#!/usr/bin/env python
from ConfigParser import SafeConfigParser
+import fnmatch
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
import urlparse
-
from htsworkflow.util import api
-from htsworkflow.pipelines.sequences import \
- create_sequence_table, \
- scan_for_sequences
+from htsworkflow.util.rdfhelp import \
+ dafTermOntology, \
+ fromTypedNode, \
+ get_model, \
+ get_serializer, \
+ load_into_model, \
+ submissionOntology
+from htsworkflow.submission.daf import DAFMapper, get_submission_uri
+from htsworkflow.submission.condorfastq import CondorFastqExtract
+
+logger = logging.getLogger('ucsc_gather')
-def make_submission_name(ininame):
- base = os.path.basename(ininame)
- name, ext = os.path.splitext(base)
- return name + '.tgz'
+def main(cmdline=None):
+ parser = make_parser()
+ opts, args = parser.parse_args(cmdline)
+
+ if opts.debug:
+ logging.basicConfig(level = logging.DEBUG )
+ elif opts.verbose:
+ logging.basicConfig(level = logging.INFO )
+ else:
+ logging.basicConfig(level = logging.WARNING )
+
+ apidata = api.make_auth_from_opts(opts, parser)
-def make_ddf_name(pathname):
- base = os.path.basename(pathname)
- name, ext = os.path.splitext(base)
- return name + '.ddf'
+ model = get_model(opts.load_model)
+ mapper = DAFMapper(opts.name, opts.daf, model)
+ submission_uri = get_submission_uri(opts.name)
+ if opts.load_rdf is not None:
+ load_into_model(model, 'turtle', opts.load_rdf, submission_uri)
-def make_condor_name(pathname):
- base = os.path.basename(pathname)
- name, ext = os.path.splitext(base)
- return name + '.condor'
+ 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")
+
+ library_result_map = []
+ for a in args:
+ library_result_map.extend(read_library_result_map(a))
+
+ if opts.make_tree_from is not None:
+ make_tree_from(opts.make_tree_from, library_result_map)
+
+ #if opts.daf is not None:
+ # link_daf(opts.daf, library_result_map)
+
+ if opts.fastq:
+ extractor = CondorFastqExtract(opts.host, apidata, opts.sequence,
+ force=opts.force)
+ extractor.build_fastqs(library_result_map)
+
+ if opts.scan_submission:
+ scan_submission_dirs(mapper, library_result_map)
+
+ if opts.makeddf:
+ make_all_ddfs(mapper, library_result_map, force=opts.force)
+
+ if opts.print_rdf:
+ writer = get_serializer()
+ print writer.serialize_model_to_string(model)
+
+
+def make_parser():
+ parser = OptionParser()
+
+ parser.add_option('--name', help="Set submission name")
+ parser.add_option('--load-model', default=None,
+ help="Load model database")
+ parser.add_option('--load-rdf', default=None,
+ help="load rdf statements into model")
+ parser.add_option('--print-rdf', action="store_true", default=False,
+ help="print ending model state")
+
+ # commands
+ parser.add_option('--make-tree-from',
+ help="create directories & link data files",
+ default=None)
+ parser.add_option('--fastq', help="generate scripts for making fastq files",
+ default=False, action="store_true")
+
+ parser.add_option('--scan-submission', default=False, action="store_true",
+ help="Import metadata for submission into our model")
-def make_submit_script(target, header, body_list):
- """
- write out a text file
+ parser.add_option('--makeddf', help='make the ddfs', default=False,
+ action="store_true")
+
+ parser.add_option('--daf', default=None, help='specify daf name')
+ parser.add_option('--force', default=False, action="store_true",
+ help="Force regenerating fastqs")
- this was intended for condor submit scripts
+ # 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)
- Args:
- target (str or stream):
- if target is a string, we will open and close the file
- if target is a stream, the caller is responsible.
-
- header (str);
- header to write at the beginning of the file
- body_list (list of strs):
- a list of blocks to add to the file.
+ return parser
+
+def make_tree_from(source_path, library_result_map):
+ """Create a tree using data files from source path.
