2 from ConfigParser import SafeConfigParser
8 from optparse import OptionParser, OptionGroup
10 from pprint import pprint, pformat
12 from StringIO import StringIO
23 from htsworkflow.util import api
24 from htsworkflow.util.rdfhelp import \
32 from htsworkflow.submission.daf import \
34 MetadataLookupException, \
36 from htsworkflow.submission.condorfastq import CondorFastqExtract
38 logger = logging.getLogger('ucsc_gather')
40 def main(cmdline=None):
41 parser = make_parser()
42 opts, args = parser.parse_args(cmdline)
46 logging.basicConfig(level = logging.DEBUG )
48 logging.basicConfig(level = logging.INFO )
50 logging.basicConfig(level = logging.WARNING )
52 apidata = api.make_auth_from_opts(opts, parser)
54 model = get_model(opts.model, opts.db_path)
56 mapper = DAFMapper(opts.name, opts.daf, model)
57 if opts.library_url is not None:
58 mapper.library_url = opts.library_url
59 submission_uri = get_submission_uri(opts.name)
62 if opts.load_rdf is not None:
63 if submission_uri is None:
64 parser.error("Please specify the submission name")
65 load_into_model(model, 'turtle', opts.load_rdf, submission_uri)
67 if opts.make_ddf and opts.daf is None:
68 parser.error("Please specify your daf when making ddf files")
70 library_result_map = []
72 library_result_map.extend(read_library_result_map(a))
74 if opts.make_tree_from is not None:
75 make_tree_from(opts.make_tree_from, library_result_map)
79 parser.error("Please specify daf filename with --daf")
80 link_daf(opts.daf, library_result_map)
83 extractor = CondorFastqExtract(opts.host, apidata, opts.sequence,
85 extractor.build_fastqs(library_result_map)
87 if opts.scan_submission:
88 scan_submission_dirs(mapper, library_result_map)
91 make_all_ddfs(mapper, library_result_map, opts.daf, force=opts.force)
94 sparql_query(model, opts.sparql)
97 writer = get_serializer()
98 print writer.serialize_model_to_string(model)
102 parser = OptionParser()
104 model = OptionGroup(parser, 'model')
105 model.add_option('--name', help="Set submission name")
106 model.add_option('--db-path', default=None,
107 help="set rdf database path")
108 model.add_option('--model', default=None,
109 help="Load model database")
110 model.add_option('--load-rdf', default=None,
111 help="load rdf statements into model")
112 model.add_option('--sparql', default=None, help="execute sparql query")
113 model.add_option('--print-rdf', action="store_true", default=False,
114 help="print ending model state")
115 parser.add_option_group(model)
117 commands = OptionGroup(parser, 'commands')
118 commands.add_option('--make-tree-from',
119 help="create directories & link data files",
121 commands.add_option('--fastq', default=False, action="store_true",
122 help="generate scripts for making fastq files")
123 commands.add_option('--scan-submission', default=False, action="store_true",
124 help="Import metadata for submission into our model")
125 commands.add_option('--link-daf', default=False, action="store_true",
126 help="link daf into submission directories")
127 commands.add_option('--make-ddf', help='make the ddfs', default=False,
129 parser.add_option_group(commands)
131 parser.add_option('--force', default=False, action="store_true",
132 help="Force regenerating fastqs")
133 parser.add_option('--daf', default=None, help='specify daf name')
134 parser.add_option('--library-url', default=None,
135 help="specify an alternate source for library information")
137 parser.add_option('--verbose', default=False, action="store_true",
138 help='verbose logging')
139 parser.add_option('--debug', default=False, action="store_true",
140 help='debug logging')
142 api.add_auth_options(parser)
146 def make_tree_from(source_path, library_result_map):
147 """Create a tree using data files from source path.
