if reduce(operator.or_,
[x in chunk.lower() for x in useless_lines]): continue
if len(chunk) < 2: continue
outfile.write('%010d|%s\n' % (chunkid, chunk))
chunkmap[fakename].append(chunkid)
chunkid += 1
outfile.close()
print "Saving chunkmap"
pickle.dump(chunkmap, open(outmapname, "wb"), pickle.HIGHEST_PROTOCOL)
print "These files couldn't be processed:"
print '\n'.join(skipped)
print "Opening (or creating) cache in", sys.argv[2]
the_cache=StringDBDict(os.path.join(sys.argv[2], DEFAULT_CACHE_NAME),
file_mode='c')
PubMed.download_many([str(x) for x in known_articles if str(x) not in
the_cache.keys()], download_callback,
parser=Medline.RecordParser())
mti_filename=sys.argv[1]+'.mti'
print "Finished processing the cache. Using the cache to build", \
mti_filename
mti_file=open(mti_filename, "w")
chunkmap={}
hexfinder=re.compile(r'\\x[a-f0-9][a-f0-9]', re.IGNORECASE)
for article in known_articles:
try:
article_record=the_cache[str(article)]
except KeyError:
print "Article doesn't exist in cache. Skipping."
continue
if article_record.abstract=='':
print "Article", article, "has no abstract. Skipping."
else:
raise "unknown format"
if options.clipboard:
os.system('dcop klipper klipper setClipboardContents \"%s\"' % '\n'.join(outlines))
else:
print "\n".join(outlines)
if options.library:
nids = []
if os.path.exists( options.library ):
pmids = map(lambda x: x[:-1].split(" ")[1], filter(lambda x: re.match( "PMID", x), open(options.library).readlines()))
for id in ids:
if id in pmids: continue
nids.append(id)
outfile = open(options.library, "a")
else:
outfile = open(options.library, "w")
f = lambda id, x: outfile.write( "\n%s\n" % (str(x)))
PubMed.download_many( nids, f )
outfile.close()
print "added %i out of %i entries to library %s" % (len(nids), len(ids), options.library)
Experiment.Stop()
