def get_decoded_raw(name):
from calibre.ebooks.chardet import xml_to_unicode, force_encoding
with open(name, 'rb') as f:
raw = f.read()
syntax = syntax_from_mime(name, guess_type(name))
if syntax is None:
try:
raw = raw.decode('utf-8')
except ValueError:
pass
elif syntax != 'raster_image':
if syntax in {'html', 'xml'}:
raw = xml_to_unicode(raw, verbose=True)[0]
else:
m = re.search(br"coding[:=]\s*([-\w.]+)", raw[:1024], flags=re.I)
if m is not None and m.group(1) != '8bit':
enc = m.group(1)
if enc == b'unicode':
enc = 'utf-8'
else:
enc = force_encoding(raw, verbose=True)
try:
raw = raw.decode(enc)
except (LookupError, ValueError):
try:
raw = raw.decode('utf-8')
except ValueError:
pass
return raw, syntax
def get_decoded_raw(name):
from calibre.ebooks.chardet import xml_to_unicode, force_encoding
with open(name, 'rb') as f:
raw = f.read()
syntax = syntax_from_mime(name, guess_type(name))
if syntax is None:
try:
raw = raw.decode('utf-8')
except ValueError:
pass
elif syntax != 'raster_image':
if syntax in {'html', 'xml'}:
raw = xml_to_unicode(raw, verbose=True)[0]
else:
m = re.search(br"coding[:=]\s*([-\w.]+)", raw[:1024], flags=re.I)
if m is not None and m.group(1) != '8bit':
enc = m.group(1)
if enc == b'unicode':
enc = 'utf-8'
else:
enc = force_encoding(raw, verbose=True)
try:
raw = raw.decode(enc)
except (LookupError, ValueError):
try:
raw = raw.decode('utf-8')
except ValueError:
pass
return raw, syntax
def run(self, path_to_ebook):
print("reformatter: "+path_to_ebook)
f = open(path_to_ebook, 'r')
raw = f.read()
encoding = force_encoding(raw, True)
print("Detected encoding: ", encoding)
txt = unicode(raw, encoding, errors='replace')
# reformat
if prefs['reformat']:
print("reformatting...")
from ptxt2ftxt import ptxt2ftxt, ftxtclean
from ftxt2markdown import ftxt2markdown
txt = ptxt2ftxt(txt, para_by_mark=prefs['para_by_mark'])
txt = ftxtclean(txt, pretty_quote=prefs['pretty_quote'], correct_word_break=prefs['correct_word_break'])
txt = ftxt2markdown(txt, guessChapter=prefs['guess_chapter'], guessParaSep=prefs['insert_empty_paragraph'])
# save as temporary file
tempfile = self.temporary_file('.txt')
tempfile.write( txt.encode('utf-8') )
tempfile.close()
print("save as ", tempfile.name)
return tempfile.name