"""
- if type(target) in types.StringTypes:
- f = open(target,'w')
- else:
- f = target
- f.write(header)
- for entry in body_list:
- f.write(entry)
- if type(target) in types.StringTypes:
- f.close()
+ for lib_id, lib_path in library_result_map:
+ if not os.path.exists(lib_path):
+ logging.info("Making dir {0}".format(lib_path))
+ os.mkdir(lib_path)
+ source_lib_dir = os.path.join(source_path, lib_path)
+ if os.path.exists(source_lib_dir):
+ pass
+ for filename in os.listdir(source_lib_dir):
+ source_pathname = os.path.join(source_lib_dir, filename)
+ target_pathname = os.path.join(lib_path, filename)
+ if not os.path.exists(source_pathname):
+ raise IOError("{0} does not exist".format(source_pathname))
+ if not os.path.exists(target_pathname):
+ os.symlink(source_pathname, target_pathname)
+ logging.info(
+ 'LINK {0} to {1}'.format(source_pathname, target_pathname))
+
-def parse_filelist(file_string):
- return file_string.split(',')
+def link_daf(daf_path, library_result_map):
+ if not os.path.exists(daf_path):
+ raise RuntimeError("%s does not exist, how can I link to it?" % (daf_path,))
-def validate_filelist(files):
- """
- Die if a file doesn't exist in a file list
+ 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)
+
+
+def scan_submission_dirs(view_map, library_result_map):
+ """Look through our submission directories and collect needed information
"""
- for f in files:
- if not os.path.exists(f):
- raise RuntimeError("%s does not exist" % (f,))
+ for lib_id, result_dir in library_result_map:
+ view_map.import_submission_dir(result_dir, lib_id)
+
+def make_all_ddfs(view_map, library_result_map, make_condor=True, force=False):
+ dag_fragment = []
+ for lib_id, result_dir in library_result_map:
+ submissionNode = view_map.get_submission_node(result_dir)
+ dag_fragment.extend(
+ make_ddf(view_map, submissionNode, make_condor, result_dir)
+ )
+
+ if make_condor and len(dag_fragment) > 0:
+ dag_filename = 'submission.dagman'
+ if not force and os.path.exists(dag_filename):
+ logging.warn("%s exists, please delete" % (dag_filename,))
+ else:
+ f = open(dag_filename,'w')
+ f.write( os.linesep.join(dag_fragment))
+ f.write( os.linesep )
+ f.close()
+
-def read_ddf_ini(filename, output=sys.stdout):
+def make_ddf(view_map, submissionNode, make_condor=False, outdir=None):
"""
- Read a ini file and dump out a tab delmited text file
+ Make ddf files, and bonus condor file
"""
- file_list = []
- config = SafeConfigParser()
- config.read(filename)
+ dag_fragments = []
+ curdir = os.getcwd()
+ if outdir is not None:
+ os.chdir(outdir)
+ output = sys.stdout
- order_by = shlex.split(config.get("config", "order_by"))
+ name = fromTypedNode(view_map.model.get_target(submissionNode, submissionOntology['name']))
+ if name is None:
+ logging.error("Need name for %s" % (str(submissionNode)))
+ return []
+
+ ddf_name = name + '.ddf'
+ output = sys.stdout
+ # output = open(ddf_name,'w')
- output.write("\t".join(order_by))
+ # filename goes first
+ variables = ['filename']
+ variables.extend(view_map.get_daf_variables())
+ output.write('\t'.join(variables))
output.write(os.linesep)
- sections = config.sections()
- sections.sort()
- for section in sections:
- if section == "config":
- # skip the config block
- continue
- values = []
- for key in order_by:
- v = config.get(section, key)
- values.append(v)
- if key == 'files':
- file_list.extend(parse_filelist(v))
-
- output.write("\t".join(values))
+
+ submission_views = view_map.model.get_targets(submissionNode, submissionOntology['has_view'])
+ file_list = []
+ for viewNode in submission_views:
+ record = []
+ for variable_name in variables:
+ varNode = dafTermOntology[variable_name]
+ values = [fromTypedNode(v) for v in list(view_map.model.get_targets(viewNode, varNode))]
+ if variable_name == 'filename':
+ file_list.extend(values)
+ if len(values) == 0:
+ attribute = "#None#"
+ elif len(values) == 1:
+ attribute = values[0]
+ else:
+ attribute = ",".join(values)
+ record.append(attribute)
+ output.write('\t'.join(record))
output.write(os.linesep)
- return file_list
+ logging.info(
+ "Examined {0}, found files: {1}".format(
+ str(submissionNode), ", ".join(file_list)))
+
+ #file_list.append(daf_name)
+ #if ddf_name is not None:
+ # file_list.append(ddf_name)
+ #
+ #if make_condor:
+ # archive_condor = make_condor_archive_script(ininame, file_list)
+ # upload_condor = make_condor_upload_script(ininame)
+ #
+ # dag_fragments.extend(
+ # make_dag_fragment(ininame, archive_condor, upload_condor)
+ # )
+
+ os.chdir(curdir)
+
+ return dag_fragments
+
+
+def read_library_result_map(filename):
+ """
+ Read a file that maps library id to result directory.
