switch to version v3.0.1b15

ASCII (issue #89)

python/obitools3/apps/optiongroups/__init__.py

@@ -297,6 +297,29 @@ def __addExportOutputOption(optionManager):
                      const=b'tabular',
                      help="Output file is in tabular format")
 
+    group.add_argument('--metabaR-output',
+                     action="store_const", dest="obi:outputformat",
+                     default=None,
+                     const=b'metabaR',
+                     help="Export the files needed by the obifiles_to_metabarlist function of the metabaR package")
+
+    group.add_argument('--metabaR-prefix',
+                     action="store", dest="obi:metabarprefix",
+                     type=str,
+                     help="Prefix for the files when using --metabaR-output option")
+
+    group.add_argument('--metabaR-ngsfilter',
+                     action="store", dest="obi:metabarngsfilter",
+                     type=str,
+                     default=None,
+                     help="URI to the ngsfilter view when using --metabaR-output option (if not provided, it is not exported)")
+
+    group.add_argument('--metabaR-samples',
+                     action="store", dest="obi:metabarsamples",
+                     type=str,
+                     default=None,
+                     help="URI to the sample metadata view when using --metabaR-output option (if not provided, it is built as just a list of the sample names)")
+
     group.add_argument('--only-keys',
                        action="append", dest="obi:only_keys",
                        type=str,

python/obitools3/commands/export.pyx

@@ -6,6 +6,9 @@ from obitools3.apps.config import logger
 from obitools3.dms import DMS
 from obitools3.dms.obiseq import Nuc_Seq
 from obitools3.dms.capi.obiview cimport QUALITY_COLUMN
+from obitools3.writers.tab import TabWriter
+from obitools3.format.tab import TabFormat
+from obitools3.utils cimport tobytes, tostr
 
 from obitools3.apps.optiongroups import addMinimalInputOption, \
                                         addExportOutputOption, \
@@ -76,6 +79,13 @@ def run(config):
     else:
         pb = ProgressBar(withoutskip - skip, config)
 
+    if config['obi']['outputformat'] == b'metabaR':
+        # Check prefix
+        if "metabarprefix" not in config["obi"]:
+            raise Exception("Prefix needed when exporting for metabaR (--metabaR-prefix option)")
+        else:
+            metabaRprefix = config["obi"]["metabarprefix"]
+        
     i=0
     for seq in iview :
         PyErr_CheckSignals()
@@ -91,6 +101,81 @@ def run(config):
         pb(i, force=True)
         print("", file=sys.stderr)
 
+    if config['obi']['outputformat'] == b'metabaR':
+        
+        # Export ngsfilter file if view provided
+        if 'metabarngsfilter' in config['obi']:
+            ngsfilter_input = open_uri(config['obi']['metabarngsfilter'])
+            if ngsfilter_input is None:
+                raise Exception("Could not read ngsfilter view for metabaR output")
+            ngsfilter_view = ngsfilter_input[1]
+    
+            ngsfilter_output = open(config['obi']['metabarprefix']+'.ngsfilter', 'w')
+            
+            for line in ngsfilter_view:
+
+                line_to_print = b""                
+                line_to_print += line[b'experiment']
+                line_to_print += b"\t"
+                line_to_print += line[b'sample']
+                line_to_print += b"\t"
+                line_to_print += line[b'forward_tag']
+                line_to_print += b":"
+                line_to_print += line[b'reverse_tag']
+                line_to_print += b"\t"
+                line_to_print += line[b'forward_primer']
+                line_to_print += b"\t"
+                line_to_print += line[b'reverse_primer']
+                line_to_print += b"\t"
+                line_to_print += line[b'additional_info']
+                
+                print(tostr(line_to_print), file=ngsfilter_output)
+            
+            if ngsfilter_input[0] != input[0]:
+                ngsfilter_input[0].close()
+            ngsfilter_output.close()
+            
+        # Export sample metadata
+        samples_output = open(config['obi']['metabarprefix']+'_samples.csv', 'w')
+
+        # Export sample metadata file if view provided
+        if 'metabarsamples' in config['obi']:
+            samples_input = open_uri(config['obi']['metabarsamples'])
+            if samples_input is None:
+                raise Exception("Could not read sample view for metabaR output")
+            samples_view = samples_input[1]
+        
+            # Export with tab formatter
+            TabWriter(TabFormat(header=True, sep='\t',), 
+                      samples_output,
+                      header=True)
+            
+            if samples_input[0] != input[0]:
+                samples_input[0].close()
+        
+        # Else export just sample names from main view
+        else:
+            
+            sample_list = []
+            if 'MERGED_sample' in iview:
+                sample_list = iview['MERGED_sample'].keys()
+            elif 'sample' not in iview:
+                for seq in iview:
+                    sample = seq['sample']
+                    if sample not in sample_list:
+                        sample_list.append(sample)
+            else:
+                logger("warning", "Can not read sample list from main view for metabaR sample list export")            
+            
+            print("sample_id", file=samples_output)
+            for sample in sample_list:
+                line_to_print = b""                
+                line_to_print += sample
+                line_to_print += b"\t"
+                print(tostr(line_to_print), file=samples_output)
+        
+        samples_output.close()
+
     # TODO save command in input dms?
     
