Browse Source

Updated greper.

production
Michael Herzberg 5 years ago
parent
commit
77ddcc23e4
  1. 129
      grepper

129
grepper

@ -21,7 +21,6 @@ import os
import sys
import glob
import tarfile
import tempfile
import traceback
import jsbeautifier
from multiprocessing import Pool
@ -33,73 +32,38 @@ from ExtensionCrawler.config import *
from ExtensionCrawler.util import *
def get_name(overview_path):
if os.path.exists(overview_path):
with open(overview_path) as overview_file:
contents = overview_file.read()
# Extract extension name
match = re.search("""<meta itemprop="name" content="(.*?)"\s*/>""",
contents)
if match:
return match.group(1)
def get_downloads(overview_path):
if os.path.exists(overview_path):
with open(overview_path) as overview_file:
contents = overview_file.read()
match = re.search(
"""<meta itemprop="interactionCount" content="UserDownloads:((:?\d|,)+)""",
contents)
if match:
return int(match.group(1).replace(",", ''))
def help():
print("grepper [OPTION] BASEDIR GREP1 [GREP2...]")
print(" GREP1 [GREP2...] regex patterns")
print(" BASEDIR directory for output")
print(" -h print this help text")
print(" -a=<DIR> archive directory")
print(" -p=<PREFIX> three-letter-prefix")
print(" -t=<THREADS> number of parallel threads")
def process_date(extdir, ext_id, date, greps, out_f):
overview_path = os.path.join(extdir, "overview.html")
crxpath = next(iter(glob.glob(os.path.join(extdir, "*.crx"))), None)
if crxpath and os.path.getsize(crxpath) > 0:
contents = {}
with ZipFile(crxpath) as f:
for in_zip_f in f.infolist():
if not in_zip_f.filename.endswith(".js"):
continue
with f.open(in_zip_f) as f2:
content = f2.read().decode(errors="surrogateescape")
contents[in_zip_f.filename] = jsbeautifier.beautify(
content)
jslocs = sum(
[len(content.splitlines()) for content in contents.values()])
name = get_name(overview_path)
downloads = get_downloads(overview_path)
for filename, content in contents.items():
for i, line in enumerate(content.splitlines()):
for gr in greps:
if re.search(gr, line):
print(" -b beautify JavaScript before matching")
print(" -a <DIR> archive directory")
print(" -p <PREFIX> three-letter-prefix")
print(" -t <THREADS> number of threads to use")
def process_crx(ext_id, date, crx, greps, beautify, out_f):
with ZipFile(crx) as z:
for zip_file_info in z.infolist():
if not zip_file_info.filename.endswith(".js"):
continue
with z.open(zip_file_info) as f:
content = f.read().decode(errors="surrogateescape")
if beautify:
content = jsbeautifier.beautify(content)
for i, line in enumerate(content.splitlines()):
for gr in greps:
if not re.search(gr, line):
continue
args = [
ext_id, date,
name.replace("|", "<PIPE>") if name is not None else "",
str(downloads) if downloads is not None else "",
str(jslocs), gr,
filename + " (line " + str(i + 1) + ")",
line
ext_id, date, gr, zip_file_info.filename +
" (line " + str(i + 1) + ")", line
]
print("|".join(args), file=out_f)
def process_id(archivedir, outdir, greps, verbose, ext_id):
def process_id(archivedir, outdir, greps, beautify, verbose, ext_id):
txt = ""
txt = logmsg(verbose, txt, "Processing {} ...\n".format(ext_id))
@ -108,22 +72,19 @@ def process_id(archivedir, outdir, greps, verbose, ext_id):
if os.path.exists(greppath):
os.remove(greppath)
with open(greppath, "w") as out_f:
with tempfile.TemporaryDirectory() as tmpdir:
with tarfile.open(tarpath) as t:
t.extractall(tmpdir)
iddir = os.path.join(tmpdir, ext_id)
for date in sorted(os.listdir(iddir)):
try:
process_date(
os.path.join(iddir, date), ext_id, date, greps,
out_f)
except Exception:
txt = logmsg(
verbose, txt,
"Exception when handling {} on {}:\n".format(
ext_id, date))
txt = logmsg(verbose, txt, traceback.format_exc())
with tarfile.open(tarpath, 'r') as t:
for tar_info in t.getmembers():
if not tar_info.name.endswith(".crx") or tar_info.size is 0:
continue
date = tar_info.name.split("/")[1]
try:
with t.extractfile(tar_info) as crx:
process_crx(ext_id, date, crx, greps, beautify, out_f)
except Exception:
txt = logmsg(
verbose, txt,
"Exception when handling {}:\n".format(tar_info.name))
txt = logmsg(verbose, txt, traceback.format_exc())
return txt
@ -131,10 +92,11 @@ def process_id(archivedir, outdir, greps, verbose, ext_id):
def main(argv):
archive = "archive"
prefix = ""
beautify = False
parallel = 8
try:
opts, args = getopt.getopt(argv, "ha:p:t:",
["archive=", "prefix=", "threads="])
opts, args = getopt.getopt(
argv, "ha:p:bt:", ["archive=", "prefix=", "beautify", "threads="])
except getopt.GetoptError:
help()
sys.exit(2)
@ -146,6 +108,8 @@ def main(argv):
archive = arg
elif opt in ("-p", "--prefix"):
prefix = arg
elif opt in ("-b", "--beautify"):
beautify = True
elif opt in ("-t", "--threads"):
parallel = int(arg)
@ -158,15 +122,24 @@ def main(argv):
archivedir = os.path.join(archive, "data")
threeletterdirs = glob.glob(os.path.join(archivedir, prefix + "*"))
print("Using archive '{}'".format(archive))
print("Using prefix '{}'".format(prefix))
print("Using beautifier? '{}'".format(beautify))
print("Using {} threads".format(parallel))
print("Found {} three-letter-dirs".format(len(threeletterdirs)))
for threeletterdir in threeletterdirs:
ext_ids = list(set([d[:32] for d in os.listdir(threeletterdir)]))
with Pool(parallel) as p:
for txt in p.imap(
partial(process_id, archivedir, outdir, greps, True),
ext_ids):
partial(process_id, archivedir, outdir, greps, beautify,
True), ext_ids):
sys.stdout.write(txt)
sys.stdout.flush()
print("Sucessfully finished grepping")
if __name__ == "__main__":
main(sys.argv[1:])

Loading…
Cancel
Save