from glob import glob
import json
import logging
+import netrc
from optparse import OptionParser
import os
from pprint import pprint, pformat
import shlex
from StringIO import StringIO
-import time
+import stat
+from subprocess import Popen, PIPE
import sys
+import time
import types
import urllib
import urllib2
else:
logging.basicConfig(level = logging.WARNING )
- apidata = {'apiid': opts.apiid, 'apikey': opts.apikey }
+ apidata = api.make_auth_from_opts(opts, parser)
- if opts.host is None or opts.apiid is None or opts.apikey is None:
- parser.error("Please specify host url, apiid, apikey")
+ if opts.makeddf and opts.daf is None:
+ parser.error("Please specify your daf when making ddf files")
if len(args) == 0:
parser.error("I need at least one library submission-dir input file")
def make_parser():
- # Load defaults from the config files
- config = SafeConfigParser()
- config.read([os.path.expanduser('~/.htsworkflow.ini'), '/etc/htsworkflow.ini'])
-
- sequence_archive = None
- apiid = None
- apikey = None
- apihost = None
- SECTION = 'sequence_archive'
- if config.has_section(SECTION):
- sequence_archive = config.get(SECTION, 'sequence_archive',sequence_archive)
- sequence_archive = os.path.expanduser(sequence_archive)
- apiid = config.get(SECTION, 'apiid', apiid)
- apikey = config.get(SECTION, 'apikey', apikey)
- apihost = config.get(SECTION, 'host', apihost)
-
parser = OptionParser()
# commands
parser.add_option('--force', default=False, action="store_true",
help="Force regenerating fastqs")
- # configuration options
- parser.add_option('--apiid', default=apiid, help="Specify API ID")
- parser.add_option('--apikey', default=apikey, help="Specify API KEY")
- parser.add_option('--host', default=apihost,
- help="specify HTSWorkflow host",)
- parser.add_option('--sequence', default=sequence_archive,
- help="sequence repository")
-
# debugging
parser.add_option('--verbose', default=False, action="store_true",
help='verbose logging')
parser.add_option('--debug', default=False, action="store_true",
help='debug logging')
+ api.add_auth_options(parser)
+
return parser
base_daf = os.path.basename(daf_path)
for lib_id, result_dir in library_result_map:
+ if not os.path.exists(result_dir):
+ raise RuntimeError("Couldn't find target directory %s" %(result_dir,))
submission_daf = os.path.join(result_dir, base_daf)
if not os.path.exists(submission_daf):
+ if not os.path.exists(daf_path):
+ raise RuntimeError("Couldn't find daf: %s" %(daf_path,))
os.link(daf_path, submission_daf)
for lib_id, result_dir in library_result_map:
order_by = ['order_by=files', 'view', 'replicate', 'cell',
- 'readType', 'mapAlgorithm', 'insertLength' ]
+ 'readType', 'mapAlgorithm', 'insertLength', 'md5sum' ]
inifile = ['[config]']
inifile += [" ".join(order_by)]
inifile += ['']
attributes = view_map.find_attributes(f, lib_id)
if attributes is None:
raise ValueError("Unrecognized file: %s" % (f,))
+ attributes['md5sum'] = "None"
ext = attributes["extension"]
if attributes['view'] is None:
fastqs.setdefault(ext, set()).add(f)
fastq_attributes[ext] = attributes
else:
+ md5sum = make_md5sum(os.path.join(result_dir,f))