149 for lib_id, lib_path in library_result_map:
150 if not os.path.exists(lib_path):
151 logger.info("Making dir {0}".format(lib_path))
153 source_lib_dir = os.path.abspath(os.path.join(source_path, lib_path))
154 if os.path.exists(source_lib_dir):
156 for filename in os.listdir(source_lib_dir):
157 source_pathname = os.path.join(source_lib_dir, filename)
158 target_pathname = os.path.join(lib_path, filename)
159 if not os.path.exists(source_pathname):
160 raise IOError("{0} does not exist".format(source_pathname))
161 if not os.path.exists(target_pathname):
162 os.symlink(source_pathname, target_pathname)
164 'LINK {0} to {1}'.format(source_pathname, target_pathname))
167 def link_daf(daf_path, library_result_map):
168 if not os.path.exists(daf_path):
169 raise RuntimeError("%s does not exist, how can I link to it?" % (daf_path,))
171 base_daf = os.path.basename(daf_path)
173 for lib_id, result_dir in library_result_map:
174 if not os.path.exists(result_dir):
175 raise RuntimeError("Couldn't find target directory %s" %(result_dir,))
176 submission_daf = os.path.join(result_dir, base_daf)
177 if not os.path.exists(submission_daf):
178 if not os.path.exists(daf_path):
179 raise RuntimeError("Couldn't find daf: %s" %(daf_path,))
180 os.link(daf_path, submission_daf)
183 def scan_submission_dirs(view_map, library_result_map):
184 """Look through our submission directories and collect needed information
186 for lib_id, result_dir in library_result_map:
187 logger.info("Importing %s from %s" % (lib_id, result_dir))
189 view_map.import_submission_dir(result_dir, lib_id)
190 except MetadataLookupException, e:
191 logger.error("Skipping %s: %s" % (lib_id, str(e)))
193 def make_all_ddfs(view_map, library_result_map, daf_name, make_condor=True, force=False):
195 for lib_id, result_dir in library_result_map:
196 submissionNode = view_map.get_submission_node(result_dir)
198 make_ddf(view_map, submissionNode, daf_name, make_condor, result_dir)
201 if make_condor and len(dag_fragment) > 0:
202 dag_filename = 'submission.dagman'
203 if not force and os.path.exists(dag_filename):
204 logger.warn("%s exists, please delete" % (dag_filename,))
206 f = open(dag_filename,'w')
207 f.write( os.linesep.join(dag_fragment))
208 f.write( os.linesep )
212 def make_ddf(view_map, submissionNode, daf_name, make_condor=False, outdir=None):
214 Make ddf files, and bonus condor file
216 query_template = """PREFIX libraryOntology: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
217 PREFIX submissionOntology: <http://jumpgate.caltech.edu/wiki/UcscSubmissionOntology#>
218 PREFIX ucscDaf: <http://jumpgate.caltech.edu/wiki/UcscDaf#>
220 select ?submitView ?files ?md5sum ?view ?cell ?antibody ?sex ?control ?strain ?controlId ?labExpId ?labVersion ?treatment ?protocol ?readType ?insertLength ?replicate ?mapAlgorithm
222 ?file ucscDaf:filename ?files ;
223 ucscDaf:md5sum ?md5sum .
224 ?submitView ucscDaf:has_file ?file ;
225 ucscDaf:view ?dafView ;
226 ucscDaf:submission <%(submission)s> .
227 ?dafView ucscDaf:name ?view .
228 <%(submission)s> submissionOntology:library ?library ;
230 OPTIONAL { ?library libraryOntology:antibody ?antibody }
231 OPTIONAL { ?library libraryOntology:cell_line ?cell }
232 OPTIONAL { <%(submission)s> ucscDaf:control ?control }
233 OPTIONAL { <%(submission)s> ucscDaf:controlId ?controlId }
234 OPTIONAL { ?library ucscDaf:sex ?sex }
235 OPTIONAL { ?library libraryOntology:library_id ?labExpId }
236 OPTIONAL { ?library libraryOntology:library_id ?labVersion }
237 OPTIONAL { ?library libraryOntology:replicate ?replicate }
238 OPTIONAL { ?library libraryOntology:condition ?treatment }
239 OPTIONAL { ?library ucscDaf:protocol ?protocol }
240 OPTIONAL { ?library ucscDaf:readType ?readType }
241 OPTIONAL { ?library ucscDaf:strain ?strain }
242 OPTIONAL { ?library libraryOntology:insert_size ?insertLength }
243 OPTIONAL { ?library ucscDaf:mapAlgorithm ?mapAlgorithm }
245 ORDER BY ?submitView"""
248 name = fromTypedNode(view_map.model.get_target(submissionNode, submissionOntology['name']))
250 logger.error("Need name for %s" % (str(submissionNode)))
253 ddf_name = name + '.ddf'
254 if outdir is not None:
255 outfile = os.path.join(outdir, ddf_name)
256 output = open(outfile,'w')
261 formatted_query = query_template % {'submission': str(submissionNode.uri)}
263 query = RDF.SPARQLQuery(formatted_query)
264 results = query.execute(view_map.model)
266 # filename goes first
267 variables = view_map.get_daf_variables()
269 output.write('\t'.join(variables))
270 output.write(os.linesep)
275 viewname = fromTypedNode(row['view'])
276 current = all_views.setdefault(viewname, {})
277 for variable_name in variables:
278 value = str(fromTypedNode(row[variable_name]))
279 if value is None or value == 'None':
280 logger.warn("{0}: {1} was None".format(outfile, variable_name))
281 if variable_name in ('files', 'md5sum'):
282 current.setdefault(variable_name,[]).append(value)
284 current[variable_name] = value
286 for view in all_views.keys():
288 for variable_name in variables:
289 if variable_name in ('files', 'md5sum'):
290 line.append(','.join(all_views[view][variable_name]))
292 line.append(all_views[view][variable_name])
293 output.write("\t".join(line))
294 output.write(os.linesep)
295 all_files.extend(all_views[view]['files'])
298 "Examined {0}, found files: {1}".format(
299 str(submissionNode), ", ".join(all_files)))
301 all_files.append(daf_name)
302 all_files.append(ddf_name)
305 archive_condor = make_condor_archive_script(name, all_files, outdir)
306 upload_condor = make_condor_upload_script(name, outdir)
308 dag_fragments.extend(
309 make_dag_fragment(name, archive_condor, upload_condor)
315 def read_library_result_map(filename):
