef class Line :
 
 
     def __repr__(self):
+        return bytes2str(self.repr_bytes())
+
+
+    cpdef repr_bytes(self):
         formatter = TabFormat(header=False)
-        return bytes2str(formatter(self))
+        return formatter(self)
 
 
     # View property getter

python/obitools3/format/tab.pyx

@@ -3,8 +3,9 @@
 cimport cython
 from obitools3.dms.view.view cimport Line
 from obitools3.utils cimport bytes2str_object, str2bytes, tobytes
-from obitools3.dms.column.column cimport Column_line
+from obitools3.dms.column.column cimport Column_line, Column_multi_elts
 
+import sys
 
 cdef class TabFormat:
     
@@ -22,23 +23,45 @@ cdef class TabFormat:
         if self.first_line:
             self.tags = [k for k in data.keys()]
         
+        if self.header and self.first_line:
+            for k in self.tags:
+                if isinstance(data.view[k], Column_multi_elts):
+                    keys = data.view[k].keys()
+                    keys.sort()
+                    for k2 in keys:
+                        line.append(tobytes(k)+b':'+tobytes(k2))
+                else:
+                    line.append(tobytes(k))
+            r = self.sep.join(value for value in line)
+            r += b'\n'
+            line = []
+                    
         for k in self.tags:
-            
-            if self.header and self.first_line:
-                value = tobytes(k)
+            value = data[k]
+            if isinstance(data.view[k], Column_multi_elts):
+                keys = data.view[k].keys()
+                keys.sort()
+                if value is None:  # all keys at None
+                    for k2 in keys: # TODO could be much more efficient
+                        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:
-                value = data[k]
                 if value is not None:
-                    if type(value) == Column_line:
-                        value = value.bytes()
-                    else:
-                        value = str2bytes(str(bytes2str_object(value))) # genius programming
-                if value is None:
-                    value = self.NAString
-            
-            line.append(value)
-      	
+                    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/libalign/_solexapairend.pyx

@@ -183,12 +183,13 @@ def buildConsensus(ali, seq, ref_tags=None):
         # doesn't work because uint8_t* are forced into bytes by cython (nothing read/kept beyond 0 values)
         #obi_set_qual_int_with_elt_idx_and_col_p_in_view(view_p, col_p, seq.index, 0, ali.consensus_qual, ali.consensus_len)
         seq.set(ref_tags.id+b"_CONS", ali.consensus_seq, quality=ali.consensus_qual)
-        seq[b'ali_length'] = ali.consensus_len
-        seq[b'score_norm']=float(ali.score)/ali.consensus_len
+        seq[b"seq_length"] = ali.consensus_len
+        seq[b"overlap_length"] = ali.overlap_len
+        seq[b'score_norm']=round(float(ali.score)/ali.overlap_len, 3)
         seq[b'shift']=ali.shift
     else:
         if len(ali[0])>999:   # TODO why?
-            raise AssertionError,"Too long alignemnt"
+            raise AssertionError,"Too long alignment"
     
         ic=IterOnConsensus(ali)
     
@@ -250,11 +251,22 @@ def buildJoinedSequence(ali, reverse, seq, forward=None):
     quality.extend(reverse.quality)
     seq.set(forward.id +b"_PairedEnd", s, definition=forward.definition, quality=quality)
     seq[b"score"]=ali.score
-    seq[b"ali_direction"]=ali.direction
+    if len(ali.direction) > 0:
+        seq[b"ali_direction"]=ali.direction
+    else:
+        seq[b"ali_direction"]=None
     seq[b"mode"]=b"joined"
-    seq[b"pairedend_limit"]=len(forward)    
+    seq[b"pairedend_limit"]=len(forward)
+    seq[b"seq_length"] = ali.consensus_len
+    seq[b"overlap_length"] = ali.overlap_len
+    if ali.overlap_len > 0:
+        seq[b'score_norm']=round(float(ali.score)/ali.overlap_len, 3)
+    else:
+        seq[b"score_norm"]=0.0
+  
     for tag in forward:
-        if tag != REVERSE_SEQUENCE_COLUMN and tag != REVERSE_QUALITY_COLUMN:
+        if tag != REVERSE_SEQUENCE_COLUMN and tag != REVERSE_QUALITY_COLUMN and \
+            tag != NUC_SEQUENCE_COLUMN and tag != QUALITY_COLUMN:
             seq[tag] = forward[tag]
     return seq
 

python/obitools3/parsers/embl.pyx

@@ -156,6 +156,9 @@ def emblIterator_file(lineiterator,
         yield seq
         read+=1
     
+    # Last sequence
+    seq = emblParser(entry)
+        
     yield seq
     
     free(entry)
@@ -174,7 +177,7 @@ def emblIterator_dir(dir_path,
     for filename in files:
         if read==only:
             return
-        print("Parsing file %s (%d/%d)" % (tostr(filename), read_files, len(files)))
+        print("Parsing file %s (%d/%d)" % (tostr(filename), read_files+1, len(files)))
         f = uopen(filename)
         if only is not None:
             only_f = only-read

python/obitools3/parsers/fasta.pyx

@@ -104,6 +104,7 @@ def fastaNucIterator(lineiterator,
     cdef bytes      sequence
     cdef int        skipped, ionly, read
     cdef Nuc_Seq    seq
+    cdef bint       stop
     
     if only is None:
         ionly = -1
@@ -130,7 +131,8 @@ def fastaNucIterator(lineiterator,
     else:
         line = firstline       
     
