def test_output_results_option_when_false(self):
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor, output_results=False)
runner.run('test', output_dir = self.results_dir)
result_file_path = os.path.join(self.results_dir, 'SelfExtractor.xml')
self.assertFalse(os.path.isfile(result_file_path))
def test_output_results_option_defaults_to_true(self):
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor)
runner.run('test', output_dir = self.results_dir)
result_file_path = os.path.join(self.results_dir, 'SelfExtractor.xml')
self.assertTrue(os.path.isfile(result_file_path))
def test_file_prefix_option_works(self):
runner = ExtractionRunner()
runner.add_runnable(ImplTestFileExtractor)
runner.run('whatever', output_dir=self.results_dir, file_prefix = 'prefix.')
result_file_path = os.path.join(self.results_dir, 'prefix.test.txt')
self.assertTrue(os.path.isfile(result_file_path))
self.assertEqual(open(result_file_path, 'r').read(), 'test test')
def test_file_name_result_works(self):
runner = ExtractionRunner()
runner.add_runnable(SelfChangeNameExtractor)
runner.run('pizza', output_dir=self.results_dir)
result_file_path = os.path.join(self.results_dir, SelfChangeNameExtractor.result_file_name)
self.assertTrue(os.path.isfile(result_file_path))
self.assertEqual(ET.parse(result_file_path).getroot().text, 'pizza')
def test_run_from_file(self):
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor)
runner.run_from_file(self.f1_path, output_dir=self.results_dir)
result_file_path = os.path.join(self.results_dir, 'SelfExtractor.xml')
self.assertTrue(os.path.isfile(result_file_path))
xml = ET.parse(result_file_path).getroot()
self.assertEqual(xml.text, 'file 1')
def test_dependency_results_work(self):
runner = ExtractionRunner()
runner.add_runnable(ImplTestFileExtractor)
runner.add_runnable(DepsOnTestFileExtractor)
# an error will be thrown if dependency doesn't work
# so no need to assert anything in this test
runner.run('whatever', output_dir=self.results_dir)
runner = ExtractionRunner()
runner.add_runnable(DepsOnTestFileExtractor)
self.assertRaises(LookupError, runner.run, 'whatever', output_dir=self.results_dir)
def test_run_batch(self):
batch = ['test 0', 'test 1', 'test 2']
prefixes = ['1', '2', '3']
output_dirs = [self.results_dir] * 3
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor)
runner.run_batch(batch, output_dirs, file_prefixes=prefixes)
for prefix, text in zip(prefixes, batch):
result_file_path = os.path.join(self.results_dir, '{0}SelfExtractor.xml'.format(prefix))
self.assertTrue(os.path.isfile(result_file_path))
xml = ET.parse(result_file_path).getroot()
self.assertEqual(xml.text, text)
def test_disable_logs_works(self):
runner = ExtractionRunner()
results_log_path = os.path.join(self.results_dir, 'results')
runnables_log_path = os.path.join(self.results_dir, 'runnables')
runner.enable_logging(results_log_path, runnables_log_path)
runner.disable_logging()
runner.add_runnable(SelfLogExtractor)
runner.run('abc', output_dir=self.results_dir, run_name='RUN!')
log_list = glob.glob(results_log_path + "*.log")
self.assertFalse(log_list)
log_list = glob.glob(runnables_log_path + "*.log")
self.assertFalse(log_list)
def test_run_from_file_batch(self):
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor)
paths = [self.f1_path, self.f2_path, self.f3_path]
prefixes = ['1', '2', '3']
output_dirs = [self.results_dir] * 3
runner.run_from_file_batch(paths, output_dirs, file_prefixes=prefixes)
result_file_paths = [os.path.join(self.results_dir, '{0}SelfExtractor.xml'.format(i)) for i in prefixes]
file_content = ['file {0}'.format(i) for i in prefixes]
for path, content in zip(result_file_paths, file_content):
self.assertTrue(os.path.isfile(path))
xml = ET.parse(path).getroot()
self.assertEqual(xml.text, content)
def test_extractor_errors_cascade_no_write_dep_errors(self):
runner = ExtractionRunner()
runner.add_runnable(ErrorExtractor)
runner.add_runnable(DepsOnErrorExtractor)
runner.add_runnable(DepsOnErrorExtractor2)
runner.run('Test', output_dir=self.results_dir)
ee_path = os.path.join(self.results_dir, 'ErrorExtractor.xml')
self.assertTrue(os.path.isfile(ee_path))
self.assertEqual(ET.parse(ee_path).getroot().tag, 'error')
doee_path = os.path.join(self.results_dir, 'DepsOnErrorExtractor.xml')
self.assertFalse(os.path.isfile(doee_path))
doee2_path = os.path.join(self.results_dir,
'DepsOnErrorExtractor2.xml')