317 Read a file that maps library id to result directory.
318 Does not support spaces in filenames.
323 stream = open(filename,'r')
328 if not line.startswith('#') and len(line) > 0 :
329 library_id, result_dir = line.split()
330 results.append((library_id, result_dir))
334 def make_condor_archive_script(name, files, outdir=None):
335 script = """Universe = vanilla
337 Executable = /bin/tar
338 arguments = czvhf ../%(archivename)s %(filelist)s
340 Error = compress.out.$(Process).log
341 Output = compress.out.$(Process).log
342 Log = /tmp/submission-compress-%(user)s.log
343 initialdir = %(initialdir)s
344 environment="GZIP=-3"
352 pathname = os.path.join(outdir, f)
353 if not os.path.exists(pathname):
354 raise RuntimeError("Missing %s from %s" % (f,outdir))
356 context = {'archivename': make_submission_name(name),
357 'filelist': " ".join(files),
358 'initialdir': os.path.abspath(outdir),
359 'user': os.getlogin()}
361 condor_script = os.path.join(outdir, make_condor_name(name, 'archive'))
362 condor_stream = open(condor_script,'w')
363 condor_stream.write(script % context)
364 condor_stream.close()
368 def make_condor_upload_script(name, outdir=None):
369 script = """Universe = vanilla
371 Executable = /usr/bin/lftp
372 arguments = -c put ../%(archivename)s -o ftp://%(ftpuser)s:%(ftppassword)s@%(ftphost)s/%(archivename)s
374 Error = upload.out.$(Process).log
375 Output = upload.out.$(Process).log
376 Log = /tmp/submission-upload-%(user)s.log
377 initialdir = %(initialdir)s
384 auth = netrc.netrc(os.path.expanduser("~diane/.netrc"))
386 encodeftp = 'encodeftp.cse.ucsc.edu'
387 ftpuser = auth.hosts[encodeftp][0]
388 ftppassword = auth.hosts[encodeftp][2]
389 context = {'archivename': make_submission_name(name),
390 'initialdir': os.path.abspath(outdir),
391 'user': os.getlogin(),
393 'ftppassword': ftppassword,
394 'ftphost': encodeftp}
396 condor_script = os.path.join(outdir, make_condor_name(name, 'upload'))
397 condor_stream = open(condor_script,'w')
398 condor_stream.write(script % context)
399 condor_stream.close()
400 os.chmod(condor_script, stat.S_IREAD|stat.S_IWRITE)
405 def make_dag_fragment(ininame, archive_condor, upload_condor):
407 Make the couple of fragments compress and then upload the data.
409 cur_dir = os.getcwd()
410 archive_condor = os.path.join(cur_dir, archive_condor)
411 upload_condor = os.path.join(cur_dir, upload_condor)
412 job_basename = make_base_name(ininame)
415 fragments.append('JOB %s_archive %s' % (job_basename, archive_condor))
416 fragments.append('JOB %s_upload %s' % (job_basename, upload_condor))
417 fragments.append('PARENT %s_archive CHILD %s_upload' % (job_basename, job_basename))
422 def get_library_info(host, apidata, library_id):
423 url = api.library_url(host, library_id)
424 contents = api.retrieve_info(url, apidata)
428 def make_submission_section(line_counter, files, attributes):
430 Create a section in the submission ini file
432 inifile = [ "[line%s]" % (line_counter,) ]
433 inifile += ["files=%s" % (",".join(files))]
435 for k,v in attributes.items():
436 inifile += ["%s=%s" % (k,v)]
440 def make_base_name(pathname):
441 base = os.path.basename(pathname)
442 name, ext = os.path.splitext(base)
446 def make_submission_name(ininame):
447 name = make_base_name(ininame)
451 def make_ddf_name(pathname):
452 name = make_base_name(pathname)
456 def make_condor_name(pathname, run_type=None):
457 name = make_base_name(pathname)
459 if run_type is not None:
460 elements.append(run_type)
461 elements.append("condor")
462 return ".".join(elements)
465 def parse_filelist(file_string):
466 return file_string.split(",")
469 def validate_filelist(files):
471 Die if a file doesn't exist in a file list
474 if not os.path.exists(f):
475 raise RuntimeError("%s does not exist" % (f,))
477 if __name__ == "__main__":