def to_unicode(self, file_content):
encoding_info = chardet.detect(file_content)
logging.info("[Sender] file encoding detection result: %s",
encoding_info)
if not encoding_info or "encoding" not in encoding_info:
logging.info(
"[Sender] failed to detect file encoding, fall back to default encoding"
)
return file_content.decode(_DEFAULT_ENCODING_ASSUMPTION).encode(
"utf-8")
encoding = encoding_info.get("encoding").lower()
if encoding.startswith("utf"):
logging.info("[Sender] file in utf already, no need to convert")
return file_content
for parent_encoding, child_encodings in _ENCODING_MAP.iteritems():
if encoding in child_encodings:
logging.info("[Sender] switch to parent encoding %s",
parent_encoding)
return file_content.decode(parent_encoding).encode("utf-8")
try:
return file_content.decode(encoding).encode("utf-8")
except Exception:
logging.error(traceback.format_exc())
logging.info(
"[Sender] failed to convert from %s to utf-8, fall back to default coding",
encoding)
return file_content.decode(_DEFAULT_ENCODING_ASSUMPTION).encode(
"utf-8")
def makeUnicode(string,encoding='utf-8'):
if isinstance(string,unicode): return string
try:
ret = unicode.decode(string,encoding)
if DEBUG: LOG('util.makeUnicode(): Successfully used: %s' % encoding)
return ret
except:
detected_encoding = chardet.detect(string)
if DEBUG: LOG('util.makeUnicode(): Detected encoding: %s' % detected_encoding)
try:
string = unicode(string,detected_encoding['encoding'])
return string
except:
return unicode(string,encoding,'replace')
def unicode(self, string, encoding=None):
if not encoding:
encoding = self._encoding
else:
encoding = "utf-8"
try:
return string.decode(encoding).encode("utf-8")
except:
detected_encoding = chardet.detect(string)
try:
string = string.decode(detected_encoding["encoding"])
self.updateEncoding(detected_encoding["encoding"], detected_encoding["confidence"])
return string.encode("utf-8")
except:
pass
return string.encode("utf-8", "replace")
def detect_with_chardet(self):
len_limit = 500 * 1024
dct = {}
if self.txt_byte:
dct = chardet.detect(self.txt_byte[:len_limit])
else:
f = open(self.filename, 'r')
detector = universaldetector.UniversalDetector()
numbytes = 0
for line_byte in f.xreadlines():
detector.feed(line_byte)
numbytes += len(line_byte)
if detector.done: break
if numbytes > len_limit: break # don't read forever
detector.close()
dct = detector.result
return dct.get('encoding')
def detect_with_chardet(self):
len_limit = 500*1024
dct = {}
if self.txt_byte:
dct = chardet.detect(self.txt_byte[:len_limit])
else:
f = open(self.filename, 'r')
detector = universaldetector.UniversalDetector()
numbytes = 0
for line_byte in f.xreadlines():
detector.feed(line_byte)
numbytes += len(line_byte)
if detector.done: break
if numbytes > len_limit: break # don't read forever
detector.close()
dct = detector.result
return dct.get('encoding')
def txt_to_html(self, input_txt, output):
"""
Convert txt string to file with name "output".