def test_extractor_errors_cascade_no_write_dep_errors(self):
runner = ExtractionRunner()
runner.add_runnable(ErrorExtractor)
runner.add_runnable(DepsOnErrorExtractor)
runner.add_runnable(DepsOnErrorExtractor2)
runner.run('Test', output_dir = self.results_dir)
ee_path = os.path.join(self.results_dir, 'ErrorExtractor.xml')
self.assertTrue(os.path.isfile(ee_path))
self.assertEqual(ET.parse(ee_path).getroot().tag, 'error')
doee_path = os.path.join(self.results_dir, 'DepsOnErrorExtractor.xml')
self.assertFalse(os.path.isfile(doee_path))
doee2_path = os.path.join(self.results_dir, 'DepsOnErrorExtractor2.xml')
def test_disable_logs_works(self):
runner = ExtractionRunner()
results_log_path = os.path.join(self.results_dir, 'results')
runnables_log_path = os.path.join(self.results_dir, 'runnables')
runner.enable_logging(results_log_path, runnables_log_path)
runner.disable_logging()
runner.add_runnable(SelfLogExtractor)
runner.run('abc', output_dir = self.results_dir, run_name = 'RUN!')
log_list = glob.glob(results_log_path + "*.log")
self.assertFalse(log_list)
log_list = glob.glob(runnables_log_path + "*.log")
self.assertFalse(log_list)
def test_output_results_option_when_false(self):
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor, output_results=False)
runner.run('test', output_dir=self.results_dir)
result_file_path = os.path.join(self.results_dir, 'SelfExtractor.xml')
self.assertFalse(os.path.isfile(result_file_path))
def test_output_results_option_defaults_to_true(self):
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor)
runner.run('test', output_dir=self.results_dir)
result_file_path = os.path.join(self.results_dir, 'SelfExtractor.xml')
self.assertTrue(os.path.isfile(result_file_path))
def test_files_get_written(self):
runner = ExtractionRunner()
runner.add_runnable(ImplTestFileExtractor)
runner.run('whatever', output_dir=self.results_dir)
result_file_path = os.path.join(self.results_dir, 'test.txt')
self.assertTrue(os.path.isfile(result_file_path))
self.assertEqual(open(result_file_path, 'r').read(), 'test test')
def test_file_name_result_works(self):
runner = ExtractionRunner()
runner.add_runnable(SelfChangeNameExtractor)
runner.run('pizza', output_dir=self.results_dir)
result_file_path = os.path.join(
self.results_dir, SelfChangeNameExtractor.result_file_name)
self.assertTrue(os.path.isfile(result_file_path))
self.assertEqual(ET.parse(result_file_path).getroot().text, 'pizza')
def test_run_from_file(self):
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor)
runner.run_from_file(self.f1_path, output_dir=self.results_dir)
result_file_path = os.path.join(self.results_dir, 'SelfExtractor.xml')
self.assertTrue(os.path.isfile(result_file_path))
xml = ET.parse(result_file_path).getroot()
self.assertEqual(xml.text, 'file 1')
def test_logs_work(self):
runner = ExtractionRunner()
results_log_path = os.path.join(self.results_dir, 'results')
runnables_log_path = os.path.join(self.results_dir, 'runnables')
runner.enable_logging(results_log_path, runnables_log_path)
runner.add_runnable(SelfLogExtractor)
runner.run('abc', output_dir=self.results_dir, run_name='RUN!')
results_log = glob.glob(results_log_path + "*.log")[0]
log_data = open(results_log, 'r').read()
self.assertTrue('[SUCCESS]' in log_data)
self.assertTrue('RUN!' in log_data)
runnables_log = glob.glob(runnables_log_path + "*.log")[0]
log_data = open(runnables_log, 'r').read()
self.assertTrue('abc' in log_data)
self.assertTrue('SelfLogExtractor' in log_data)
self.assertTrue('RUN!' in log_data)
def test_run_batch(self):
batch = ['test 0', 'test 1', 'test 2']
prefixes = ['1', '2', '3']
output_dirs = [self.results_dir] * 3
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor)
runner.run_batch(batch, output_dirs, file_prefixes=prefixes)
for prefix, text in zip(prefixes, batch):
result_file_path = os.path.join(
self.results_dir, '{0}SelfExtractor.xml'.format(prefix))
self.assertTrue(os.path.isfile(result_file_path))
xml = ET.parse(result_file_path).getroot()
self.assertEqual(xml.text, text)
def test_logs_work(self):
runner = ExtractionRunner()
results_log_path = os.path.join(self.results_dir, 'results')
runnables_log_path = os.path.join(self.results_dir, 'runnables')
runner.enable_logging(results_log_path, runnables_log_path)
runner.add_runnable(SelfLogExtractor)
runner.run('abc', output_dir = self.results_dir, run_name = 'RUN!')