-    while True:
+    stop=False
+    while not stop:
                 
         if ionly >= 0 and read >= ionly:
             break
@@ -153,7 +155,7 @@ def fastaNucIterator(lineiterator,
                 s.append(line[0:-1])
                 line = next(iterator)
         except StopIteration:
-            pass
+            stop=True
         
         sequence  = b"".join(s)        
         

python/obitools3/parsers/genbank.pyx

@@ -22,11 +22,12 @@ 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)
-_cleanSeq      = re.compile(b'[ \n0-9]+')
+_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)
 _deMatcher     = re.compile(b'(?<=^DEFINITION  ).+\n( .+\n)*',re.M)
 _cleanDe       = re.compile(b'\n *')
@@ -42,7 +43,8 @@ def genbankParser(bytes text):
         ft     = _featureMatcher.search(text).group()
         
         s      = _seqMatcher.search(text).group()
-        s      = _cleanSeq.sub(b'', s).upper()
+        s      = _cleanSeq1.sub(b'', s)
+        s      = _cleanSeq2.sub(b'', s)
         
         acs    = _acMatcher.search(text).group()
         acs    = acs.split()
@@ -51,23 +53,23 @@ def genbankParser(bytes text):
         
         de     = _deMatcher.search(header).group()
         de     = _cleanDe.sub(b' ',de).strip().strip(b'.')
-
+    
+        tags = {}
+        extractTaxon(ft, tags)
+            
+        seq = Nuc_Seq(ac,
+                      s,
+                      definition=de,
+                      quality=None,
+                      offset=-1,
+                      tags=tags)
+    
     except Exception as e:
         print("\nCould not import sequence id:", text.split()[1], "(error raised:", e, ")")
         # Do not raise any Exception if you need the possibility to resume the generator
         # (Python generators can't resume after any exception is raised)
         return None
-    
-    tags = {}
-    extractTaxon(ft, tags)
-        
-    seq = Nuc_Seq(ac,
-                  s,
-                  definition=de,
-                  quality=None,
-                  offset=-1,
-                  tags=tags)
-    
+
     return seq
 
 
@@ -153,6 +155,9 @@ def genbankIterator_file(lineiterator,
         yield seq
         read+=1
     
+    # Last sequence
+    seq = genbankParser(entry)
+    
     yield seq
     
     free(entry)
@@ -168,10 +173,12 @@ def genbankIterator_dir(dir_path,
     read = 0
     read_files = 0
     files = [filename for filename in glob.glob(os.path.join(path, b'*.gbff*'))]
+    files.extend([filename for filename in glob.glob(os.path.join(path, b'*.seq*'))])  # new genbank extension
+    files = list(set(files))
     for filename in files:
         if read==only:
             return
-        print("Parsing file %s (%d/%d)" % (tostr(filename), read_files, len(files)))
+        print("Parsing file %s (%d/%d)" % (tostr(filename), read_files+1, len(files)))
         f = uopen(filename)
         if only is not None:
             only_f = only-read

python/obitools3/uri/decode.pyx

@@ -210,10 +210,11 @@ def open_uri(uri,
     
     error = None
     
-    if scheme==b"dms" or \
-        (scheme==b"" and \
-         (((not input) and "outputformat" not in config["obi"]) or \
-         (input and "inputformat" not in config["obi"]))):   # TODO maybe not best way
+    if urib != b"-" and \
+        (scheme==b"dms" or \
+         (scheme==b"" and \
+          (((not input) and "outputformat" not in config["obi"]) or \
+          (input and "inputformat" not in config["obi"])))):   # TODO maybe not best way
         
         if default_dms is not None and b"/" not in urip.path:   # assuming view to open in default DMS (TODO discuss)
             dms=(default_dms, urip.path)
@@ -275,11 +276,11 @@ def open_uri(uri,
             iseq = urib
             objclass = bytes
     else:  # TODO update uopen to be able to write? 
-        if urip.path:
-            file = open(urip.path, 'wb')
-        else:
+        if not urip.path or urip.path == b'-':
             file = sys.stdout.buffer
-        
+        else:
+            file = open(urip.path, 'wb')
+
     if file is not None:
         qualifiers=parse_qs(urip.query)
         
@@ -463,7 +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/utils.pxd

@@ -2,7 +2,7 @@
 
 from obitools3.dms.capi.obitypes cimport obitype_t, index_t
 
-cpdef bytes format_separator(bytes format)
+cpdef bytes format_uniq_pattern(bytes format)
 cpdef int count_entries(file, bytes format)
 
 cdef obi_errno_to_exception(index_t line_nb=*, object elt_id=*, str error_message=*)

python/obitools3/utils.pyx

@@ -24,11 +24,11 @@ import glob
 import gzip
 
 
-cpdef bytes format_separator(bytes format):
+cpdef bytes format_uniq_pattern(bytes format):
     if format == b"fasta":
         return b"\n>"
     elif format == b"fastq":
-        return b"\n@"
+        return b"\n\+\n"
     elif format == b"ngsfilter" or format == b"tabular":
         return b"\n"
     elif format == b"genbank" or format == b"embl":
@@ -42,7 +42,7 @@ cpdef bytes format_separator(bytes format):
 cpdef int count_entries(file, bytes format):
     
     try:
-        sep = format_separator(format)
+        sep = format_uniq_pattern(format)
         if sep is None:
             return -1
         sep = re.compile(sep)
@@ -72,7 +72,7 @@ cpdef int count_entries(file, bytes format):
                 return -1
             mmapped_file = mmap.mmap(f.fileno(), 0, access=mmap.ACCESS_READ)
             total_count += len(re.findall(sep, mmapped_file))
-            if format != b"ngsfilter" and format != b"tabular":
+            if format != b"ngsfilter" and format != b"tabular" and format != b"embl" and format != b"genbank" and format != b"fastq":
                 total_count += 1 # adding +1 for 1st entry because separators include \n (ngsfilter and tabular already count one more because of last \n)
             
     except:
@@ -166,7 +166,9 @@ cdef object bytes2str_object(object value):  # Only works if complex types are d
                     value[k] = bytes2str(v)
             if type(k) == bytes:
                 value[bytes2str(k)] = value.pop(k)
-    elif isinstance(value, list):
+    elif isinstance(value, list) or isinstance(value, tuple):
+        if isinstance(value, tuple):
+            value = list(value)
         for i in range(len(value)):
             if isinstance(value[i], list) or isinstance(value[i], dict):
                 value[i] = bytes2str_object(value[i])

python/obitools3/version.py

@@ -1,5 +1,5 @@
 major = 3
 minor = 0
-serial= '0-beta10'
+serial= '1b1'
 