+ Does not support spaces in filenames.
+
+ For example:
+ 10000 result/foo/bar
+ """
+ stream = open(filename,'r')
+
+ results = []
+ for line in stream:
+ line = line.rstrip()
+ if not line.startswith('#') and len(line) > 0 :
+ library_id, result_dir = line.split()
+ results.append((library_id, result_dir))
+ return results
+
+
def make_condor_archive_script(ininame, files):
script = """Universe = vanilla
Executable = /bin/tar
-arguments = czvf ../%(archivename)s %(filelist)s
+arguments = czvhf ../%(archivename)s %(filelist)s
Error = compress.err.$(Process).log
Output = compress.out.$(Process).log
-Log = compress.log
+Log = /tmp/submission-compress-%(user)s.log
initialdir = %(initialdir)s
+environment="GZIP=-3"
+request_memory = 20
queue
"""
context = {'archivename': make_submission_name(ininame),
'filelist': " ".join(files),
- 'initialdir': os.getcwd()}
+ 'initialdir': os.getcwd(),
+ 'user': os.getlogin()}
- condor_script = make_condor_name(ininame)
+ condor_script = make_condor_name(ininame, 'archive')
condor_stream = open(condor_script,'w')
condor_stream.write(script % context)
condor_stream.close()
+ return condor_script
-def make_ddf(ininame, daf_name, guess_ddf=False, make_condor=False, outdir=None):
- """
- Make ddf files, and bonus condor file
- """
- curdir = os.getcwd()
- if outdir is not None:
- os.chdir(outdir)
- output = sys.stdout
- ddf_name = None
- if guess_ddf:
- ddf_name = make_ddf_name(ininame)
- print ddf_name
- output = open(ddf_name,'w')
-
- file_list = read_ddf_ini(ininame, output)
- file_list.append(daf_name)
- if ddf_name is not None:
- file_list.append(ddf_name)
-
- if make_condor:
- make_condor_archive_script(ininame, file_list)
-
- os.chdir(curdir)
+def make_condor_upload_script(ininame):
+ script = """Universe = vanilla
+Executable = /usr/bin/lftp
+arguments = -c put ../%(archivename)s -o ftp://%(ftpuser)s:%(ftppassword)s@%(ftphost)s/%(archivename)s
-def get_library_info(host, apidata, library_id):
- url = api.library_url(host, library_id)
- contents = api.retrieve_info(url, apidata)
- return contents
-
-def read_library_result_map(filename):
- stream = open(filename,'r')
-
- results = []
- for line in stream:
- library_id, result_dir = line.strip().split()
- results.append((library_id, result_dir))
- return results
+Error = upload.err.$(Process).log
+Output = upload.out.$(Process).log
+Log = /tmp/submission-upload-%(user)s.log
+initialdir = %(initialdir)s
-def condor_srf_to_fastq(srf_file, target_pathname):
- script = """output=%(target_pathname)s
-arguments="-c %(srf_file)s"
-queue
+queue
"""
- params = {'srf_file': srf_file,
- 'target_pathname': target_pathname}
+ auth = netrc.netrc(os.path.expanduser("~diane/.netrc"))
- return script % params
+ 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(),
+ 'ftpuser': ftpuser,
+ 'ftppassword': ftppassword,
+ 'ftphost': encodeftp}
-def condor_qseq_to_fastq(qseq_file, target_pathname):
- script = """
-arguments="-i %(qseq_file)s -o %(target_pathname)s"
-queue
-"""
- params = {'qseq_file': qseq_file,
- 'target_pathname': target_pathname}
-
- return script % params
+ 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)
-def find_archive_sequence_files(host, apidata, sequences_path,
- library_result_map):
- """
- Find all the archive sequence files possibly associated with our results.
