FilesExpand file tree

 master

/

merge-rulesets.py

Copy path

Blame

More file actions

Blame

More file actions

Latest commit

 

History

History

History

101 lines (83 loc) · 3.23 KB

 master

/

merge-rulesets.py

Copy path

Top

File metadata and controls

• Code
• Blame

101 lines (83 loc) · 3.23 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

#!/usr/bin/env python2.7

# Merge all the .xml rulesets into a single "default.rulesets" file -- this

# prevents inodes from wasting disk space, but more importantly, works around

# the fact that zip does not perform well on a pile of small files.

# currently a very literal translation of merge-rulesets.sh, but about five

# times faster

from __future__ import print_function

pass

import os

from glob import glob

from subprocess import call

import sys

import traceback

import re

import unicodedata

def normalize(f):

"""

OSX and Linux filesystems encode composite characters differently in filenames.

We should normalize to NFC: http://unicode.org/reports/tr15/.

"""

f = unicodedata.normalize('NFC', unicode(f, 'utf-8')).encode('utf-8')

return f

os.chdir("src")

rulesets_fn="chrome/content/rules/default.rulesets"

xml_ruleset_files = map(normalize, glob("chrome/content/rules/*.xml"))

# cleanup after bugs :/

misfile = rulesets_fn + "r"

if os.path.exists(misfile):

print("Cleaning up malformed rulesets file...")

os.unlink(misfile)

if "--fast" in sys.argv:

library_compiled_time = os.path.getmtime(rulesets_fn)

newest_xml = max([os.path.getmtime(f) for f in xml_ruleset_files])

if library_compiled_time >= newest_xml:

print("Library is newer that all rulesets, skipping rebuild...")

sys.exit(0)

print("Creating ruleset library...")

# Under git bash, sed -i issues errors and sets the file "read only". Thanks.

if os.path.isfile(rulesets_fn):

os.system("chmod u+w " + rulesets_fn)

def rulesize():

return len(open(rulesets_fn).read())

def clean_up(rulefile):

"""Remove extra whitespace, comments and tests from a ruleset"""

comment_and_newline_pattern = re.compile(r"<!--.*?-->|\n|\r", flags=re.DOTALL)

rulefile = comment_and_newline_pattern.sub('', rulefile)

to_and_from_pattern = re.compile(r'\s*(from=)')

rulefile = to_and_from_pattern.sub(r' \1', rulefile)

rulefile = re.sub(r'"\s*(to=)', r'" \1', rulefile)

rulefile = re.sub(r">\s*<", r"><", rulefile)

rulefile = re.sub(r"</ruleset>\s*", r"</ruleset>\n", rulefile)

rulefile = re.sub(r"\s*(/>|<ruleset)", r"\1", rulefile)

rulefile = re.sub(r"<test.+?/>", r"", rulefile)

return rulefile

library = open(rulesets_fn,"w")

try:

commit_id = os.environ["GIT_COMMIT_ID"]

library.write('<rulesetlibrary gitcommitid="%s">' % commit_id)

except:

# Chromium

library.write('<rulesetlibrary>')

# Include the filename.xml as the "f" attribute

print("Removing whitespaces and comments...")

for rfile in sorted(xml_ruleset_files):

ruleset = open(rfile).read()

fn = os.path.basename(rfile)

ruleset = ruleset.replace("<ruleset", '<ruleset f="%s"' % fn, 1)

library.write(clean_up(ruleset))

library.write("</rulesetlibrary>\n")

library.close()

try:

if 0 == call(["xmllint", "--noout", rulesets_fn]):

print(rulesets_fn, "passed XML validity test.")

else:

print("ERROR:", rulesets_fn, "failed XML validity test!")

sys.exit(1)

except OSError as e:

if "No such file or directory" not in traceback.format_exc():

raise

print("WARNING: xmllint not present; validation of", rulesets_fn, " skipped.")

# We make default.rulesets at build time, but it shouldn't have a variable

# timestamp

call(["touch", "-r", "install.rdf", rulesets_fn])