Initial port to python3

[htsworkflow.git] / htsworkflow / submission / submission.py

diff --git a/htsworkflow/submission/submission.py b/htsworkflow/submission/submission.py
index 77a1b68e4db763ab83eb50135f9613e8264ce973..cd2a8df03629296cc2db17dd582a8f072121db04 100644 (file)
--- a/htsworkflow/submission/submission.py
+++ b/htsworkflow/submission/submission.py
@@ -8,16 +8,12 @@ import RDF
 
 from htsworkflow.util.rdfhelp import \
      blankOrUri, \
-     dafTermOntology, \
      dump_model, \
+     fromTypedNode, \
      get_model, \
-     libraryOntology, \
-     owlNS, \
-     rdfNS, \
-     submissionLog, \
-     submissionOntology, \
-     toTypedNode, \
-     fromTypedNode
+     stripNamespace, \
+     toTypedNode
+from htsworkflow.util.rdfns import *
 from htsworkflow.util.hashfile import make_md5sum
 from htsworkflow.submission.fastqname import FastqName
 from htsworkflow.submission.daf import \
@@ -41,11 +37,11 @@ class Submission(object):
     def scan_submission_dirs(self, result_map):
         """Examine files in our result directory
         """
-        for lib_id, result_dir in result_map.items():
+        for lib_id, result_dir in list(result_map.items()):
             LOGGER.info("Importing %s from %s" % (lib_id, result_dir))
             try:
                 self.import_analysis_dir(result_dir, lib_id)
-            except MetadataLookupException, e:
+            except MetadataLookupException as e:
                 LOGGER.error("Skipping %s: %s" % (lib_id, str(e)))
 
     def import_analysis_dir(self, analysis_dir, library_id):
@@ -61,6 +57,13 @@ class Submission(object):
             pathname = os.path.abspath(os.path.join(analysis_dir, filename))
             self.construct_file_attributes(analysis_dir, libNode, pathname)
 
+    def analysis_nodes(self, result_map):
+        """Return an iterable of analysis nodes
+        """
+        for result_dir in list(result_map.values()):
+            an_analysis = self.get_submission_node(result_dir)
+            yield an_analysis
+
     def construct_file_attributes(self, analysis_dir, libNode, pathname):
         """Looking for the best extension
         The 'best' is the longest match
@@ -122,6 +125,11 @@ class Submission(object):
             RDF.Statement(fileNode,
                           rdfNS['type'],
                           file_type))
+        self.model.add_statement(
+            RDF.Statement(fileNode,
+                          libraryOntology['library'],
+                          libNode))
+                          
         LOGGER.debug("Done.")
 
     def make_file_node(self, pathname, submissionNode):
@@ -129,7 +137,8 @@ class Submission(object):
         """
         # add file specific information
         path, filename = os.path.split(pathname)
-        fileNode = RDF.Node(RDF.Uri('file://'+ os.path.abspath(pathname)))
+        pathname = os.path.abspath(pathname)
+        fileNode = RDF.Node(RDF.Uri('file://'+ pathname))
         self.model.add_statement(
             RDF.Statement(submissionNode,
                           dafTermOntology['has_file'],
@@ -138,6 +147,10 @@ class Submission(object):
             RDF.Statement(fileNode,
                           dafTermOntology['filename'],
                           filename))
+        self.model.add_statement(
+            RDF.Statement(fileNode,
+                          dafTermOntology['relative_path'],
+                          os.path.relpath(pathname)))
         return fileNode
 
     def add_md5s(self, filename, fileNode, analysis_dir):
@@ -176,7 +189,12 @@ class Submission(object):
                               libraryOntology['has_mappings'],
                               dafTermOntology['has_file']))
         parser = RDF.Parser(name='rdfa')
-        new_statements = parser.parse_as_stream(libNode.uri)
+        try:
+            new_statements = parser.parse_as_stream(libNode.uri)
+        except RDF.RedlandError as e:
+            LOGGER.error(e)
+            return
+        LOGGER.debug("Scanning %s", str(libNode.uri))
         toadd = []
         for s in new_statements:
             # always add "collections"
@@ -206,7 +224,7 @@ class Submission(object):
             LOGGER.debug("Importing %s" % (lane.uri,))
             try:
                 parser.parse_into_model(self.model, lane.uri)
-            except RDF.RedlandError, e:
+            except RDF.RedlandError as e:
                 LOGGER.error("Error accessing %s" % (lane.uri,))
                 raise e
 
@@ -218,7 +236,7 @@ class Submission(object):
             self.__view_map = self._get_filename_view_map()
 
         results = []
-        for pattern, view in self.__view_map.items():
+        for pattern, view in list(self.__view_map.items()):
             if re.match(pattern, filename):
                 results.append(view)
 
@@ -247,7 +265,7 @@ class Submission(object):
             LOGGER.debug("Found: %s" % (literal_re,))
             try:
                 filename_re = re.compile(literal_re)
-            except re.error, e:
+            except re.error as e:
                 LOGGER.error("Unable to compile: %s" % (literal_re,))
             patterns[literal_re] = view_name
         return patterns
@@ -310,7 +328,8 @@ class Submission(object):
         paired = ['Barcoded Illumina',
                   'Multiplexing',
                   'Nextera',
-                  'Paired End (non-multiplexed)',]
+                  'Paired End (non-multiplexed)',
+                  'Dual Index Illumina',]
         if library_type in single:
             return False
         elif library_type in paired:
@@ -330,7 +349,25 @@ class Submission(object):
         results = []
         for record in rdfstream:
             d = {}
-            for key, value in record.items():
+            for key, value in list(record.items()):
                 d[key] = fromTypedNode(value)
             results.append(d)
         return results
+
+
+def list_submissions(model):
+    """Return generator of submissions in this model.
+    """
+    query_body = """
+      PREFIX subns: <http://jumpgate.caltech.edu/wiki/UcscSubmissionOntology#>
+
+      select distinct ?submission
+      where { ?submission subns:has_submission ?library_dir }
+    """
+    query = RDF.SPARQLQuery(query_body)
+    rdfstream = query.execute(model)
+    for row in rdfstream:
+        s = stripNamespace(submissionLog, row['submission'])
+        if s[-1] in ['#', '/', '?']:
+            s = s[:-1]
+        yield s