diff --git a/datatypes_conf.xml.sample b/datatypes_conf.xml.sample index ace054e4c90..72c5515ab80 100644 --- a/datatypes_conf.xml.sample +++ b/datatypes_conf.xml.sample @@ -135,6 +135,7 @@ + diff --git a/lib/galaxy/datatypes/converters/__init__.py b/lib/galaxy/datatypes/converters/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/lib/galaxy/datatypes/converters/fasta_to_2bit.xml b/lib/galaxy/datatypes/converters/fasta_to_2bit.xml new file mode 100644 index 00000000000..b4890808a13 --- /dev/null +++ b/lib/galaxy/datatypes/converters/fasta_to_2bit.xml @@ -0,0 +1,13 @@ + + + + faToTwoBit $input $output + + + + + + + + + diff --git a/lib/galaxy/datatypes/converters/fasta_to_len.py b/lib/galaxy/datatypes/converters/fasta_to_len.py new file mode 100644 index 00000000000..26f0856bd4e --- /dev/null +++ b/lib/galaxy/datatypes/converters/fasta_to_len.py @@ -0,0 +1,48 @@ +#!/usr/bin/env python +""" +Input: fasta, int +Output: tabular +Return titles with lengths of corresponding seq +""" + +import sys, os + +assert sys.version_info[:2] >= ( 2, 4 ) + +def compute_fasta_length( fasta_file, out_file, keep_first_char ): + + infile = fasta_file + out = open( out_file, 'w') + keep_first_char = int( keep_first_char ) + + fasta_title = '' + seq_len = 0 + + # number of char to keep in the title + if keep_first_char == 0: + keep_first_char = None + else: + keep_first_char += 1 + + first_entry = True + + for line in open( infile ): + line = line.strip() + if not line or line.startswith( '#' ): + continue + if line[0] == '>': + if first_entry == False: + out.write( "%s\t%d\n" % ( fasta_title[ 1:keep_first_char ], seq_len ) ) + else: + first_entry = False + fasta_title = line + seq_len = 0 + else: + seq_len += len(line) + + # last fasta-entry + out.write( "%s\t%d\n" % ( fasta_title[ 1:keep_first_char ], seq_len ) ) + out.close() + +if __name__ == "__main__" : + compute_fasta_length( sys.argv[1], sys.argv[2], sys.argv[3] ) \ No newline at end of file diff --git a/lib/galaxy/datatypes/converters/fasta_to_len.xml b/lib/galaxy/datatypes/converters/fasta_to_len.xml new file mode 100644 index 00000000000..8e1a1a0b966 --- /dev/null +++ b/lib/galaxy/datatypes/converters/fasta_to_len.xml @@ -0,0 +1,13 @@ + + + + fasta_to_len.py $input $output 0 + + + + + + + + + diff --git a/lib/galaxy/datatypes/converters/len_to_linecount.xml b/lib/galaxy/datatypes/converters/len_to_linecount.xml new file mode 100644 index 00000000000..9a91d8723b8 --- /dev/null +++ b/lib/galaxy/datatypes/converters/len_to_linecount.xml @@ -0,0 +1,13 @@ + + + + wc -l $input | awk '{print $1}' > $output + + + + + + + + + diff --git a/lib/galaxy/datatypes/data.py b/lib/galaxy/datatypes/data.py index 4e9874a299e..2149ef3c04b 100644 --- a/lib/galaxy/datatypes/data.py +++ b/lib/galaxy/datatypes/data.py @@ -446,6 +446,13 @@ class Text( Data ): dataset.peek = 'file does not exist' dataset.blurb = 'file purged from disk' +class LineCount( Text ): + """ + Dataset contains a single line with a single integer that denotes the + line count for a related dataset. Used for custom builds. + """ + pass + class Newick( Text ): pass diff --git a/lib/galaxy/web/controllers/user.py b/lib/galaxy/web/controllers/user.py index ce7b68f91c6..fe24d9f041e 100644 --- a/lib/galaxy/web/controllers/user.py +++ b/lib/galaxy/web/controllers/user.py @@ -1165,6 +1165,9 @@ class User( BaseController, UsesFormDefinitions ): @web.expose @web.require_login() def dbkeys( self, trans, **kwds ): + # + # Process arguments and add/delete build. + # user = trans.user message = None lines_skipped = 0 @@ -1178,57 +1181,79 @@ class User( BaseController, UsesFormDefinitions ): else: dbkeys = from_json_string(user.preferences['dbkeys']) if 'delete' in kwds: + # Delete a build. key = kwds.get('key', '') if key and key in dbkeys: del dbkeys[key] elif 'add' in kwds: - name = kwds.get('name', '') - key = kwds.get('key', '') - len_file = kwds.get('len_file', None) - if getattr(len_file, "file", None): # Check if it's a FieldStorage object - len_text = len_file.file.read() - else: - len_text = kwds.get('len_text', '') - if not name or not key or not len_text: + # Add new custom build. + name = kwds.get('name', '') + key = kwds.get('key', '') + dataset_id = kwds.get('dataset_id', '') + if not name or not key or not dataset_id: message = "You must specify values for all the fields." elif key in dbkeys: message = "There is already a custom build with that key. Delete it first if you want to replace it." else: - # Create new len file - new_len = trans.app.model.HistoryDatasetAssociation( extension="len", create_dataset=True, sa_session=trans.sa_session ) - trans.sa_session.add( new_len ) - new_len.name = name - new_len.visible = False - new_len.state = trans.app.model.Job.states.OK - new_len.info = "custom build .len file" - trans.sa_session.flush() - counter = 0 - f = open(new_len.file_name, "w") - # LEN files have format: - # - for line in len_text.split("\n"): - lst = line.strip().rsplit(None, 1) # Splits at the last whitespace in the line - if not lst or len(lst) < 2: - lines_skipped += 1 - continue - chrom, length = lst[0], lst[1] - try: - length = int(length) - except ValueError: - lines_skipped += 1 - continue - counter += 1 - f.write("%s\t%s\n" % (chrom, length)) - f.close() - dbkeys[key] = { "name": name, "len": new_len.id, "count": counter } + dataset_id = trans.security.decode_id( dataset_id ) + dbkeys[key] = { "name": name, "fasta": dataset_id } + # Save builds. + # TODO: use database table to save builds. user.preferences['dbkeys'] = to_json_string(dbkeys) trans.sa_session.flush() + + # + # Display custom builds page. + # + + # Add chrom/contig count to dbkeys dict. + updated = False + for key, attributes in dbkeys.items(): + if 'count' in attributes: + # Already have count, so do nothing. + continue + + # Get len file. + fasta_dataset = trans.app.model.HistoryDatasetAssociation.get( attributes[ 'fasta' ] ) + len_dataset = fasta_dataset.get_converted_dataset( trans, "len" ) + # HACK: need to request dataset again b/c get_converted_dataset() + # doesn't return dataset (as it probably should). + len_dataset = fasta_dataset.get_converted_dataset( trans, "len" ) + if len_dataset.state == trans.app.model.Job.states.ERROR: + # Can't use len dataset. + continue + + # Get chrom count file. + # NOTE: this conversion doesn't work well with set_metadata_externally=False + # because the conversion occurs before metadata can be set; the + # dataset is marked as deleted and a subsequent conversion is run. + chrom_count_dataset = len_dataset.get_converted_dataset( trans, "linecount" ) + if not chrom_count_dataset or chrom_count_dataset.state != trans.app.model.Job.states.OK: + # No valid linecount dataset. + continue + else: + # Set chrom count. + chrom_count = int( open( chrom_count_dataset.file_name ).readline() ) + attributes[ 'count' ] = chrom_count + updated = True + + if updated: + user.preferences['dbkeys'] = to_json_string(dbkeys) + trans.sa_session.flush() + + + # Potential genome data for custom builds is limited to fasta datasets in current history for now. + fasta_hdas = trans.sa_session.query( model.HistoryDatasetAssociation ) \ + .filter_by( history=trans.history, extension="fasta", deleted=False ) \ + .order_by( model.HistoryDatasetAssociation.hid.desc() ) + return trans.fill_template( 'user/dbkeys.mako', user=user, dbkeys=dbkeys, message=message, installed_len_files=self.installed_len_files, lines_skipped=lines_skipped, + fasta_hdas=fasta_hdas, use_panels=kwds.get( 'use_panels', None ) ) @web.expose @web.require_login() diff --git a/templates/user/dbkeys.mako b/templates/user/dbkeys.mako index 86b145f7743..bd07f350d71 100644 --- a/templates/user/dbkeys.mako +++ b/templates/user/dbkeys.mako @@ -103,6 +103,8 @@ ## % if 'count' in dct: ${dct['count']} + % else: + working % endif
@@ -133,10 +135,12 @@
- -
- - + +
diff --git a/tools/fasta_tools/fasta_compute_length.py b/tools/fasta_tools/fasta_compute_length.py index 3b055e3d5ba..57a257bb7df 100644 --- a/tools/fasta_tools/fasta_compute_length.py +++ b/tools/fasta_tools/fasta_compute_length.py @@ -1,47 +1,9 @@ #!/usr/bin/env python """ -Input: fasta, int -Output: tabular -Return titles with lengths of corresponding seq +Uses fasta_to_len converter code. """ -import sys, os +import sys +from galaxy.datatypes.converters.fasta_to_len import compute_fasta_length -assert sys.version_info[:2] >= ( 2, 4 ) - -def __main__(): - - infile = sys.argv[1] - out = open( sys.argv[2], 'w') - keep_first_char = int( sys.argv[3] ) - - fasta_title = '' - seq_len = 0 - - # number of char to keep in the title - if keep_first_char == 0: - keep_first_char = None - else: - keep_first_char += 1 - - first_entry = True - - for line in open( infile ): - line = line.strip() - if not line or line.startswith( '#' ): - continue - if line[0] == '>': - if first_entry == False: - out.write( "%s\t%d\n" % ( fasta_title[ 1:keep_first_char ], seq_len ) ) - else: - first_entry = False - fasta_title = line - seq_len = 0 - else: - seq_len += len(line) - - # last fasta-entry - out.write( "%s\t%d\n" % ( fasta_title[ 1:keep_first_char ], seq_len ) ) - out.close() - -if __name__ == "__main__" : __main__() \ No newline at end of file +compute_fasta_length( sys.argv[1], sys.argv[2], sys.argv[3]) \ No newline at end of file