-version ="%d.%02d.%s" % (major,minor,serial)
+version ="%d.%d.%s" % (major,minor,serial)

requirements.txt

@@ -0,0 +1,5 @@
+--extra-index-url https://pypi.python.org/simple/
+Cython>=0.24
+Sphinx>=1.2.0
+ipython>=3.0.0
+breathe>=4.0.0

setup.py

@@ -5,8 +5,9 @@ import re
 import subprocess
 
 from distutils import log
-from distutils.core import setup
-    
+#from distutils.core import setup
+from setuptools import setup    # to work with pip
+
 from distutils.core import Extension
 from distutils.sysconfig import get_python_lib
 
@@ -26,10 +27,11 @@ class Distribution(ori_Distribution):
         
         ori_Distribution.__init__(self, attrs)
         
-        self.global_options.insert(0,('cobitools3', None, "intall location of the C library"
+        self.global_options.insert(0,('cobitools3', None, "install location of the C library"
                                      ))
 
 from distutils.command.build import build as build_ori
+from setuptools.command.bdist_egg import bdist_egg as bdist_egg_ori
 from distutils.core import Command
 
 
@@ -70,6 +72,12 @@ class build(build_ori):
         build_ori.run(self)
 
 
+class bdist_egg(bdist_egg_ori):
+    def run(self):
+        self.run_command('build_clib')
+        bdist_egg_ori.run(self)
+
+
 sys.path.append(os.path.abspath("python"))
 
 
@@ -88,9 +96,10 @@ PACKAGE     = "OBITools3"
 VERSION     = version
 AUTHOR      = 'Celine Mercier'
 EMAIL       = 'celine.mercier@metabarcoding.org'
-URL         = "http://metabarcoding.org/obitools3"
+URL         = "https://metabarcoding.org/obitools3"
+PLATFORMS   = "posix"
 LICENSE     = "CeCILL-V2"
-DESCRIPTION = "Tools and library for DNA metabarcoding",
+DESCRIPTION = "A package for the management of analyses and data in DNA metabarcoding."
 PYTHONMIN   = '3.5'
 
 SRC       = 'python'
@@ -147,17 +156,24 @@ classifiers=['Development Status :: 4 - Beta',
              'Topic :: Utilities',
              ]
 
+with open("README.md", "r") as fh:
+    long_description = fh.read()
+
 setup(name=PACKAGE,
       description=DESCRIPTION,
+      long_description=long_description,
+      long_description_content_type="text/markdown",
       classifiers=classifiers,
       version=VERSION,
       author=AUTHOR,
       author_email=EMAIL,
+      platforms=PLATFORMS,
       license=LICENSE,
       url=URL,
       ext_modules=xx,
       distclass=Distribution,
       cmdclass={'build': build,
+                'bdist_egg': bdist_egg,
                 'build_clib': build_clib},
       cobitools3=get_python_lib(),
       packages = findPackage('python'),

src/build_reference_db.c

@@ -157,7 +157,7 @@ int build_reference_db(const char* dms_name,
 	ecotx_t* lca_2 = NULL;
 	ecotx_t* lca = NULL;
 	index_t idx1, idx2;
-	index_t i, j, k;
+	index_t i, j, k, count;
 	int32_t taxid_array_length;
 	int32_t score_array_length;
 	int32_t taxid_array_writable_length;
@@ -185,6 +185,7 @@ int build_reference_db(const char* dms_name,
 	matrix_view_name = strcpy(matrix_view_name, o_view_name);
 	strcat(matrix_view_name, "_matrix");
 
+	fprintf(stderr, "Aligning queries with reference database...\n");
 	if (obi_lcs_align_one_column(dms_name,
 								 refs_view_name,
 								 "",
@@ -242,6 +243,7 @@ int build_reference_db(const char* dms_name,
 							false,
 							false,
 							false,
+							false,
 							"",
 							"",
 							-1,
@@ -320,13 +322,19 @@ int build_reference_db(const char* dms_name,
 		return -1;
 	}
 
+	count = (matrix_with_lca_view->infos)->line_count;
+	fprintf(stderr, "Computing LCAs...\n");
+
 	// Compute all the LCAs
 		// For each pair
-	for (i=0; i<(matrix_with_lca_view->infos)->line_count; i++)
+	for (i=0; i<count; i++)
 	{
 		if (! keep_running)
 			return -1;
 
+		if (i%1000 == 0)
+			fprintf(stderr,"\rDone : %f %%       ", (i / (float) count)*100);
+
 		// Read all taxids associated with the first sequence and compute their LCA
 		// Read line index
 		idx1 = obi_get_int_with_elt_idx_and_col_p_in_view(matrix_with_lca_view, matrix_idx1_column, i, 0);
@@ -363,6 +371,7 @@ int build_reference_db(const char* dms_name,
 			return -1;
 		}
 	}
+	fprintf(stderr,"\rDone : 100 %%           \n");
 
 	// Clone refs view, add 2 arrays columns for lca and score, compute and write them
 
@@ -384,6 +393,7 @@ int build_reference_db(const char* dms_name,
 							1,
 							"",
 							false,
+							false,
 							true,
 							false,
 							"",
@@ -407,6 +417,7 @@ int build_reference_db(const char* dms_name,
 							1,
 							"",
 							false,
+							false,
 							true,
 							false,
 							"",
@@ -442,13 +453,18 @@ int build_reference_db(const char* dms_name,
 		return -1;
 	}
 