+ return condor_script
- """
- logging.debug("Searching for sequence files in: %s" %(sequences_path,))
- lib_db = {}
- seq_dirs = set()
- #seq_dirs = set(os.path.join(sequences_path, 'srfs'))
- candidate_lanes = {}
- for lib_id, result_dir in library_result_map:
- lib_info = get_library_info(host, apidata, lib_id)
- lib_db[lib_id] = lib_info
-
- for lane in lib_info['lane_set']:
- lane_key = (lane['flowcell'], lane['lane_number'])
- candidate_lanes[lane_key] = lib_id
- seq_dirs.add(os.path.join(sequences_path,
- 'flowcells',
- lane['flowcell']))
- logging.debug("Seq_dirs = %s" %(unicode(seq_dirs)))
- candidate_seq_list = scan_for_sequences(seq_dirs)
-
- # at this point we have too many sequences as scan_for_sequences
- # returns all the sequences in a flowcell directory
- # so lets filter out the extras
-
- for seq in candidate_seq_list:
- lane_key = (seq.flowcell, seq.lane)
- lib_id = candidate_lanes.get(lane_key, None)
- if lib_id is not None:
- lib_info = lib_db[lib_id]
- lanes = lib_info.setdefault('lanes', {})
- lanes.setdefault(lane_key, set()).add(seq)
-
- return lib_db
-
-def build_fastqs(host, apidata, sequences_path, library_result_map,
- paired=True ):
+def make_dag_fragment(ininame, archive_condor, upload_condor):
"""
- Generate condor scripts to build any needed fastq files
-
- Args:
- host (str): root of the htsworkflow api server
- apidata (dict): id & key to post to the server
- sequences_path (str): root of the directory tree to scan for files
- library_result_map (list): [(library_id, destination directory), ...]
- paired: should we assume that we are processing paired end records?
- if False, we will treat this as single ended.
+ Make the couple of fragments compress and then upload the data.
"""
- qseq_condor_header = """
-Universe=vanilla
-executable=/woldlab/rattus/lvol0/mus/home/diane/proj/gaworkflow/scripts/qseq2fastq
-error=qseqfastq.err.$(process).log
-output=qseqfastq.out.$(process).log
-log=qseqfastq.log
+ cur_dir = os.getcwd()
+ archive_condor = os.path.join(cur_dir, archive_condor)
+ upload_condor = os.path.join(cur_dir, upload_condor)
+ job_basename = make_base_name(ininame)
-"""
- qseq_condor_entries = []
- srf_condor_header = """
-Universe=vanilla
-executable=/woldlab/rattus/lvol0/mus/home/diane/bin/srf2fastq
-output=srf2fastq.out.$(process).log
-error=srf2fastq.err.$(process).log
-log=srffastq.log
+ fragments = []
+ fragments.append('JOB %s_archive %s' % (job_basename, archive_condor))
+ fragments.append('JOB %s_upload %s' % (job_basename, upload_condor))
+ fragments.append('PARENT %s_archive CHILD %s_upload' % (job_basename, job_basename))
-"""
- srf_condor_entries = []
- fastq_paired_template = '%(lib_id)s_%(flowcell)s_c%(cycle)s_l%(lane)s_r%(read)s.fastq'
- fastq_single_template = '%(lib_id)s_%(flowcell)s_c%(cycle)s_l%(lane)s.fastq'
- lib_db = find_archive_sequence_files(host,
- apidata,
- sequences_path,
- library_result_map)
-
- # find what targets we're missing
- needed_targets = {}
- for lib_id, result_dir in library_result_map:
- lib = lib_db[lib_id]
- for lane_key, sequences in lib['lanes'].items():
- for seq in sequences:
- filename_attributes = {
- 'flowcell': seq.flowcell,
- 'lib_id': lib_id,
- 'lane': seq.lane,
- 'read': seq.read,
- 'cycle': seq.cycle
- }
- # throw out test runs
- # FIXME: this should probably be configurable
- if seq.cycle < 50:
- continue
- if seq.flowcell == '30CUUAAXX':
- # 30CUUAAXX run sucked
- continue
-
- # end filters
- if paired:
- target_name = fastq_paired_template % filename_attributes
- else:
- target_name = fastq_single_template % filename_attributes
-
- target_pathname = os.path.join(result_dir, target_name)
- if not os.path.exists(target_pathname):
- t = needed_targets.setdefault(target_pathname, {})
- t[seq.filetype] = seq
-
- for target_pathname, available_sources in needed_targets.items():
