def parse(self, doc):
# FIXME: don't create these if they already exists
self.lagrum_parser = LegalRef(LegalRef.LAGRUM)
self.rattsfall_parser = LegalRef(LegalRef.RATTSFALL)
docfile = self.store.downloaded_path(doc.basefile)
intermediatefile = self.store.intermediate_path(doc.basefile)
r = WordReader()
intermediatefile, filetype = r.read(docfile, intermediatefile)
with codecs.open(intermediatefile, encoding="utf-8") as fp:
patchedtext, patchdesc = self.patch_if_needed(doc.basefile,
fp.read())
# The second step is to mangle the crappy XML produced by
# antiword (docbook) or Word 2007 (OOXML) into a nice pair of
# structures. rawhead is a simple dict that we'll later transform
# into a rdflib Graph. rawbody is a list of plaintext strings, each
# representing a paragraph.
#
# long-term FIXME: WordReader should expose a unified
# interface for handling both kinds of word files so that we
# wouldn't need both parse_ooxml() and
# parse_antiword_docbook(). This might require some other tool
# than antiword for old .doc files, as this throws away a LOT
# of info.
if filetype == "docx":
rawhead, rawbody = self.parse_ooxml(patchedtext, doc.basefile)
else:
rawhead, rawbody = self.parse_antiword_docbook(patchedtext, doc.basefile)
doc.uri = self.polish_metadata(rawhead, doc)
if patchdesc:
doc.meta.add((URIRef(doc.uri),
self.ns['ferenda'].patchdescription,
patchdesc))
doc.body = self.format_body(rawbody) # FIXME: Write a
class Read(unittest.TestCase):
def setUp(self):
self.maxDiff = None
self.datadir = tempfile.mkdtemp()
self.reader = WordReader()
def tearDown(self):
shutil.rmtree(self.datadir)
def test_doc(self):
path = self.datadir + os.sep + "out.xml"
out, type = self.reader.read("test/files/wordreader/sample.doc",
path)
self.assertEqual(out, path)
self.assertEqual(type, "doc")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("book", tree.getroot().tag)
xpath = '//*[contains(text(), "simple document in .doc format")]'
self.assertTrue(tree.getroot().xpath(xpath))
# test that spaces in filename work (requires more cmdline quoting)
os.unlink(path)
out, type = self.reader.read("test/files/wordreader/spaces in filename.doc",
path)
self.assertEqual(out, path)
self.assertEqual(type, "doc")
def test_docx(self):
path = self.datadir + os.sep + "out.xml"
out, type = self.reader.read("test/files/wordreader/sample.docx",
path)
self.assertEqual(out, path)
self.assertEqual(type, "docx")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("{http://schemas.openxmlformats.org/wordprocessingml/2006/main}document",
tree.getroot().tag)
xpath = '//*[contains(text(), "simple document in OOXML (.docx) format")]'
self.assertTrue(tree.getroot().xpath(xpath))
def test_mislabeled(self):
path = self.datadir + os.sep + "out.xml"
out, type = self.reader.read("test/files/wordreader/mislabeled.doc",
path)
self.assertEqual(out, path)
self.assertEqual(type, "docx")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("{http://schemas.openxmlformats.org/wordprocessingml/2006/main}document",
tree.getroot().tag)
xpath = '//*[contains(text(), "mis-labeled as a .doc file")]'
self.assertTrue(tree.getroot().xpath(xpath))
def setUp(self):
self.maxDiff = None
self.datadir = tempfile.mkdtemp()
self.reader = WordReader()
class Read(unittest.TestCase):
def setUp(self):
self.maxDiff = None
self.datadir = tempfile.mkdtemp()
self.reader = WordReader()
def tearDown(self):
shutil.rmtree(self.datadir)
def test_doc(self):
path = self.datadir + os.sep + "out.xml"
try:
with open(path, "wb") as fp:
filetype = self.reader.read("test/files/wordreader/sample.doc",
fp)