+	fprintf(stderr, "Building LCA arrays...\n");
+
 	// For each sequence, look for all its alignments in the matrix, and for each different LCA taxid/score, order them and write them
 	// Going through matrix once, filling refs arrays on the go for efficiency
-	for (i=0; i<(matrix_with_lca_view->infos)->line_count; i++)
+	for (i=0; i<count; i++)
 	{
 		if (! keep_running)
 			return -1;
 
+		if (i%1000 == 0)
+			fprintf(stderr,"\rDone : %f %%       ", (i / (float) count)*100);
+
 		// Read ref line indexes
 		idx1 = obi_get_int_with_elt_idx_and_col_p_in_view(matrix_with_lca_view, matrix_idx1_column, i, 0);
 		idx2 = obi_get_int_with_elt_idx_and_col_p_in_view(matrix_with_lca_view, matrix_idx2_column, i, 0);
@@ -464,6 +480,8 @@ int build_reference_db(const char* dms_name,
 		// Read alignment score
 		score = obi_get_float_with_elt_idx_and_col_p_in_view(matrix_with_lca_view, matrix_score_column, i, 0);
 
+		//fprintf(stderr, "\n\ntaxid_lca=%d, score=%f, idx1=%d, idx2=%d", taxid_lca, score, idx1, idx2);
+
 		///////////////// Compute for first sequence \\\\\\\\\\\\\\\\\\\\\\\     (TODO function)
 
 		// Read arrays
@@ -480,9 +498,11 @@ int build_reference_db(const char* dms_name,
 //			return -1;
 //		}
 
+		//fprintf(stderr, "\n1st sequence");
 		// If empty, add values
 		if (taxid_array_length == 0)
 		{
+			//fprintf(stderr, "\nEmpty, add value");
 			if (obi_set_array_with_col_p_in_view(o_view, final_lca_taxid_a_column, idx1, &taxid_lca, (uint8_t) (obi_sizeof(OBI_INT) * 8), 1) < 0)
 			{
 				obidebug(1, "\nError setting a LCA taxid array in a column when building a reference database");
@@ -496,6 +516,8 @@ int build_reference_db(const char* dms_name,
 		}
 		else
 		{
+			//fprintf(stderr, "\nNot empty");
+
 			j = 0;
 			modified = false;
 			while (j < taxid_array_length)
@@ -509,6 +531,9 @@ int build_reference_db(const char* dms_name,
 						memcpy(score_array_writable, score_array, score_array_length*sizeof(obifloat_t));
 						modified = true;
 
+						//fprintf(stderr, "\nSame LCA, replace %d and %f with %d and %f", lca_taxid_array_writable[j],
+						//		score_array_writable[j], taxid_lca, score);
+
 						// Better score for the same LCA, replace this LCA/score pair
 						lca_taxid_array_writable[j] = taxid_lca;
 						score_array_writable[j] = score;
@@ -535,6 +560,8 @@ int build_reference_db(const char* dms_name,
 				{
 					if (score > score_array[j])
 					{
+						//fprintf(stderr, "\nInsert new");
+
 						memcpy(lca_taxid_array_writable, lca_taxid_array, taxid_array_length*sizeof(obiint_t));
 						memcpy(score_array_writable, score_array, score_array_length*sizeof(obifloat_t));
 						modified = true;
@@ -579,10 +606,15 @@ int build_reference_db(const char* dms_name,
 				memcpy(score_array_writable, score_array, score_array_length*sizeof(obifloat_t));
 				modified = true;
 
+				//fprintf(stderr, "\nAppend at the end");
+
 				// Append LCA
 				lca_taxid_array_writable[taxid_array_writable_length] = taxid_lca;
 				score_array_writable[score_array_writable_length] = score;
 
+				taxid_array_writable_length++;
+				score_array_writable_length++;
+
 				// Remove the previous (children) LCAs from the array if their score is equal or lower
 				while ((j>0) && (score_array_writable[j-1] <= score))
 				{
@@ -603,6 +635,13 @@ int build_reference_db(const char* dms_name,
 			// Write new arrays
 			if (modified)
 			{
+//				fprintf(stderr, "\n\nnew array:");
+//				for (k=0;k<taxid_array_writable_length;k++)
+//				{
+//					lca = obi_taxo_get_taxon_with_taxid(tax, lca_taxid_array_writable[k]);
+//					fprintf(stderr, "\nLCA=%d, %s, score=%f", lca_taxid_array_writable[k], lca->name, score_array_writable[k]);
+//				}
+
 				if (obi_set_array_with_col_p_in_view(o_view, final_lca_taxid_a_column, idx1, lca_taxid_array_writable, (uint8_t) (obi_sizeof(OBI_INT) * 8), taxid_array_writable_length) < 0)
 				{
 					obidebug(1, "\nError setting a LCA taxid array in a column when building a reference database");
@@ -632,9 +671,13 @@ int build_reference_db(const char* dms_name,
 //			return -1;
 //		}
 
+		//fprintf(stderr, "\n2nd sequence");
+
 		// If empty, add values
 		if (taxid_array_length == 0)
 		{
+			//fprintf(stderr, "\nEmpty, add value");
+
 			if (obi_set_array_with_col_p_in_view(o_view, final_lca_taxid_a_column, idx2, &taxid_lca, (uint8_t) (obi_sizeof(OBI_INT) * 8), 1) < 0)
 			{
 				obidebug(1, "\nError setting a LCA taxid array in a column when building a reference database");
@@ -648,6 +691,8 @@ int build_reference_db(const char* dms_name,
 		}
 		else
 		{
+			//fprintf(stderr, "\nNot empty");
+
 			j = 0;
 			modified = false;
 			while (j < taxid_array_length)
@@ -661,6 +706,9 @@ int build_reference_db(const char* dms_name,
 						memcpy(score_array_writable, score_array, score_array_length*sizeof(obifloat_t));
 						modified = true;
 