results_log = glob.glob(results_log_path + "*.log")[0]
log_data = open(results_log, 'r').read()
self.assertTrue('[SUCCESS]' in log_data)
self.assertTrue('RUN!' in log_data)
runnables_log = glob.glob(runnables_log_path + "*.log")[0]
log_data = open(runnables_log, 'r').read()
self.assertTrue('abc' in log_data)
self.assertTrue('SelfLogExtractor' in log_data)
self.assertTrue('RUN!' in log_data)
def test_run_from_file_batch(self):
runner = ExtractionRunner()
runner.add_runnable(SelfExtractor)
paths = [self.f1_path, self.f2_path, self.f3_path]
prefixes = ['1', '2', '3']
output_dirs = [self.results_dir] * 3
runner.run_from_file_batch(paths, output_dirs, file_prefixes=prefixes)
result_file_paths = [
os.path.join(self.results_dir, '{0}SelfExtractor.xml'.format(i))
for i in prefixes
]
file_content = ['file {0}'.format(i) for i in prefixes]
for path, content in zip(result_file_paths, file_content):
self.assertTrue(os.path.isfile(path))
xml = ET.parse(path).getroot()
self.assertEqual(xml.text, content)
def get_extraction_runner(modules):
runner = ExtractionRunner()
if modules['fulltext'] == 'True':
if modules['fulltext_pdfbox'] == 'True':
runner.add_runnable(pdfbox.PDFBoxPlainTextExtractor)
if modules['academicfilter'] == 'True':
runner.add_runnable(filters.AcademicPaperFilter)
if modules['fulltext'] == 'True':
if modules['fulltext_grobid'] == 'True':
runner.add_runnable(grobid.GrobidTEIExtractor)
if modules['fulltext_tei_to_csx'] == 'True':
runner.add_runnable(tei.TEItoPlainTextExtractor)
if modules['header'] == 'True':
if modules['header_grobid'] == 'True':
runner.add_runnable(grobid.GrobidHeaderTEIExtractor)
if modules['header_tei_to_csx'] == 'True':
runner.add_runnable(tei.TEItoHeaderExtractor)
if modules['citation'] == 'True':
if modules['citation_parscit'] == 'True':
runner.add_runnable(parscit.ParsCitCitationExtractor)
if modules['citation_grobid'] == 'True':
runner.add_runnable(grobid.GrobidCitationTEIExtractor)
if modules['figures'] == 'True':
runner.add_runnable(figures.PDFFiguresExtractor)
if modules['algorithms'] == 'True':
runner.add_runnable(algorithms.AlgorithmsExtractor)
return runner
def test_no_extraction_result_works(self):
runner = ExtractionRunner()
runner.add_runnable(NothingExtractor)
runner.run('pizza', output_dir=self.results_dir)
self.assertFalse(os.listdir(self.results_dir))
def test_filter_results_cascade(self):
runner = ExtractionRunner()
runner.add_runnable(FailFilter)
runner.add_runnable(FailingDepsExtractor)
runner.run('Test', output_dir=self.results_dir)
fde_path = os.path.join(self.results_dir, 'FailingDepsExtractor.xml')
self.assertFalse(os.path.isfile(fde_path))
runner.run('Test', output_dir=self.results_dir, write_dep_errors=True)
self.assertTrue(os.path.isfile(fde_path))
self.assertEqual(ET.parse(fde_path).getroot().tag, 'error')
runner = ExtractionRunner()
runner.add_runnable(PassFilter)
runner.add_runnable(PassingDepsExtractor)
runner.run('Test', output_dir=self.results_dir)
pde_path = os.path.join(self.results_dir, 'PassingDepsExtractor.xml')
self.assertTrue(os.path.isfile(pde_path))
self.assertEqual(ET.parse(pde_path).getroot().text, 'Test')
os.remove(pde_path)