self.assertEqual(filetype, "doc")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("book", tree.getroot().tag)
xpath = '//*[contains(text(), "simple document in .doc format")]'
self.assertTrue(tree.getroot().xpath(xpath))
# test that spaces in filename work (requires more cmdline quoting)
os.unlink(path)
with open(path, "wb") as fp:
filetype = self.reader.read("test/files/wordreader/spaces in filename.doc",
fp)
self.assertEqual(filetype, "doc")
except ExternalCommandError as e:
raise unittest.SkipTest("Antiword does not seem to be installed")
def test_docx(self):
path = self.datadir + os.sep + "out.xml"
with open(path, "wb") as fp:
filetype = self.reader.read("test/files/wordreader/sample.docx",
fp)
self.assertEqual(filetype, "docx")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("{http://schemas.openxmlformats.org/wordprocessingml/2006/main}document",
tree.getroot().tag)
xpath = '//*[contains(text(), "simple document in OOXML (.docx) format")]'
self.assertTrue(tree.getroot().xpath(xpath))
def test_mislabeled(self):
path = self.datadir + os.sep + "out.xml"
try:
with silence():
with open(path, "wb") as fp:
filetype = self.reader.read("test/files/wordreader/mislabeled.doc",
fp)
self.assertEqual(filetype, "docx")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("{http://schemas.openxmlformats.org/wordprocessingml/2006/main}document",
tree.getroot().tag)
xpath = '//*[contains(text(), "mis-labeled as a .doc file")]'
self.assertTrue(tree.getroot().xpath(xpath))
except ExternalCommandError as e:
raise unittest.SkipTest("Antiword does not seem to be installed")
def setUp(self):
self.maxDiff = None
self.datadir = tempfile.mkdtemp()
self.reader = WordReader()
class Read(unittest.TestCase):
def setUp(self):
self.maxDiff = None
self.datadir = tempfile.mkdtemp()
self.reader = WordReader()
def tearDown(self):
shutil.rmtree(self.datadir)
def test_doc(self):
path = self.datadir + os.sep + "out.xml"
try:
with open(path, "wb") as fp:
filetype = self.reader.read("test/files/wordreader/sample.doc",
fp)
self.assertEqual(filetype, "doc")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("book", tree.getroot().tag)
xpath = '//*[contains(text(), "simple document in .doc format")]'
self.assertTrue(tree.getroot().xpath(xpath))
# test that spaces in filename work (requires more cmdline quoting)
os.unlink(path)
with open(path, "wb") as fp:
filetype = self.reader.read("test/files/wordreader/spaces in filename.doc",
fp)
self.assertEqual(filetype, "doc")
except ExternalCommandError as e:
raise unittest.SkipTest("Antiword does not seem to be installed")
def test_docx(self):
path = self.datadir + os.sep + "out.xml"
with open(path, "wb") as fp:
filetype = self.reader.read("test/files/wordreader/sample.docx",
fp)
self.assertEqual(filetype, "docx")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("{http://schemas.openxmlformats.org/wordprocessingml/2006/main}document",
tree.getroot().tag)
xpath = '//*[contains(text(), "simple document in OOXML (.docx) format")]'
self.assertTrue(tree.getroot().xpath(xpath))
def test_mislabeled(self):
path = self.datadir + os.sep + "out.xml"
try:
with silence():
with open(path, "wb") as fp:
filetype = self.reader.read("test/files/wordreader/mislabeled.doc",
fp)
self.assertEqual(filetype, "docx")
self.assertTrue(os.path.exists(path))
tree = etree.parse(path)
self.assertEqual("{http://schemas.openxmlformats.org/wordprocessingml/2006/main}document",
tree.getroot().tag)
xpath = '//*[contains(text(), "mis-labeled as a .doc file")]'
self.assertTrue(tree.getroot().xpath(xpath))
except ExternalCommandError as e:
raise unittest.SkipTest("Antiword does not seem to be installed")