"""
charset = chardet.detect(input_txt)
temp = input_txt.decode(charset['encoding'])
#temp = temp.replace('&', '&')
#temp = temp.replace('<', '<')
#temp = temp.replace('>', '>')
#temp = temp.replace('"', '"')
#temp = temp.replace("'", '»')
paragraphs = temp.split('\n')
html_template = get_template("epuber-templates/html-template.xhtml")
html_context = template.Context({'paragraphs': paragraphs})
output_file = open(output, 'w')
output_file.write(html_template.render(html_context).encode("utf-8"))
output_file.close()
def convertHTMLCodes(html,FB=None,encoding=None):
if not encoding:
if FB:
encoding = FB.getEncoding()
else:
encoding = 'utf-8'
try:
html = html.decode(encoding).encode('utf-8')
return html
except:
detected_encoding = chardet.detect(html)
if util.DEBUG: util.LOG(detected_encoding)
try:
html = html.decode(detected_encoding['encoding']).encode('utf-8')
if FB: FB.updateEncoding(detected_encoding['encoding'],detected_encoding['confidence'])
except:
html = html.encode('utf-8','replace')
try:
html = re.sub('&#(\d{1,5});',cUConvert,html)
html = re.sub('&(\w+?);',cTConvert,html)
except:
pass
return html
def getTitleLink(link):
title = u''
des = u''
try:
response = webfetch(link, follow_redirects=True, deadline=30, validate_certificate=False)
except:
return None,None
if response == None:
return None,None
content = response.content
uri = urlparse.urlparse(link)
loc = uri[1]
if short(loc):
link = response.final_url or link
if not link.startswith('http') :
return None,None
if linkitem.all().filter('url =',link).filter('valid =',0).get() != None :
return None,None
# filter url like "aaa.com", no site url, focus article
uri = urlparse.urlparse(link)
if uri[1] in filtersite :
return None,None
if uri[4] == '':
if uri[2] == '':
return None,None
if uri[2] == '/':
return None,None
if uri[2] == '#':
return None,None
title = re.search('<title>([\S\s]*?)</title>', content, re.IGNORECASE)
if title == None:
return None,None
title = title.group(1)
if title=='':
title=link
if is_snake(title) : #or is_snake(content):
url = linkitem()
url.valid = 0
url.url = link
url.put()
return None,None
else:
url = linkitem()
url.valid = 1
url.url = link
url.put()
charset = 'utf-8'
if 'charset' in response.headers:
charset1 = response.headers['charset']
else:
charset1 = 'utf-8'
if len(content) >= (1 * 1024 * 1024):
return title, link
try:
content = content.decode(charset1, 'ignore')
charset = charset1
except:
try:
m_charset = re.search('<meta\s*http-equiv="?Content-Type"? content="text/html;\s*charset=([\w\d-]+?)"', content, re.IGNORECASE)
charset2 = m_charset.group(1)
charset = charset2
except:
try:
charset3 = chardet.detect(content)['encoding']
content = content.decode(charset3,'ignore')
charset = charset3
except:
try:
content = content.decode("utf-8",'ignore')
charset = 'utf-8'
except:
return None,None
title = title.decode(charset)
art = article()
art.title = title
art.txt = content
art.url = link
art.put()
return title, link
def getTitleLink(link):
title = u""
des = u""
try:
response = webfetch(link, follow_redirects=True, deadline=30, validate_certificate=False)
except:
return None, None
if response == None:
return None, None
content = response.content
uri = urlparse.urlparse(link)
loc = uri[1]
if short(loc):
link = response.final_url or link
if not link.startswith("http"):
return None, None
if linkitem.all().filter("url =", link).filter("valid =", 0).get() != None:
return None, None
# filter url like "aaa.com", no site url, focus article
uri = urlparse.urlparse(link)
if uri[1] in filtersite:
return None, None
if uri[4] == "":
if uri[2] == "":
return None, None
if uri[2] == "/":
return None, None
if uri[2] == "#":
return None, None
title = re.search("<title>([\S\s]*?)</title>", content, re.IGNORECASE)
if title == None:
return None, None
title = title.group(1)
if title == "":
return None, None
if title.startswith("Page Not Found"):
return None, None
if title.startswith("Oops! Something Bad Happened"):
return None, None
if is_snake(title): # or is_snake(content):
url = linkitem()
url.valid = 0
url.url = link
url.put()
return None, None
else:
url = linkitem()
url.valid = 1
url.url = link
url.put()
charset = "utf-8"
if "charset" in response.headers:
charset1 = response.headers["charset"]
else:
charset1 = "utf-8"
if len(content) >= (1 * 1024 * 1024):
return title, link
try:
content = content.decode(charset1, "ignore")
charset = charset1
except:
try:
m_charset = re.search(
'<meta\s*http-equiv="?Content-Type"? content="text/html;\s*charset=([\w\d-]+?)"', content, re.IGNORECASE
)
charset2 = m_charset.group(1)
charset = charset2
except:
try:
charset3 = chardet.detect(content)["encoding"]
content = content.decode(charset3, "ignore")
charset = charset3
except:
try:
content = content.decode("utf-8", "ignore")
charset = "utf-8"
except:
return None, None
title = title.decode(charset)
art = article()
art.title = title
art.txt = content
art.url = link
art.put()
return title, link