def _convert_row(doc_date, xml):
"""
Given a default date (for the whole document) and a (date, entry)
row, return
* a (partial) iso string for the date
* the text for the entry
"""
ths = list(xml.iter('th'))
tds = list(xml.iter('td'))
if len(ths) < 1:
date = None
elif len(ths) > 1:
ET.dump(xml)
raise Exception("Did not expect more than one th node")
else:
th_text = _clean_date(ths[0].text or "")
date = read_date(th_text, prefix=doc_date, fuzzy=True) # or doc_date
columns = ths + tds
text = "\n".join(_column_to_text(x) for x in columns)
if text:
return date or doc_date, text
else:
return None
def _convert_row(doc_date, xml):
"""
Given a default date (for the whole document) and a (date, entry)
row, return
* a (partial) iso string for the date
* the text for the entry
"""
ths = list(xml.iter('th'))
tds = list(xml.iter('td'))
if len(ths) < 1:
date = None
elif len(ths) > 1:
ET.dump(xml)
raise Exception("Did not expect more than one th node")
else:
th_text = _clean_date(ths[0].text or "")
date = read_date(th_text, prefix=doc_date, fuzzy=True) # or doc_date
columns = ths + tds
text = "\n".join(_column_to_text(x) for x in columns)
if text:
return date or doc_date, text
else:
return None
def _convert_section(xml):
"""
string representation of entire document
(WARNING: mutates the tree)
"""
some = lambda l: [x for x in l if x is not None]
dates = some(read_date(_clean_date(x.text)) for x in xml.iter('head'))
section_date = dates[0] if dates else None
for br_node in xml.iter('br'):
br_node.text = "\n"
return [_convert_row(section_date, r) for r in xml.iter('tr')]
def _convert_section(xml):
"""
string representation of entire document
(WARNING: mutates the tree)
"""
some = lambda l: [x for x in l if x is not None]
dates = some(read_date(_clean_date(x.text))
for x in xml.iter('head'))
section_date = dates[0] if dates else None
for br_node in xml.iter('br'):
br_node.text = "\n"
return [_convert_row(section_date, r) for r in xml.iter('tr')]
def _write_membrane(ntext, oprefix):
"""
Write out text for an individual membrane
(may involve multiple files)
"""
lines = ntext.split("\n")
mname = _membrane_name(lines[0])
subentries = lines[1:]
digits = _digits(subentries)
for i, line in enumerate(subentries):
date_str = " ".join(line.split()[:3])
try:
date = read_date(date_str, fuzzy=True)
except ValueError as _:
date = None
filename = "-".join([oprefix, mname, str(i + 1).zfill(digits)])
with codecs.open(filename, 'w', 'utf-8') as stream:
if date is not None:
print(date, file=stream)
print(line, file=stream)
def _write_membrane(ntext, oprefix):
"""
Write out text for an individual membrane
(may involve multiple files)
"""
lines = ntext.split("\n")
mname = _membrane_name(lines[0])
subentries = lines[1:]
digits = _digits(subentries)
for i, line in enumerate(subentries):
date_str = " ".join(line.split()[:3])
try:
date = read_date(date_str, fuzzy=True)
except ValueError as _:
date = None
filename = "-".join([oprefix,
mname,
str(i+1).zfill(digits)])
with codecs.open(filename, 'w', 'utf-8') as stream:
if date is not None:
print(date, file=stream)
print(line, file=stream)
def assertDateEqual(self, expected, dstr, **kwargs):
"assert that date string parses as expected"
self.assertEqual(expected, read_date(dstr, **kwargs))
def assertDateEqual(self, expected, dstr, **kwargs):
"assert that date string parses as expected"
self.assertEqual(expected, read_date(dstr, **kwargs))