FilesExpand file tree

 develop

/

htmls.py

Copy path

Blame

More file actions

Blame

More file actions

Latest commit

 

History

History

History

115 lines (95 loc) · 3.46 KB

 develop

/

htmls.py

Copy path

Top

File metadata and controls

• Code
• Blame

115 lines (95 loc) · 3.46 KB

Raw

Copy raw file

Download raw file

Open symbols panel

Edit and raw actions

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

from cleaners import normalize_spaces, clean_attributes

from encoding import get_encoding

from lxml.html import tostring

import logging

import lxml.html

import re

logging.getLogger().setLevel(logging.DEBUG)

utf8_parser = lxml.html.HTMLParser(encoding='utf-8')

def build_doc(page):

if isinstance(page, unicode):

page_unicode = page

else:

enc = get_encoding(page)

page_unicode = page.decode(enc, 'replace')

doc = lxml.html.document_fromstring(page_unicode.encode('utf-8', 'replace'), parser=utf8_parser)

return doc

def js_re(src, pattern, flags, repl):

return re.compile(pattern, flags).sub(src, repl.replace('$', '\\'))

def normalize_entities(cur_title):

entities = {

u'\u2014':'-',

u'\u2013':'-',

u'—': '-',

u'–': '-',

u'\u00A0': ' ',

u'\u00AB': '"',

u'\u00BB': '"',

u'"': '"',

}

for c, r in entities.iteritems():

if c in cur_title:

cur_title = cur_title.replace(c, r)

return cur_title

def norm_title(title):

return normalize_entities(normalize_spaces(title))

def get_title(doc):

title = doc.find('.//title')

if title is None or len(title.text) == 0:

return '[no-title]'

return norm_title(title.text)

def add_match(collection, text, orig):

text = norm_title(text)

if len(text.split()) >= 2 and len(text) >= 15:

if text.replace('"', '') in orig.replace('"', ''):

collection.add(text)

def shorten_title(doc):

title = doc.find('.//title')

if title is None or len(title.text) == 0:

return ''

title = orig = norm_title(title.text)

candidates = set()

for item in ['.//h1', './/h2', './/h3']:

for e in list(doc.iterfind(item)):

if e.text:

add_match(candidates, e.text, orig)

if e.text_content():

add_match(candidates, e.text_content(), orig)

for item in ['#title', '#head', '#heading', '.pageTitle', '.news_title', '.title', '.head', '.heading', '.contentheading', '.small_header_red']:

for e in doc.cssselect(item):

if e.text:

add_match(candidates, e.text, orig)

if e.text_content():

add_match(candidates, e.text_content(), orig)

if candidates:

title = sorted(candidates, key=len)[-1]

else:

for delimiter in [' | ', ' - ', ' :: ', ' / ']:

if delimiter in title:

parts = orig.split(delimiter)

if len(parts[0].split()) >= 4:

title = parts[0]

break

elif len(parts[-1].split()) >= 4:

title = parts[-1]

break

else:

if ': ' in title:

parts = orig.split(': ')

if len(parts[-1].split()) >= 4:

title = parts[-1]

else:

title = orig.split(': ', 1)[1]

if not 15 < len(title) < 150:

return orig

return title

def get_body(doc):

[ elem.drop_tree() for elem in doc.xpath('.//script | .//link | .//style') ]

raw_html = unicode(tostring(doc.body or doc))

cleaned = clean_attributes(raw_html)

try:

#BeautifulSoup(cleaned) #FIXME do we really need to try loading it?

return cleaned

except Exception: #FIXME find the equivalent lxml error

logging.error("cleansing broke html content: %s\n---------\n%s" % (raw_html, cleaned))

return raw_html