+						//fprintf(stderr, "\nSame LCA, replace %d and %f with %d and %f", lca_taxid_array_writable[j],
+						//		score_array_writable[j], taxid_lca, score);
+
 						// Better score for the same LCA, replace this LCA/score pair
 						lca_taxid_array_writable[j] = taxid_lca;
 						score_array_writable[j] = score;
@@ -687,6 +735,8 @@ int build_reference_db(const char* dms_name,
 				{
 					if (score > score_array[j])
 					{
+						//fprintf(stderr, "\nInsert new");
+
 						memcpy(lca_taxid_array_writable, lca_taxid_array, taxid_array_length*sizeof(obiint_t));
 						memcpy(score_array_writable, score_array, score_array_length*sizeof(obifloat_t));
 						modified = true;
@@ -727,6 +777,8 @@ int build_reference_db(const char* dms_name,
 
 			if (j == taxid_array_length) // same or parent LCA not found, need to be appended at the end
 			{
+				//fprintf(stderr, "\nAppend at the end");
+
 				memcpy(lca_taxid_array_writable, lca_taxid_array, taxid_array_length*sizeof(obiint_t));
 				memcpy(score_array_writable, score_array, score_array_length*sizeof(obifloat_t));
 				modified = true;
@@ -735,6 +787,9 @@ int build_reference_db(const char* dms_name,
 				lca_taxid_array_writable[taxid_array_writable_length] = taxid_lca;
 				score_array_writable[score_array_writable_length] = score;
 
+				taxid_array_writable_length++;
+				score_array_writable_length++;
+
 				// Remove the previous (children) LCAs from the array if their score is equal or lower
 				while ((j>0) && (score_array_writable[j-1] <= score))
 				{
@@ -769,11 +824,17 @@ int build_reference_db(const char* dms_name,
 			}
 		}
 	}
+	fprintf(stderr,"\rDone : 100 %%           \n");
 
+	fprintf(stderr, "Writing results...\n");
+	count = (o_view->infos)->line_count;
 	// Fill empty LCA informations (because filling from potentially sparse alignment matrix) with the sequence taxid
 	score=1.0;  // technically getting LCA of identical sequences
-	for (i=0; i<(o_view->infos)->line_count; i++)
+	for (i=0; i<count; i++)
 	{
+		if (i%1000 == 0)
+			fprintf(stderr,"\rDone : %f %%       ", (i / (float) count)*100);
+
 		obi_get_array_with_col_p_in_view(o_view, final_lca_taxid_a_column, i, &taxid_array_length);
 		if (taxid_array_length == 0)  // no LCA set
 		{
@@ -799,6 +860,7 @@ int build_reference_db(const char* dms_name,
 			}
 		}
 	}
+	fprintf(stderr,"\rDone : 100 %%           \n");
 
 	// Add information about the threshold used to build the DB
 	snprintf(threshold_str, 5, "%f", threshold);
@@ -858,7 +920,6 @@ int build_reference_db(const char* dms_name,
 	free(matrix_view_name);
 	free(matrix_with_lca_view_name);
 
-	fprintf(stderr,"\rDone : 100 %%           \n");
 	return 0;
 }
 

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
- *    C or c : 01
- *    T or t : 10
- *    G or g : 11
+ *    A or a :           00
+ *    C or c :           01
+ *    T or t or U or u : 10
+ *    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
- *      C or c : 0010
- *      G or g : 0011
- *      T or t : 0100
- *      R or r : 0101
- *      Y or y : 0110
- *      S or s : 0111
- *      W or w : 1000
- *      K or k : 1001
- *      M or m : 1010
- *      B or b : 1011
- *      D or d : 1100
- *      H or h : 1101
- *      V or v : 1110
- *      N or n : 1111
+ *		A or a :           0001
+ *      C or c :           0010
+ *      G or g :           0011
+ *      T or t or U or u : 0100
+ *      R or r :           0101
+ *      Y or y :           0110
+ *      S or s :           0111
+ *      W or w :           1000
+ *      K or k :           1001
+ *      M or m :           1010
+ *      B or b :           1011
+ *      D or d :           1100
+ *      H or h :           1101
+ *      V or v :           1110
+ *      N or n :           1111
  *
  * @warning The DNA sequence must contain only IUPAC characters.
+ * @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/kmer_similarity.c

@@ -413,7 +413,13 @@ Obi_ali_p kmer_similarity(Obiview_p view1, OBIDMS_column_p column1, index_t idx1
 	 	return NULL;
 	}
 
-	score = max_common_kmers + kmer_size - 1;  // aka the number of nucleotides in the longest stretch of kmers perfectly matching
+	if (max_common_kmers > 0)
+		score = max_common_kmers + kmer_size - 1; // aka an approximation of the number of nucleotides matching in the overlap of the alignment.
+												  // It's an approximation because one mismatch produces kmer_size kmer mismatches if in the middle of the overlap,
+												  // and less for mismatches located towards the ends of the overlap. The case where there are the most mismatches is assumed,
+												  // meaning that the score will be often underestimated and never overestimated.
+	else
+		score = 0;
 	abs_shift = abs(best_shift);
 