- logging.debug(' target : %s' % (target_pathname,))
- logging.debug(' candidate sources: %s' % (available_sources,))
- if available_sources.has_key('qseq'):
- source = available_sources['qseq']
- qseq_condor_entries.append(
- condor_qseq_to_fastq(source.path, target_pathname)
- )
- elif available_sources.has_key('srf'):
- source = available_sources['srf']
- if source.read is not None:
- logging.warn(
- "srf -> fastq paired end doesn't work yet: %s" % (source,)
- )
- else:
- srf_condor_entries.append(
- condor_srf_to_fastq(source.path, target_pathname)
- )
- else:
- print " need file", target_pathname
+ return fragments
- if len(srf_condor_entries) > 0:
- make_submit_script('srf.fastq.condor',
- srf_condor_header,
- srf_condor_entries)
- if len(qseq_condor_entries) > 0:
- make_submit_script('qseq.fastq.condor',
- qseq_condor_header,
- qseq_condor_entries)
+def get_library_info(host, apidata, library_id):
+ url = api.library_url(host, library_id)
+ contents = api.retrieve_info(url, apidata)
+ return contents
-def find_best_extension(extension_map, filename):
- """
- Search through extension_map looking for the best extension
- The 'best' is the longest match
- :Args:
- extension_map (dict): '.ext' -> { 'view': 'name' or None }
- filename (str): the filename whose extention we are about to examine
- """
- best_ext = None
- path, last_ext = os.path.splitext(filename)
-
- for ext in extension_map.keys():
- if filename.endswith(ext):
- if best_ext is None:
- best_ext = ext
- elif len(ext) > len(best_ext):
- best_ext = ext
- return best_ext
-
-def add_submission_section(line_counter, files, standard_attributes, file_attributes):
+def make_submission_section(line_counter, files, attributes):
"""
Create a section in the submission ini file
"""
- inifile = [ '[line%s]' % (line_counter,) ]
+ inifile = [ "[line%s]" % (line_counter,) ]
inifile += ["files=%s" % (",".join(files))]
- cur_attributes = {}
- cur_attributes.update(standard_attributes)
- cur_attributes.update(file_attributes)
-
- for k,v in cur_attributes.items():
+
+ for k,v in attributes.items():
inifile += ["%s=%s" % (k,v)]
return inifile
-
-def make_submission_ini(host, apidata, library_result_map):
- attributes = {
- '.bai': {'view': None}, # bam index
- '.bam': {'view': 'Signal'},
- '.condor': {'view': None},
- '.daf': {'view': None},
- '.ddf': {'view': None},
- '.splices.bam': {'view': 'Splices'},
- '.bed': {'view': 'TopHatJunctions'},
- '.ini': {'view': None},
- '.log': {'view': None},
- '_r1.fastq': {'view': 'FastqRd1'},
- '_r2.fastq': {'view': 'FastqRd2'},
- '.tar.bz2': {'view': None},
- 'novel.genes.expr': {'view': 'GeneDeNovoFPKM'},
- 'novel.transcripts.expr': {'view': 'TranscriptDeNovoFPKM'},
- '.genes.expr': {'view': 'GeneFPKM'},
- '.transcripts.expr': {'view': 'TranscriptFPKM'},
- '.stats.txt': {'view': 'InsLength'},
- '.gtf': {'view': 'CufflinksGeneModel'},
- '.wig': {'view': 'RawSignal'},
- }
-
- candidate_fastq_src = {}
- for lib_id, result_dir in library_result_map:
- inifile = ['[config]']
- inifile += ['order_by=files view cell localization rnaExtract mapAlgorithm readType replicate labVersion']
- inifile += ['']
- line_counter = 1
- lib_info = get_library_info(host, apidata, lib_id)
- result_ini = os.path.join(result_dir, result_dir+'.ini')
-
- standard_attributes = {'cell': lib_info['cell_line'],
- 'insertLength': '200', # ali
- 'labVersion': 'TopHat',
- 'localization': 'cell',
- 'mapAlgorithm': 'TopHat',
- 'readType': '2x75', #ali
- 'replicate': lib_info['replicate'],
- 'rnaExtract': 'longPolyA',
- }
-
- # write fastq line
- #fastqs = []
- #for lane in lib_info['lane_set']:
- # target_name = "%s_%s_%s.fastq" % (lane['flowcell'], lib_id, lane['lane_number'])
- # fastqs.append(target_name)
- #inifile.extend(
- # make_run_block(line_counter, fastqs, standard_attributes, attributes['.fastq'])
- #)
- #inifile += ['']