+ if md5sum is not None:
+ attributes['md5sum']=md5sum
inifile.extend(
make_submission_section(line_counter,
[f],
output = open(ddf_name,'w')
file_list = read_ddf_ini(ininame, output)
+ logging.info(
+ "Read config {0}, found files: {1}".format(
+ ininame, ", ".join(file_list)))
file_list.append(daf_name)
if ddf_name is not None:
script = """Universe = vanilla
Executable = /usr/bin/lftp
-arguments = -c put ../%(archivename)s -o ftp://detrout@encodeftp.cse.ucsc.edu/%(archivename)s
+arguments = -c put ../%(archivename)s -o ftp://%(ftpuser)s:%(ftppassword)s@%(ftphost)s/%(archivename)s
Error = upload.err.$(Process).log
Output = upload.out.$(Process).log
queue
"""
+ auth = netrc.netrc(os.path.expanduser("~diane/.netrc"))
+
+ encodeftp = 'encodeftp.cse.ucsc.edu'
+ ftpuser = auth.hosts[encodeftp][0]
+ ftppassword = auth.hosts[encodeftp][2]
context = {'archivename': make_submission_name(ininame),
'initialdir': os.getcwd(),
- 'user': os.getlogin()}
+ 'user': os.getlogin(),
+ 'ftpuser': ftpuser,
+ 'ftppassword': ftppassword,
+ 'ftphost': encodeftp}
condor_script = make_condor_name(ininame, 'upload')
condor_stream = open(condor_script,'w')
condor_stream.write(script % context)
condor_stream.close()
+ os.chmod(condor_script, stat.S_IREAD|stat.S_IWRITE)
+
return condor_script
ma = 'TH1014'
self.patterns = [
+ # for 2011 Feb 18 elements submission
+ ('final_Cufflinks_genes_*gtf', 'GeneDeNovo'),
+ ('final_Cufflinks_transcripts_*gtf', 'TranscriptDeNovo'),
+ ('final_exonFPKM-Cufflinks-0.9.3-GENCODE-v3c-*.gtf',
+ 'ExonsGencV3c'),
+ ('final_GENCODE-v3-Cufflinks-0.9.3.genes-*gtf',
+ 'GeneGencV3c'),
+ ('final_GENCODE-v3-Cufflinks-0.9.3.transcripts-*gtf',
+ 'TranscriptGencV3c'),
+ ('final_TSS-Cufflinks-0.9.3-GENCODE-v3c-*.gtf', 'TSS'),
+ ('final_junctions-*.bed6+3', 'Junctions'),
+
('*.bai', None),
('*.splices.bam', 'Splices'),
('*.bam', self._guess_bam_view),
('junctions.bed', 'Junctions'),
('*.jnct', 'Junctions'),
- ('*.plus.bigwig', 'PlusSignal'),
- ('*.minus.bigwig', 'MinusSignal'),
+ ('*unique.bigwig', None),
+ ('*plus.bigwig', 'PlusSignal'),
+ ('*minus.bigwig', 'MinusSignal'),
('*.bigwig', 'Signal'),
('*.tar.bz2', None),
('*.condor', None),
('*.daf', None),
('*.ddf', None),
+
+ ('*ufflinks?0.9.3.genes.gtf', 'GeneDeNovo'),
+ ('*ufflinks?0.9.3.transcripts.gtf', 'TranscriptDeNovo'),
+ ('*GENCODE-v3c.exonFPKM.gtf', 'ExonsGencV3c'),
+ ('*GENCODE-v3c.genes.gtf', 'GeneGencV3c'),
+ ('*GENCODE-v3c.transcripts.gtf', 'TranscriptGencV3c'),
+ ('*GENCODE-v3c.TSS.gtf', 'TSS'),
+ ('*.junctions.bed6+3', 'Junctions'),
+
('*.?ufflinks-0.9.0?genes.expr', 'GeneDeNovo'),
('*.?ufflinks-0.9.0?transcripts.expr', 'TranscriptDeNovo'),
('*.?ufflinks-0.9.0?transcripts.gtf', 'GeneModel'),
- ('*.GENCODE-v3c?genes.expr', 'GeneGencV3c'),
- ('*.GENCODE-v3c?transcripts.expr', 'TranscriptGencV3c'),
- ('*.GENCODE-v4?genes.expr', 'GeneGencV4'),
- ('*.GENCODE-v4?transcripts.expr', 'TranscriptGencV4'),
- ('*.GENCODE-v4?transcript.expr', 'TranscriptGencV4'),
+
+ ('*.GENCODE-v3c?genes.expr', 'GeneGCV3c'),
+ ('*.GENCODE-v3c?transcript*.expr', 'TranscriptGCV3c'),
+ ('*.GENCODE-v3c?transcript*.gtf', 'TranscriptGencV3c'),