 	// Save result in Obi_ali structure
@@ -423,10 +429,15 @@ Obi_ali_p kmer_similarity(Obiview_p view1, OBIDMS_column_p column1, index_t idx1
 	ali->shift = abs_shift;
 	ali->consensus_seq = NULL;
 	ali->consensus_qual = NULL;
-	if (((best_shift <= 0) && (!switched_seqs)) || ((best_shift > 0) && switched_seqs))
-		strcpy(ali->direction, "left");
+	if (score == 0)
+		ali->direction[0] = '\0';
 	else
-		strcpy(ali->direction, "right");
+	{
+		if (((best_shift <= 0) && (!switched_seqs)) || ((best_shift > 0) && switched_seqs))
+			strcpy(ali->direction, "left");
+		else
+			strcpy(ali->direction, "right");
+	}
 
 	// Build the consensus sequence if asked
 	if (build_consensus)

src/kmer_similarity.h

@@ -27,7 +27,11 @@
  * @brief Alignment structure, with informations about the similarity and to rebuild the alignment.
  */
 typedef struct Obi_ali {
-	int      score;	    		/**< Alignment score, corresponding to the number of nucleotides in the longest stretch of kmers perfectly matching.
+	int      score;	    		/**< Alignment score, corresponding to an approximation of the number of
+								 *   nucleotides matching in the overlap of the alignment.
+								 *   It's an approximation because one mismatch produces kmer_size kmer mismatches if in the middle of the overlap,
+							     *   and less for mismatches located towards the ends of the overlap. The case where there are the most mismatches is assumed,
+							     *   meaning that the score will be often underestimated and never overestimated.
 	 	 	 	 	 	 	 	 */
 	int      consensus_length; 	/**< Length of the final consensus sequence.
 	 	 	 	 	 	 	 	 */

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;
@@ -246,7 +246,16 @@ 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))
-		sample_column = NULL;
+	{
+		fprintf(stderr, "Info: No sample information provided, assuming one sample.\n");
+		sample_column = obi_view_get_column(i_view, COUNT_COLUMN);
+		if (sample_column == NULL)
+		{
+			obidebug(1, "\nError getting the COUNT column");
+			return -1;
+		}
+		sample_count = 1;
+	}
 	else
 	{
 		sample_column = obi_view_get_column(i_view, sample_column_name);
@@ -255,6 +264,13 @@ int obi_clean(const char* dms_name,
 			obidebug(1, "\nError getting the sample column");
 			return -1;
 		}
+		sample_count = (sample_column->header)->nb_elements_per_line;
+		// Check that the sample column is a merged column with all sample informations
+		if (sample_count == 1)
+		{
+			obidebug(1, "\n\nError: If a sample column is provided, it must contain 'merged' sample counts as built by obi uniq with the -m option\n");
+			return -1;
+		}
 	}
 
 	// Create the output view, or a temporary one if heads only
@@ -279,8 +295,6 @@ int obi_clean(const char* dms_name,
 		return -1;
 	}
 
-	sample_count = (sample_column->header)->nb_elements_per_line;
-
 	// Create the output columns
 	if (create_output_columns(o_view, sample_column, sample_count) < 0)
 	{
@@ -549,7 +563,7 @@ int obi_clean(const char* dms_name,
 
 	if (heads_only)
 	{
-		line_selection = malloc((o_view->infos)->line_count * sizeof(index_t));
+		line_selection = malloc((((o_view->infos)->line_count) + 1) * sizeof(index_t));
 		if (line_selection == NULL)
 		{
 			obi_set_errno(OBI_MALLOC_ERROR);

src/obi_clean.h

@@ -52,7 +52,8 @@
  *
  * @param dms A pointer on an OBIDMS.
  * @param i_view_name The name of the input view.
- * @param sample_column_name The name of the OBI_STR column in the input view where the sample information is kept.
+ * @param sample_column_name The name of the column in the input view where the sample information is kept.
+ *                           Must be merged informations as built by the obi uniq tool (checked by the function).
  *                           NULL or "" (empty string) if there is no sample information.
  * @param o_view_name The name of the output view where the results should be written (should not already exist).
  * @param o_view_comments The comments that should be associated with the output view.

src/obi_ecopcr.c

@@ -105,7 +105,8 @@ static int create_output_columns(Obiview_p o_view, bool kingdom_mode);
  * @param o_temp1_column A pointer on the output column for the temperature for the first primer.
  * @param o_temp2_column A pointer on the output column for the temperature for the second primer.
  *
- * @retval 0 if the operation was successfully completed.
+ * @retval 0 if the sequence was skipped (taxid not found, warning printed).
+ * @retval 1 if the sequence was successfully printed to the output.
  * @retval -1 if an error occurred.
  *
  * @since July 2018
@@ -149,49 +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;
@@ -200,7 +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;
@@ -209,7 +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;
@@ -217,28 +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;
@@ -247,7 +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;
@@ -256,7 +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;
@@ -264,49 +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;
@@ -366,6 +367,17 @@ static int print_seq(Obiview_p i_view, Obiview_p o_view,
 
 	// TODO add check for primer longer than MAX_PAT_LEN (32)
 
+	// Get sequence id
+	seq_id = obi_get_str_with_elt_idx_and_col_p_in_view(i_view, i_id_column, i_idx, 0);
+
+	// Get the taxon structure
+	main_taxon = obi_taxo_get_taxon_with_taxid(taxonomy, taxid);
+	if (main_taxon == NULL)
+	{
+		obidebug(1, "\nWarning: error reading the taxonomic information of a sequence. Seq id: %s, taxid: %d. Probably deprecated taxid. Skipping this sequence.", seq_id, taxid);
+		return 0;
+	}
+
 	ldelta = (pos1 <= keep_nucleotides)?pos1:keep_nucleotides;
 	rdelta = ((pos2+keep_nucleotides)>=seq_len)?seq_len-pos2:keep_nucleotides;
 
@@ -431,16 +443,7 @@ static int print_seq(Obiview_p i_view, Obiview_p o_view,
 	if (isnan(tm2))
 		tm2 = OBIFloat_NA;
 
