Last active
September 6, 2017 05:53
-
-
Save DosAmp/6484554 to your computer and use it in GitHub Desktop.
Parse output of `pm list packages -f` on Android
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import fileinput | |
import re | |
import http.client | |
import html.parser | |
URI_PREFIX = '/store/apps/details?id=' | |
REGEXP_JAVA_PACKAGE = r"(?P<pkg>([a-z_][a-z0-9_]*(\.[a-z_][a-z0-9_]*)*))$" | |
REGEXP_PACKAGE_LIST = r"package:(?P<path>/.*\.apk)=" + REGEXP_JAVA_PACKAGE | |
#REQUEST_HEADERS = {'Accept-Language': 'en'} | |
empty_none = lambda s: s or '' | |
csv = lambda i: ';'.join(i) | |
class PlayStoreAppNameHTMLParser(html.parser.HTMLParser): | |
def __init__(self): | |
super().__init__() | |
# __app_name_found replaced by truth check on app_name | |
self.__inner_div_follows = False | |
self.__app_name_follows = False | |
self.app_name = None | |
def handle_starttag(self, tag, attrs): | |
if tag != 'div' or self.app_name: | |
return | |
if self.__inner_div_follows: | |
self.__app_name_follows = True | |
self.__inner_div_follows = False | |
elif ('itemprop', 'name') in attrs: | |
self.__inner_div_follows = True | |
def handle_data(self, data): | |
if self.__app_name_follows: | |
self.app_name = data.strip() | |
self.__app_name_follows = False | |
def app_type(path): | |
if path.startswith('/system/app'): | |
return 'System' | |
elif path.startswith('/data/app'): | |
return 'Store' | |
elif path.startswith('/mnt/asec'): | |
return 'Protected' | |
else: | |
return 'Other' | |
def app_name(conn, package): | |
conn.request('GET', URI_PREFIX + package) | |
response = conn.getresponse() | |
# read all the data before next request, even if we don't have to parse it | |
data = response.read() | |
if (response.status != http.client.OK): return | |
# TODO just decoding the first ~128K seems enough, use errors='ignore' then | |
parser = PlayStoreAppNameHTMLParser() | |
parser.feed(data.decode('UTF-8')) | |
return parser.app_name | |
playstore = http.client.HTTPSConnection('play.google.com') | |
print(csv(['Name', 'Type', 'Package name'])) | |
for line in fileinput.input(): | |
linematch = re.match(REGEXP_PACKAGE_LIST, line) | |
if (not linematch): continue | |
atype = app_type(linematch.group('path')) | |
aname = app_name(playstore, linematch.group('pkg')) | |
print(csv([empty_none(aname), atype, linematch.group('pkg')]), flush=True) | |
playstore.close() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
print(csv([empty_none(aname), atype, linematch.group('pkg')]), flush=True)
^
SyntaxError: invalid syntax