- #line_counter += 1
-
- # write other lines
- submission_files = os.listdir(result_dir)
- for f in submission_files:
- best_ext = find_best_extension(attributes, f)
-
- if best_ext is not None:
- if attributes[best_ext]['view'] is None:
- continue
- inifile.extend(
- add_submission_section(line_counter,
- [f],
- standard_attributes,
- attributes[best_ext]
- )
- )
- inifile += ['']
- line_counter += 1
- else:
- raise ValueError("Unrecognized file: %s" % (f,))
- f = open(result_ini,'w')
- f.write(os.linesep.join(inifile))
- f.close()
+def make_base_name(pathname):
+ base = os.path.basename(pathname)
+ name, ext = os.path.splitext(base)
+ return name
-def link_daf(daf_path, library_result_map):
- if not os.path.exists(daf_path):
- raise RuntimeError("%s does not exist, how can I link to it?" % (daf_path,))
- base_daf = os.path.basename(daf_path)
-
- for lib_id, result_dir in library_result_map:
- submission_daf = os.path.join(result_dir, base_daf)
- if not os.path.exists(submission_daf):
- os.link(daf_path, submission_daf)
+def make_submission_name(ininame):
+ name = make_base_name(ininame)
+ return name + ".tgz"
-def make_all_ddfs(library_result_map, daf_name):
- for lib_id, result_dir in library_result_map:
- ininame = result_dir+'.ini'
- inipathname = os.path.join(result_dir, ininame)
- if os.path.exists(inipathname):
- make_ddf(ininame, daf_name, True, True, result_dir)
-
-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()
+def make_ddf_name(pathname):
+ name = make_base_name(pathname)
+ return name + ".ddf"
- # commands
- parser.add_option('--fastq', help="generate scripts for making fastq files",
- default=False, action="store_true")
- parser.add_option('--ini', help="generate submission ini file", default=False,
- action="store_true")
+def make_condor_name(pathname, run_type=None):
+ name = make_base_name(pathname)
+ elements = [name]
+ if run_type is not None:
+ elements.append(run_type)
+ elements.append("condor")
+ return ".".join(elements)
- parser.add_option('--makeddf', help='make the ddfs', default=False,
- action="store_true")
-
- parser.add_option('--daf', default=None, help='specify daf name')
- # 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")
- parser.add_option('--single', default=False, action="store_true",
- help="treat the sequences as single ended runs")
+def parse_filelist(file_string):
+ return file_string.split(",")
- # 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')
- return parser
+def validate_filelist(files):
+ """
+ Die if a file doesn't exist in a file list
+ """
+ for f in files:
+ if not os.path.exists(f):
+ raise RuntimeError("%s does not exist" % (f,))
-def main(cmdline=None):
- parser = make_parser()
- opts, args = parser.parse_args(cmdline)
-
- if opts.debug:
- logging.basicConfig(level = logging.DEBUG )
- elif opts.verbose:
- logging.basicConfig(level = logging.INFO )
+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:
- logging.basicConfig(level = logging.WARNING )
-
+ md5sum = make_md5sum_unix(filename, md5_cache)
+ return md5sum
- apidata = {'apiid': opts.apiid, 'apikey': opts.apikey }
+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 opts.host is None or opts.apiid is None or opts.apikey is None:
- parser.error("Please specify host url, apiid, apikey")
-
- if len(args) == 0:
- parser.error("I need at least one library submission-dir input file")
-
- library_result_map = []
- for a in args:
- library_result_map.extend(read_library_result_map(a))
-
- if opts.daf is not None:
- link_daf(opts.daf, library_result_map)
-
- if opts.fastq:
- build_fastqs(opts.host,
- apidata,
- opts.sequence,
- library_result_map,
- not opts.single)
-
- if opts.ini:
- make_submission_ini(opts.host, apidata, library_result_map)
-
- if opts.makeddf:
- make_all_ddfs(library_result_map, opts.daf)
-
if __name__ == "__main__":
main()