runner.run('Test', output_dir=self.results_dir, write_dep_errors=True)
self.assertTrue(os.path.isfile(pde_path))
self.assertEqual(ET.parse(pde_path).getroot().text, 'Test')
def test_nothing_run(self):
runner = ExtractionRunner()
runner.run(u'data!', output_dir=self.results_dir)
# should be no files in output_dir
self.assertFalse(os.listdir(self.results_dir))
def get_extraction_runner(self):
runner = ExtractionRunner()
web.debug('getting runner')
#runner.enable_logging('/home/huy138/logs/service/results', '/home/huy138/logs/service/runnables')
web.debug('runner gotten')
return runner
def test_dependency_results_work(self):
runner = ExtractionRunner()
runner.add_runnable(ImplTestFileExtractor)
runner.add_runnable(DepsOnTestFileExtractor)
# an error will be thrown if dependency doesn't work
# so no need to assert anything in this test
runner.run('whatever', output_dir=self.results_dir)
runner = ExtractionRunner()
runner.add_runnable(DepsOnTestFileExtractor)
self.assertRaises(LookupError,
runner.run,
'whatever',
output_dir=self.results_dir)
def test_no_extraction_result_works(self):
runner = ExtractionRunner()
runner.add_runnable(NothingExtractor)
runner.run('pizza', output_dir=self.results_dir)
self.assertFalse(os.listdir(self.results_dir))
def get_extraction_runner(modules):
runner = ExtractionRunner()
if modules['fulltext'] == 'True':
if modules['fulltext_pdfbox'] == 'True':
runner.add_runnable(pdfbox.PDFBoxPlainTextExtractor)
if modules['fulltext_grobid'] == 'True':
runner.add_runnable(grobid.GrobidTEIExtractor)
if modules['fulltext_tei_to_csx'] == 'True':
runner.add_runnable(tei.TEItoPlainTextExtractor)
if modules['academicfilter'] == 'True':
runner.add_runnable(filters.AcademicPaperFilter)
if modules['header'] == 'True':
if modules['header_grobid'] == 'True':
runner.add_runnable(grobid.GrobidHeaderTEIExtractor)
if modules['header_tei_to_csx'] == 'True':
runner.add_runnable(tei.TEItoHeaderExtractor)
if modules['citation'] == 'True':
if modules['citation_parscit'] == 'True':
runner.add_runnable(parscit.ParsCitCitationExtractor)
if modules['citation_grobid'] == 'True':
runner.add_runnable(grobid.GrobidCitationTEIExtractor)
if modules['figures'] == 'True':
runner.add_runnable(figures.PDFFiguresExtractor)
if modules['algorithms'] == 'True':
runner.add_runnable(algorithms.AlgorithmsExtractor)
return runner
(status, stdout, stderr) = utils.external_process(['awk', '/^[0-9]/ {print;}', '-'], input_data=data, timeout=5)
except subprocess.TimeoutExpired:
raise RunnableError('awk timed out')
lines = [line for line in stdout.split("\n") if line]
root = ET.Element('extraction')
for line in lines:
ele = ET.SubElement(root, 'line')
ele.text = line
return ExtractorResult(xml_result=root)
# Set up and run extraction
extraction_runner = ExtractionRunner()
extraction_runner.add_runnable(HasNumbersFilter)
extraction_runner.add_runnable(EmailExtractor)
extraction_runner.add_runnable(LinesStartWithNumberExtractor)
extraction_runner.run(u'''Random data that contains some emails [email protected]
Test lines with some @ signs now and then. Meet you@[email protected].
Line with another email embedded [email protected] in the line.
