def __init__(self, doc_dir):
convertion_style = "-raw"
self._eventextractor = EventExtractor(doc_dir)
parse = Parser(join(ROOT, 'templates', 'periodic.xml'))
self._template_metadata = parse.xml_template_metadata()
page = self._template_metadata['page']
self._preparator = Preparator(doc_dir)
self._raw_onepage_doc = self._preparator.raw_text_convertion(page, page, convertion_style)
self._linetokenized_onepage_doc = line_tokenize(self._raw_onepage_doc)
self._clean_onepage_doc = self._raw_onepage_doc.replace('\n', ' ')
class PeriodicExtractor(object):
def __init__(self, doc_dir):
convertion_style = "-raw"
self._eventextractor = EventExtractor(doc_dir)
parse = Parser(join(ROOT, 'templates', 'periodic.xml'))
self._template_metadata = parse.xml_template_metadata()
page = self._template_metadata['page']
self._preparator = Preparator(doc_dir)
self._raw_onepage_doc = self._preparator.raw_text_convertion(page, page, convertion_style)
self._linetokenized_onepage_doc = line_tokenize(self._raw_onepage_doc)
self._clean_onepage_doc = self._raw_onepage_doc.replace('\n', ' ')
## Event authors metadata extractor extends method to periodic author extractor
def _author_metadata(self):
self.authors = self._eventextractor._author_metadata()
return self.authors
def _abstract_metadata(self):
regex = re.compile(r'resumo:* (.*?) (palavr(a|as)(.|\s)chav(e|es).|unitermos|descritores)')
self.abstract = regex.search(self._clean_onepage_doc).group(1).strip().capitalize()
return self.abstract
def all_metadata(self):
if self._preparator.doc_ext == '.pdf':
try:
pdf_embed_metadata = self._preparator.pdf_embed_metadata()
self._pdf_num_pages = pdf_embed_metadata.numPages
except:
print 'Encripted document'
self._pdf_num_pages = 0
else:
self._pdf_num_pages = 0
metadata = {'author_metadata': self._author_metadata(),
'abstract_metadata': self._abstract_metadata(),
'number_pages': self._pdf_num_pages
}
try:
self._preparator.remove_converted_document()
except OSError:
print 'Temporary document already removed..'
return metadata
class TestEventExtractor(unittest.TestCase):
def setUp(self):
self.doc_dir = join(ROOT_PATH, "testdocs", "event", "1_pt-br.pdf")
self.preparator = Preparator(self.doc_dir)
self.extractor = EventExtractor(self.doc_dir)
self.parse = Parser("event.xml")
self.xml_template_metadata = self.parse.xml_template_metadata()
def test_metadata_extractor_generates_metadata_dict(self):
self.extractor.all_metadata() | should_not | be_empty
def test_event_document_has_an_abstract_metadata_pattern_found_by_regex(self):
doc = self.extractor._clean_onepage_doc
matches = re.search(r"resumo:* (.*?) (palavr(a|as)(.|\s)chav(e|es).|unitermos|descritores)", doc)
matches.group() | should | start_with("resumo")
self.extractor._abstract_metadata | should_not | equal_to("")
def test_event_document_has_author_type_metadata(self):
self.extractor._author_metadata() | should_not | be_empty
def test_event_document_has_title_type_metadata(self):
self.extractor._title_metadata() | should_not | be_empty
def setUp(self):
self.doc_dir = join(ROOT_PATH, 'testdocs', 'event', '1_pt-br.pdf')
self.preparator = Preparator(self.doc_dir)
self.extractor = EventExtractor(self.doc_dir)
self.parse = Parser('event.xml')
self.xml_template_metadata = self.parse.xml_template_metadata()
def setUp(self):
self.doc_dir = join(ROOT_PATH, "testdocs", "event", "1_pt-br.pdf")
self.preparator = Preparator(self.doc_dir)
self.extractor = EventExtractor(self.doc_dir)
self.parse = Parser("event.xml")
self.xml_template_metadata = self.parse.xml_template_metadata()