-	// Get the taxon structure
-	main_taxon = obi_taxo_get_taxon_with_taxid(taxonomy, taxid);
-	if (main_taxon == NULL)
-	{
-		obidebug(1, "\nError reading the taxonomic information of a sequence");
-		return -1;
-	}
-
 	// Write sequence id
-	seq_id = obi_get_str_with_elt_idx_and_col_p_in_view(i_view, i_id_column, i_idx, 0);
 	if (obi_set_str_with_elt_idx_and_col_p_in_view(o_view, o_id_column, o_idx, 0, seq_id) < 0)
 	{
 		obidebug(1, "\nError writing the sequence id");
@@ -629,7 +632,7 @@ static int print_seq(Obiview_p i_view, Obiview_p o_view,
 		return -1;
 	}
 
-	return 0;
+	return 1;
 }
 
 
@@ -698,6 +701,7 @@ int obi_ecopcr(const char* i_dms_name,
 
 	obiint_t      taxid;
 	char*         sequence;
+	int			  printed;
 
 	SeqPtr        apatseq=NULL;
 	int32_t       o1Hits;
@@ -1057,14 +1061,14 @@ int obi_ecopcr(const char* i_dms_name,
 											length = 0;
 											if (posj > posi)
 												length = posj - posi - o1->patlen - o2->patlen;
-											if (posj < posi)
+											else if (circular > 0)
 												length = posj + apatseq->seqlen - posi - o1->patlen - o2->patlen;
 											if ((length>0) &&	// For when primers touch or overlap
 												(!min_len || (length >= min_len)) &&
 												(!max_len || (length <= max_len)))
 											{
 												// Print the found amplicon
-												if (print_seq(i_view, o_view,
+												printed = print_seq(i_view, o_view,
 														  	  i_idx, o_idx,
 															  taxonomy,
 															  sequence,
@@ -1090,12 +1094,14 @@ int obi_ecopcr(const char* i_dms_name,
 															  o_strand_column,
 															  o_primer1_column, o_primer2_column,
 															  o_error1_column, o_error2_column,
-															  o_temp1_column,  o_temp2_column) < 0)
+															  o_temp1_column,  o_temp2_column);
+												if (printed < 0)
 												{
 													obidebug(1, "\nError writing the ecopcr result");
 													return -1;
 												}
-												o_idx++;
+												else if (printed > 0)
+													o_idx++;
 											}
 										}
 									}
@@ -1145,14 +1151,14 @@ int obi_ecopcr(const char* i_dms_name,
 											length = 0;
 											if (posj > posi)
 												length = posj - posi + 1  - o2->patlen - o1->patlen; /* - o1->patlen : deleted by <EC> (prior to the OBITools3) */
-											if (posj < posi)
+											else if (circular > 0)
 												length = posj + apatseq->seqlen - posi - o1->patlen - o2->patlen;
 											if ((length>0) &&	// For when primers touch or overlap
 												(!min_len || (length >= min_len)) &&
 												(!max_len || (length <= max_len)))
 											{
 												// Print the found amplicon
-												if (print_seq(i_view, o_view,
+												printed = print_seq(i_view, o_view,
 														  	  i_idx, o_idx,
 															  taxonomy,
 															  sequence,
@@ -1178,12 +1184,14 @@ int obi_ecopcr(const char* i_dms_name,
 															  o_strand_column,
 															  o_primer1_column, o_primer2_column,
 															  o_error1_column, o_error2_column,
-															  o_temp1_column,  o_temp2_column) < 0)
+															  o_temp1_column,  o_temp2_column);
+												if (printed < 0)
 												{
 													obidebug(1, "\nError writing the ecopcr result");
 													return -1;
 												}
-												o_idx++;
+												else if (printed > 0)
+													o_idx++;
 											}
 										}
 									}
@@ -1224,7 +1232,7 @@ int obi_ecopcr(const char* i_dms_name,
 		return -1;
 	}
 
-	fprintf(stderr,"\rDone : 100 %%           ");
+	fprintf(stderr,"\rDone : 100 %%           \n");
 	return 0;
 
 	return 0;

src/obi_ecopcr.h

@@ -81,8 +81,8 @@
  * @param o_dms_name The path to the output DMS.
  * @param o_view_name The name of the output view.
  * @param o_view_comments The comments to associate with the output view.
- * @param primer1 The first primer.
- * @param primer2 The second primer.
+ * @param primer1 The first primer, length must be less than or equal to 32 (because of apat lib limitation).
+ * @param primer2 The second primer, length must be less than or equal to 32 (because of apat lib limitation).
  * @param error_max The maximum number of errors allowed per primer for amplification.
  * @param min_len The minimum length of an amplicon.
  * @param max_len The maximum length of an amplicon.

src/obi_ecotag.c

@@ -71,9 +71,12 @@ static int create_output_columns(Obiview_p o_view);
  * @param name The assigned scientific name.
  * @param assigned_status_column A pointer on the column where the assigned status should be written.
  * @param assigned The assigned status (whether the sequence was assigned to a taxon or not).
- * @param best_match_column A pointer on the column where the list of ids of the best matches should be written.
+ * @param best_match_ids_column A pointer on the column where the list of ids of the best matches should be written.
  * @param best_match_ids The list of ids of the best matches as an array of the concatenated ids separated by '\0'.
  * @param best_match_ids_length The total length of the array of ids of best matches.
+ * @param best_match_taxids_column A pointer on the column where the list of taxids of the best matches should be written.
+ * @param best_match_taxids The list of taxids of the best matches as an array of the taxids.
+ * @param best_match_taxids_length The length of the array of taxids of best matches.
  * @param score_column A pointer on the column where the score should be written.
  * @param score The similarity score of the sequence with its best match(es).
  *
@@ -87,7 +90,8 @@ int print_assignment_result(Obiview_p output_view, index_t line,
 							OBIDMS_column_p assigned_taxid_column, int32_t taxid,
 							OBIDMS_column_p assigned_name_column, const char* name,
 							OBIDMS_column_p assigned_status_column, bool assigned,
-							OBIDMS_column_p best_match_column, const char* best_match_ids, int best_match_ids_length,
+							OBIDMS_column_p best_match_ids_column, const char* best_match_ids, int best_match_ids_length,
+							OBIDMS_column_p best_match_taxids_column, const int32_t* best_match_taxids, int best_match_taxids_length,
 							OBIDMS_column_p score_column, double score);
 