     if not BrokenPipeError and not IOError:

python/obitools3/commands/grep.pyx

@@ -91,7 +91,7 @@ def addOptions(parser):
                        metavar="<ATTRIBUTE_NAME>",
                        help="Select records with the attribute <ATTRIBUTE_NAME> "
                             "defined (not set to NA value). "
-                            "Several -a options can be used on the same "
+                            "Several -A options can be used on the same "
                             "command line.")
     
     group.add_argument("-L", "--lmax",

python/obitools3/commands/split.pyx

@@ -0,0 +1,105 @@
+#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.apps.config import logger
+from obitools3.utils cimport tobytes
+
+import sys
+from cpython.exc cimport PyErr_CheckSignals
+
+ 
+__title__="Split"
+
+ 
+def addOptions(parser):
+    
+    addMinimalInputOption(parser)
+    addNoProgressBarOption(parser)
+
+    group=parser.add_argument_group("obi split specific options")
+ 
+    group.add_argument('-p','--prefix',
+                       action="store", dest="split:prefix",
+                       metavar="<PREFIX>",
+                       help="Prefix added to each subview name (included undefined)")
+
+    group.add_argument('-t','--tag-name',
+                       action="store", dest="split:tagname",
+                       metavar="<TAG_NAME>",
+                       help="Attribute/tag used to split the input")
+      
+    group.add_argument('-u','--undefined',
+                       action="store", dest="split:undefined",
+                       default=b'UNDEFINED',
+                       metavar="<VIEW_NAME>",
+                       help="Name of the view where undefined sequenced are stored (will be PREFIX_VIEW_NAME)")
+
+
+def run(config):
+     
+    DMS.obi_atexit()
+    
+    logger("info", "obi split")
+
+    # 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]
+
+    # Initialize the progress bar
+    if config['obi']['noprogressbar'] == False:
+        pb = ProgressBar(len(i_view), config)
+    else:
+        pb = None
+    
+    tag_to_split = config["split"]["tagname"]
+    undefined = tobytes(config["split"]["undefined"])
+    selections = {}
+    
+    # Go through input view and split
+    for i in range(len(i_view)):
+        PyErr_CheckSignals()
+        if pb is not None:
+            pb(i)
+        line = i_view[i]
+        if tag_to_split not in line or line[tag_to_split] is None or len(line[tag_to_split])==0:
+            value = undefined
+        else:
+            value = line[tag_to_split]
+        if value not in selections:
+            selections[value] = Line_selection(i_view)
+        selections[value].append(i)
+        
+    if pb is not None:
+        pb(len(i_view), force=True)
+        print("", file=sys.stderr)
+
+    # Create output views with the line selection
+    try:
+        for cat in selections:
+            o_view_name = config["split"]["prefix"].encode()+cat
+            o_view = selections[cat].materialize(o_view_name)
+            # Save command config in View and DMS comments
+            command_line = " ".join(sys.argv[1:])
+            input_dms_name=[input[0].name]
+            input_view_name=[input[1].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, "split", command_line, input_dms_name=input_dms_name, input_view_name=input_view_name)
+            o_view.close()
+    except Exception, e:
+        raise RollbackException("obi split error, rollbacking view: "+str(e), o_view)
+
+    i_dms.record_command_line(command_line)
+    i_dms.close(force=True)
+
+    logger("info", "Done.")
+

python/obitools3/dms/view/view.pyx

@@ -799,7 +799,8 @@ cdef class Line :
  
  
     def keys(self):
-        return self._view.keys()
+        cdef bytes key
+        return [key for key in self._view.keys()]
  
  
     def __contains__(self, object column_name):

python/obitools3/format/tab.pxd

@@ -6,4 +6,6 @@ cdef class TabFormat:
     cdef bytes NAString
     cdef set   tags
     cdef bytes sep
-    cdef bint NAIntTo0
+    cdef bint NAIntTo0
+    cdef bint metabaR
+    cdef bint ngsfilter

python/obitools3/format/tab.pyx

@@ -10,13 +10,15 @@ import sys
 
 cdef class TabFormat:
     
