def check_tag(pipeline, sentence, tag, o1, o2): """Return True if sentence has tag between offsets o1 and o2.""" # NOTE: apart from the options this is the same function as in unittests.py td = tarsqi.process_string(sentence, pipeline=pipeline, loglevel='1') tags = td.tags.find_tags(tag) for t in tags: if t.begin == o1 and t.end == o2: return True return False, tags
def parse_text(text): doc = tarsqi.process_string(text, pipeline='PREPROCESSOR,EVITA,GUTIME') doc.sourcedoc.lif = lif.LIF(json_object={ "text": { "@value": text }, "views": [] }) doc.sourcedoc.lif.add_tarsqi_view(doc) purge_metadata(doc.sourcedoc.lif.views[0].metadata) return doc
def run_pipeline(pipeline, sentence): """Run the sentence through the pipeline and return the resulting TarsqiDocument. The process_string method does not trap errors.""" return tarsqi.process_string(sentence, pipeline=pipeline, loglevel=1)
def print_summary(): print "\nSUMMARY:\n" print " %-15s %-10s %-10s %s" % ('', ' PASS', ' FAIL', 'ERROR') for module, p, f, e in SUMMARY: print " %-15s %5s %5s %5s" % (module, p, f, e) print if f or e: print " THERE WERE FAILURES OR ERRORS\n" if __name__ == '__main__': # a dummy run just to get the tagger messages out of the way tarsqi.process_string("Fido barks.") run_gutime_tests = False run_evita_tests = False run_slinket_tests = False run_s2t_tests = False run_blinker_tests = False run_sputlink_tests = False components = ['evita', 'gutime', 'slinket', 's2t', 'blinker', 'sputlink'] opts, args = getopt.getopt(sys.argv[1:], '', components + ['show-errors']) for opt, val in opts: if opt == '--show-errors': SHOW_ERRORS = True if opt == '--evita': run_evita_tests = True if opt == '--gutime': run_gutime_tests = True if opt == '--slinket': run_slinket_tests = True
def load_tagger(): # a dummy run just to get the tagger messages out of the way #tarsqi.process_string("Fido barks.", [('--pipeline', 'PREPROCESSOR')]) tarsqi.process_string("Fido barks.", pipeline='PREPROCESSOR')