 
@@ -100,37 +104,44 @@ 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, ECOTAG_SCORE_COLUMN_NAME, 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, ECOTAG_TAXID_COLUMN_NAME, 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, ECOTAG_NAME_COLUMN_NAME, 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, ECOTAG_STATUS_COLUMN_NAME, 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, ECOTAG_BEST_MATCH_IDS_COLUMN_NAME, 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 the best match in ecotag");
+		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, 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;
 	}
 
@@ -142,7 +153,8 @@ int print_assignment_result(Obiview_p output_view, index_t line,
 							OBIDMS_column_p assigned_taxid_column, int32_t taxid,
 							OBIDMS_column_p assigned_name_column, const char* name,
 							OBIDMS_column_p assigned_status_column, bool assigned,
-							OBIDMS_column_p best_match_column, const char* best_match_ids, int best_match_ids_length,
+							OBIDMS_column_p best_match_ids_column, const char* best_match_ids, int best_match_ids_length,
+							OBIDMS_column_p best_match_taxids_column, const int32_t* best_match_taxids, int best_match_taxids_length,
 							OBIDMS_column_p score_column, double score)
 {
 	// Write the assigned taxid
@@ -167,9 +179,16 @@ int print_assignment_result(Obiview_p output_view, index_t line,
 	}
 
 	// Write the best match ids
-	if (obi_set_array_with_col_p_in_view(output_view, best_match_column, line, best_match_ids, (uint8_t)(sizeof(char)*8), best_match_ids_length) < 0)
+	if (obi_set_array_with_col_p_in_view(output_view, best_match_ids_column, line, best_match_ids, (uint8_t)(sizeof(char)*8), best_match_ids_length) < 0)
 	{
-		obidebug(1, "\nError writing a assignment status in a column when writing ecotag results");
+		obidebug(1, "\nError writing the array of best match ids in a column when writing ecotag results");
+		return -1;
+	}
+
+	// Write the best match taxids
+	if (obi_set_array_with_col_p_in_view(output_view, best_match_taxids_column, line, best_match_taxids, (uint8_t)(sizeof(OBI_INT)*8), best_match_taxids_length) < 0)
+	{
+		obidebug(1, "\nError writing the array of best match taxids in a column when writing ecotag results");
 		return -1;
 	}
 
@@ -199,7 +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
@@ -220,6 +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;
@@ -235,6 +256,8 @@ int obi_ecotag(const char* dms_name,
 	char*     		best_match_ids;
 	char*			best_match_ids_to_store;
 	int32_t         best_match_ids_length;
+	int32_t*   		best_match_taxids;
+	int32_t*		best_match_taxids_to_store;
 	int				best_match_count;
 	int             buffer_size;
 	int 			best_match_ids_buffer_size;
@@ -263,7 +286,8 @@ int obi_ecotag(const char* dms_name,
 	OBIDMS_column_p assigned_taxid_column = NULL;
 	OBIDMS_column_p assigned_name_column = NULL;
 	OBIDMS_column_p assigned_status_column = NULL;
-	OBIDMS_column_p best_match_column = NULL;
+	OBIDMS_column_p best_match_ids_column = NULL;
+	OBIDMS_column_p best_match_taxids_column = NULL;
 	OBIDMS_column_p lca_taxid_a_column = NULL;
 	OBIDMS_column_p score_a_column = NULL;
 	OBIDMS_column_p ref_taxid_column = NULL;
@@ -367,10 +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;
 	}
 
@@ -396,7 +420,8 @@ int obi_ecotag(const char* dms_name,
 	assigned_taxid_column = obi_view_get_column(output_view, ECOTAG_TAXID_COLUMN_NAME);
 	assigned_name_column = obi_view_get_column(output_view, ECOTAG_NAME_COLUMN_NAME);
 	assigned_status_column = obi_view_get_column(output_view, ECOTAG_STATUS_COLUMN_NAME);
-	best_match_column = obi_view_get_column(output_view, ECOTAG_BEST_MATCH_IDS_COLUMN_NAME);
+	best_match_ids_column = obi_view_get_column(output_view, ECOTAG_BEST_MATCH_IDS_COLUMN_NAME);
+	best_match_taxids_column = obi_view_get_column(output_view, ECOTAG_BEST_MATCH_TAXIDS_COLUMN_NAME);
 	score_column = obi_view_get_column(output_view, ECOTAG_SCORE_COLUMN_NAME);
 
 	// Open the used reference columns
@@ -453,9 +478,17 @@ int obi_ecotag(const char* dms_name,
 		return -1;
 	}
 
+	best_match_taxids = (int32_t*) malloc(buffer_size* sizeof(int32_t));
+	if (best_match_taxids == NULL)
+	{
+		obi_set_errno(OBI_MALLOC_ERROR);
+		obidebug(1, "\nError allocating memory for the best match taxid array in ecotag");
+		return -1;
+	}
+
 	for (i=0; i < query_count; i++)
 	{
-		if (i%100 == 0)
+		if (i%1000 == 0)
 			fprintf(stderr,"\rDone : %f %%       ", (i / (float) query_count)*100);
 
 		best_match_count = 0;
@@ -514,7 +547,7 @@ int obi_ecotag(const char* dms_name,