-    def __init__(self, list tags=[], header=True, bytes NAString=b"NA", bytes sep=b"\t", bint NAIntTo0=True):
+    def __init__(self, list tags=[], header=True, bytes NAString=b"NA", bytes sep=b"\t", bint NAIntTo0=True, metabaR=False, ngsfilter=False):
         self.tags = set(tags)
         self.header = header
         self.first_line = True
         self.NAString = NAString
         self.sep = sep
         self.NAIntTo0 = NAIntTo0
+        self.metabaR = metabaR
+        self.ngsfilter = ngsfilter
         
     @cython.boundscheck(False)    
     def __call__(self, object data):
@@ -34,13 +36,21 @@ cdef class TabFormat:
         if self.header and self.first_line:
             for k in ktags:
                 if k in tags:
+                    if self.metabaR:
+                        if k == b'NUC_SEQ':
+                            ktoprint = b'sequence'
+                        else:
+                            ktoprint = k.lower()
+                        ktoprint = ktoprint.replace(b'merged_', b'')
+                    else:
+                        ktoprint = k
                     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))
+                            line.append(tobytes(ktoprint)+b':'+tobytes(k2))
                     else:
-                        line.append(tobytes(k))
+                        line.append(tobytes(ktoprint))
             r = self.sep.join(value for value in line)
             r += b'\n'
             line = []

python/obitools3/parsers/genbank.pyx

@@ -22,11 +22,11 @@ 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(\s*))',re.DOTALL + re.M)
 
 _headerMatcher = re.compile(b'^LOCUS.+(?=\nFEATURES)', re.DOTALL + re.M)
-_seqMatcher    = re.compile(b'^ORIGIN .+(?=//\n)', re.DOTALL + re.M)
+_seqMatcher    = re.compile(b'^ORIGIN.+(?=//\n)', re.DOTALL + re.M)
-_cleanSeq1     = re.compile(b'ORIGIN.+\n')
+_cleanSeq1     = re.compile(b'ORIGIN(\s*)\n')
 _cleanSeq2     = re.compile(b'[ \n0-9]+')
 _acMatcher     = re.compile(b'(?<=^ACCESSION   ).+',re.M)
 _deMatcher     = re.compile(b'(?<=^DEFINITION  ).+\n( .+\n)*',re.M)
@@ -155,10 +155,10 @@ def genbankIterator_file(lineiterator,
         yield seq
         read+=1
     
-    # Last sequence
+    # Last sequence if not empty lines
-    seq = genbankParser(entry)
+    if entry.strip():
-    
+        seq = genbankParser(entry)
-    yield seq
+        yield seq
     
     free(entry)
 

python/obitools3/parsers/ngsfilter.pyx

@@ -48,13 +48,13 @@ def ngsfilterIterator(lineiterator,
         all_lines.insert(0, firstline)
         
     # Insert header for column names
-    column_names = [b"experiment", b"sample", b"forward_tag", b"reverse_tag", b"forward_primer", b"reverse_primer"]
+    column_names = [b"experiment", b"sample", b"forward_tag", b"reverse_tag", b"forward_primer", b"reverse_primer",b"additional_info"]
     header = out_sep.join(column_names)
     
     new_lines.append(header)
         
     for line in all_lines:
-        split_line = line.split()
+        split_line = line.split(maxsplit=5)
         tags = split_line.pop(2)
         tags = tags.split(b":")
         for t_idx in range(len(tags)):
@@ -64,7 +64,7 @@ def ngsfilterIterator(lineiterator,
             tags.append(tags[0])
         split_line.insert(2, tags[0])
         split_line.insert(3, tags[1])
-        new_lines.append(out_sep.join(split_line[0:6]))
+        new_lines.append(out_sep.join(split_line[0:7]))
     