+ ('*.GENCODE-v4?genes.expr', None), #'GeneGCV4'),
+ ('*.GENCODE-v4?transcript*.expr', None), #'TranscriptGCV4'),
+ ('*.GENCODE-v4?transcript*.gtf', None), #'TranscriptGencV4'),
('*_1.75mers.fastq', 'FastqRd1'),
('*_2.75mers.fastq', 'FastqRd2'),
('*_r1.fastq', 'FastqRd1'),
('*.gtf', 'GeneModel'),
('*.ini', None),
('*.log', None),
+ ('*.md5', None),
('paired-end-distribution*', 'InsLength'),
('*.stats.txt', 'InsLength'),
('*.srf', None),
('*.wig', None),
('*.zip', None),
+ ('transfer_log', None),
]
self.views = {
"GeneModel": {"MapAlgorithm": ma},
"GeneDeNovo": {"MapAlgorithm": ma},
"TranscriptDeNovo": {"MapAlgorithm": ma},
+ "ExonsGencV3c": {"MapAlgorithm": ma},
"GeneGencV3c": {"MapAlgorithm": ma},
+ "TSS": {"MapAlgorithm": ma},
+ "GeneGCV3c": {"MapAlgorithm": ma},
+ "TranscriptGCV3c": {"MapAlgorithm": ma},
"TranscriptGencV3c": {"MapAlgorithm": ma},
- "GeneGencV4": {"MapAlgorithm": ma},
- "TranscriptGencV4": {"MapAlgorithm": ma},
+ "GeneGCV4": {"MapAlgorithm": ma},
+ "TranscriptGCV4": {"MapAlgorithm": ma},
"FastqRd1": {"MapAlgorithm": "NA", "type": "fastq"},
"FastqRd2": {"MapAlgorithm": "NA", "type": "fastq"},
"Fastq": {"MapAlgorithm": "NA", "type": "fastq" },
- "GeneModel": {"MapAlgorithm": ma},
"InsLength": {"MapAlgorithm": ma},
}
# view name is one of the attributes
def _is_paired(self, lib_id, lib_info):
"""Determine if a library is paired end"""
+ # TODO: encode this information in the library type page.
+ single = (1,3,6)
if len(lib_info["lane_set"]) == 0:
- return False
+ # we haven't sequenced anything so guess based on library type
+ if lib_info['library_type_id'] in single:
+ return False
+ else:
+ return True
if not self.lib_paired.has_key(lib_id):
is_paired = 0
if not os.path.exists(f):
raise RuntimeError("%s does not exist" % (f,))
+def make_md5sum(filename):
+ """Quickly find the md5sum of a file
+ """
+ md5_cache = os.path.join(filename+".md5")
+ print md5_cache
+ if os.path.exists(md5_cache):
+ logging.debug("Found md5sum in {0}".format(md5_cache))
+ stream = open(md5_cache,'r')
+ lines = stream.readlines()
+ md5sum = parse_md5sum_line(lines, filename)
+ else:
+ md5sum = make_md5sum_unix(filename, md5_cache)
+ return md5sum
+
+def make_md5sum_unix(filename, md5_cache):
+ cmd = ["md5sum", filename]
+ logging.debug("Running {0}".format(" ".join(cmd)))
+ p = Popen(cmd, stdout=PIPE)
+ stdin, stdout = p.communicate()
+ retcode = p.wait()
+ logging.debug("Finished {0} retcode {1}".format(" ".join(cmd), retcode))
+ if retcode != 0:
+ logging.error("Trouble with md5sum for {0}".format(filename))
+ return None
+ lines = stdin.split(os.linesep)
+ md5sum = parse_md5sum_line(lines, filename)
+ if md5sum is not None:
+ logging.debug("Caching sum in {0}".format(md5_cache))
+ stream = open(md5_cache, "w")
+ stream.write(stdin)
+ stream.close()
+ return md5sum
+
+def parse_md5sum_line(lines, filename):
+ md5sum, md5sum_filename = lines[0].split()
+ if md5sum_filename != filename:
+ errmsg = "MD5sum and I disagre about filename. {0} != {1}"
+ logging.error(errmsg.format(filename, md5sum_filename))
+ return None
+ return md5sum
if __name__ == "__main__":
main()
projects / htsworkflow.git / blobdiff