[email protected] [email protected]
123 Some lines even start with numbers
Some lines don't start with numbers
004 The final line in the test data''', 'extraction/test/sample_output', run_name = 'Sample Data')
def get_extraction_runner():
runner = ExtractionRunner()
runner.enable_logging('~/logs/results', '~/logs/runnables')
runner.add_runnable(pdfbox.PDFBoxPlainTextExtractor)
runner.add_runnable(filters.AcademicPaperFilter)
runner.add_runnable(grobid.GrobidHeaderTEIExtractor)
runner.add_runnable(tei.TEItoHeaderExtractor)
runner.add_runnable(parscit.ParsCitCitationExtractor)
runner.add_runnable(figures.PDFFiguresExtractor)
runner.add_runnable(algorithms.AlgorithmsExtractor)
return runner
def test_nothing_run(self):
runner = ExtractionRunner()
runner.run(u'data!', output_dir=self.results_dir)
# should be no files in output_dir
self.assertFalse(os.listdir(self.results_dir))
def get_extraction_runner():
runner = ExtractionRunner()
runner.enable_logging('~/logs/results', '~/logs/runnables')
runner.add_runnable(pdfbox.PDFBoxPlainTextExtractor)
runner.add_runnable(filters.AcademicPaperFilter)
runner.add_runnable(grobid.GrobidHeaderTEIExtractor)
runner.add_runnable(tei.TEItoHeaderExtractor)
runner.add_runnable(parscit.ParsCitCitationExtractor)
runner.add_runnable(figures.PDFFiguresExtractor)
runner.add_runnable(algorithms.AlgorithmsExtractor)
return runner
(status, stdout, stderr) = utils.external_process(
['awk', '/^[0-9]/ {print;}', '-'], input_data=data, timeout=5)
except subprocess.TimeoutExpired:
raise RunnableError('awk timed out')
lines = [line for line in stdout.split("\n") if line]
root = ET.Element('extraction')
for line in lines:
ele = ET.SubElement(root, 'line')
ele.text = line
return ExtractorResult(xml_result=root)
# Set up and run extraction
extraction_runner = ExtractionRunner()
extraction_runner.add_runnable(HasNumbersFilter)
extraction_runner.add_runnable(EmailExtractor)
extraction_runner.add_runnable(LinesStartWithNumberExtractor)
extraction_runner.run(u'''Random data that contains some emails [email protected]
Test lines with some @ signs now and then. Meet you@[email protected].
Line with another email embedded [email protected] in the line.
[email protected] [email protected]
123 Some lines even start with numbers
Some lines don't start with numbers
004 The final line in the test data''',
'extraction/test/sample_output',
run_name='Sample Data')
def test_filter_results_cascade(self):
runner = ExtractionRunner()
runner.add_runnable(FailFilter)
runner.add_runnable(FailingDepsExtractor)
runner.run('Test', output_dir = self.results_dir)
fde_path = os.path.join(self.results_dir, 'FailingDepsExtractor.xml')
self.assertFalse(os.path.isfile(fde_path))
runner.run('Test', output_dir = self.results_dir, write_dep_errors=True)
self.assertTrue(os.path.isfile(fde_path))
self.assertEqual(ET.parse(fde_path).getroot().tag, 'error')
runner = ExtractionRunner()
runner.add_runnable(PassFilter)
runner.add_runnable(PassingDepsExtractor)
runner.run('Test', output_dir = self.results_dir)
pde_path = os.path.join(self.results_dir, 'PassingDepsExtractor.xml')
self.assertTrue(os.path.isfile(pde_path))
self.assertEqual(ET.parse(pde_path).getroot().text, 'Test')
os.remove(pde_path)
runner.run('Test', output_dir = self.results_dir, write_dep_errors=True)
self.assertTrue(os.path.isfile(pde_path))
self.assertEqual(ET.parse(pde_path).getroot().text, 'Test')
def get_extraction_runner():
runner = ExtractionRunner()
runner.enable_logging('~/logs/results', '~/logs/runnables')
# Option 1
runner.add_runnable(grobid.GrobidTEIExtractor)
runner.add_runnable(extractors.TEItoPlainTextExtractor)
runner.add_runnable(extractors.TEItoHeaderExtractor)
# OR
# Option 2
# runner.add_runnable(pdfbox.PDFBoxPlainTextExtractor)
runner.add_runnable(filters.AcademicPaperFilter)
return runner
def get_extraction_runner(modules):
runner = ExtractionRunner()
if modules["fulltext"] == "True":
if modules["fulltext_pdfbox"] == "True":
runner.add_runnable(pdfbox.PDFBoxPlainTextExtractor)
if modules["academicfilter"] == "True":
runner.add_runnable(filters.AcademicPaperFilter)
if modules["fulltext"] == "True":
if modules["fulltext_grobid"] == "True":
runner.add_runnable(grobid.GrobidTEIExtractor)
if modules["fulltext_tei_to_csx"] == "True":
runner.add_runnable(tei.TEItoPlainTextExtractor)
if modules["header"] == "True":
if modules["header_grobid"] == "True":
runner.add_runnable(grobid.GrobidHeaderTEIExtractor)
if modules["header_tei_to_csx"] == "True":
runner.add_runnable(tei.TEItoHeaderExtractor)
if modules["citation"] == "True":
if modules["citation_parscit"] == "True":
runner.add_runnable(parscit.ParsCitCitationExtractor)
if modules["citation_grobid"] == "True":
runner.add_runnable(grobid.GrobidCitationTEIExtractor)
if modules["figures"] == "True":
runner.add_runnable(figures.PDFFiguresExtractor)
if modules["algorithms"] == "True":
runner.add_runnable(algorithms.AlgorithmsExtractor)
return runner