     return tabIterator(iter(new_lines),
                        header = True,

python/obitools3/uri/decode.pyx

@@ -173,7 +173,10 @@ def open_uri(uri,
              type newviewtype=View,
              dms_only=False,
              force_file=False):
-    
+
+    if type(uri) == str and not uri.isascii():
+        raise Exception("Paths must be ASCII characters only")
+            
     cdef bytes urib = tobytes(uri)
     cdef bytes scheme
     cdef tuple dms
@@ -277,7 +280,12 @@ def open_uri(uri,
             iseq = urib
             objclass = bytes
     else:  # TODO update uopen to be able to write? 
-        if not urip.path or urip.path == b'-':
+        if config['obi']['outputformat'] == b'metabaR':
+            if 'metabarprefix' not in config['obi']:
+                raise Exception("Prefix needed when exporting for metabaR (--metabaR-prefix option)")
+            else:
+                file = open(config['obi']['metabarprefix']+'.tab', 'wb')
+        elif not urip.path or urip.path == b'-':
             file = sys.stdout.buffer
         else:
             file = open(urip.path, 'wb')
@@ -299,11 +307,11 @@ def open_uri(uri,
                     format=config["obi"][formatkey]
                 except KeyError:
                     format=None
-                
+                       
         if b'seqtype' in qualifiers:
             seqtype=qualifiers[b'seqtype'][0]
         else:
-            if format == b"ngsfilter" or format == b"tabular":  # TODO discuss
+            if format == b"ngsfilter" or format == b"tabular" or format == b"metabaR":  # TODO discuss
                 seqtype=None
             else:
                 try:
@@ -437,7 +445,7 @@ def open_uri(uri,
             try:
                 na_int_to_0=config["obi"]["na_int_to_0"]
             except KeyError:
-                if format==b"tabular":
+                if format==b"tabular" or format==b"metabaR":
                     na_int_to_0=True
                 else:
                     na_int_to_0=False
@@ -487,6 +495,13 @@ def open_uri(uri,
             except KeyError:
                 only_keys=[]
 
+        if b"metabaR_prefix" in qualifiers:
+            metabaR_prefix = tobytes(qualifiers[b"metabaR_prefix"][0][0])
+        else:
+            try:
+                metabaR_prefix = tobytes(config["obi"]["metabarprefix"])
+            except KeyError:
+                metabaR_prefix=None
 
         if format is not None:
             if seqtype==b"nuc":
@@ -552,6 +567,16 @@ def open_uri(uri,
                                                skip=skip,
                                                only=only,
                                                header=header)
+            elif format==b"metabaR":
+                objclass = dict
+                if input:
+                    raise NotImplementedError('Input data file format not implemented')
+                else:
+                    iseq = TabWriter(TabFormat(tags=only_keys, header=header, NAString=nastring, sep=sep, NAIntTo0=na_int_to_0, metabaR=True), 
+                                               file,
+                                               skip=skip,
+                                               only=only,
+                                               header=header)
             elif format==b"ngsfilter":
                 objclass = dict
                 if input:
@@ -565,7 +590,7 @@ def open_uri(uri,
                                              skip = skip,
                                              only = only)
                 else:
-                    raise NotImplementedError('Output sequence file format not implemented')
+                    raise NotImplementedError('Output data file format not implemented')
         else:
             if input:
                 iseq, objclass, format = entryIteratorFactory(file,

python/obitools3/utils.pyx

@@ -264,7 +264,7 @@ cdef obitype_t update_obitype(obitype_t obitype, object new_value) :
     if new_value == None or new_type==list or new_type==dict or new_type==tuple:
         return obitype
     
-    # TODO BOOL vers INT/FLOAT
+    # TODO BOOL to INT/FLOAT
     if new_type == str or new_type == bytes :
         if obitype == OBI_SEQ and is_a_DNA_seq(tobytes(new_value)) :
             pass

python/obitools3/version.py

@@ -1,5 +1,5 @@
 major = 3
 minor = 0
-serial= '1b12'
+serial= '1b15'
 
 version ="%d.%d.%s" % (major,minor,serial)

src/build_reference_db.c

@@ -77,6 +77,7 @@ static inline ecotx_t* get_lca_from_merged_taxids(Obiview_p view, OBIDMS_column_
 {
 	ecotx_t* taxon = NULL;
 	ecotx_t* lca = NULL;
+	ecotx_t* lca1 = NULL;
 	int32_t taxid;
 	index_t taxid_idx;
 	int64_t taxid_str_idx;
@@ -108,10 +109,11 @@ static inline ecotx_t* get_lca_from_merged_taxids(Obiview_p view, OBIDMS_column_
 			else
 			{
 				// Compute LCA
+				lca1 = lca;
 				lca = obi_taxo_get_lca(taxon, lca);
 				if (lca == NULL)
 				{
-					obidebug(1, "\nError getting the last common ancestor of two taxa when building a reference database");
+					obidebug(1, "\nError getting the last common ancestor of two taxa when building a reference database, %d %d", taxid, lca1->taxid);
 					return NULL;
 				}
 			}
@@ -185,7 +187,7 @@ 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");
+	fprintf(stderr, "Aligning sequences...\n");
 	if (obi_lcs_align_one_column(dms_name,
 								 refs_view_name,
 								 "",