Python 3: Add support in webkitpy.common.net.bugzilla
https://bugs.webkit.org/show_bug.cgi?id=202465
Reviewed by Dewei Zhu.
* Scripts/test-webkitpy-python3: Add webkitpy.common.net.bugzilla to the test list.
* Scripts/webkitpy/common/net/bugzilla/__init__.py: Use full import paths.
* Scripts/webkitpy/common/net/bugzilla/attachment_unittest.py: Ditto.
* Scripts/webkitpy/common/net/bugzilla/bug.py: Ditto.
(Bug.commit_revision): Dictionary sorting is no longer well defined in Python 3.
* Scripts/webkitpy/common/net/bugzilla/bug_unittest.py: Use full import paths.
* Scripts/webkitpy/common/net/bugzilla/bugzilla.py: Ditto.
(EditUsersParser.login_userid_pairs_from_edit_user_results): Return list, not an iterator.
(EditUsersParser.user_dict_from_edit_user_page): HTML_ENTITIES is logically connected to BeautifulSoup, not
BeautifulStoneSoup.
(BugzillaQueries._parse_attachment_ids_request_query): Explicitly cast date_tag to unicode.
(Bugzilla._parse_attachment_element): Legacy BeautifulSoup and bs4 Elements have different function calls.
(Bugzilla._file_object_for_upload): Use BytesIO instead of StringIO.
(Bugzilla.create_bug):
(reopen_bug): Access map as list, not iterator.
* Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py: Use full import paths.
* Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py: Ditto.
* Scripts/webkitpy/common/net/bugzilla/test_expectation_updater.py:
(TestExpectationUpdater._update_from_generic_attachment): Use items over iteritems().
(TestExpectationUpdater._update_from_platform_specific_attachment): Ditto.
(TestExpectationUpdater.do_update): Ditto.
* Scripts/webkitpy/common/net/buildbot/buildbot.py: Use Python 2/3 compatible urllib.
* Scripts/webkitpy/common/net/credentials.py: Use import statement to force autoinstall logic.
* Scripts/webkitpy/common/net/file_uploader.py: Use Python 2/3 compatible urllib.
(FileUploader._upload_data.callback):
(FileUploader):
* Scripts/webkitpy/common/net/resultsjsonparser.py:
(JSONTestResult._tokenize): Force map() call to return a list.
* Scripts/webkitpy/common/net/web_mock.py: Use Python 2/3 compatible urllib.
* Scripts/webkitpy/common/system/autoinstall.py:
(AutoInstaller.install): Return True if install was required, otherwise, return False.
* Scripts/webkitpy/common/system/filesystem_mock.py:
(MockFileSystem.__init__): Encode strings before writing them to the filesystem.
* Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py: Use Python 2/3 compatible urllib.
* Scripts/webkitpy/layout_tests/models/test_expectations.py:
(TestExpectationParser): Use items over iteritems().
* Scripts/webkitpy/layout_tests/models/test_failures.py: Use Python 3 compatible Pickle library.
* Scripts/webkitpy/layout_tests/models/test_results.py: Ditto.
* Scripts/webkitpy/port/darwin_testcase.py:
(DarwinTest.test_crashlog_path): Use dictionary for files.
* Scripts/webkitpy/test/main.py:
(Tester._run_tests): Update directories excluded from testing.
* Scripts/webkitpy/thirdparty/BeautifulSoup.py: Replaced.
(BeautifulSoup):Wrapper class making bs4 BeautifulSoup compatible with legacy BeautifulSoup calls.
(BeautifulStoneSoup): Ditto.
* Scripts/webkitpy/thirdparty/BeautifulSoup_legacy.py: Copied from Tools/Scripts/webkitpy/thirdparty/BeautifulSoup.py.
* Scripts/webkitpy/thirdparty/__init__.py:
(AutoinstallImportHook.__init__): Accept executive.
(AutoinstallImportHook.find_module): Add beautifulsoup.
(AutoinstallImportHook._install_beautifulsoup): bs4 must be converted via 2to3 before use in Python 3.
(AutoinstallImportHook._install): Return result from installer.
* Scripts/webkitpy/thirdparty/mock.py:
(Mock.reset_mock): Use values() instead of itervalues().
git-svn-id: http://svn.webkit.org/repository/webkit/trunk@251112 268f45cc-cd09-0410-ab3c-d52691b4dbfc
diff --git a/Tools/ChangeLog b/Tools/ChangeLog
index 9db76ae..963217c 100644
--- a/Tools/ChangeLog
+++ b/Tools/ChangeLog
@@ -1,3 +1,64 @@
+2019-10-14 Jonathan Bedard <jbedard@apple.com>
+
+ Python 3: Add support in webkitpy.common.net.bugzilla
+ https://bugs.webkit.org/show_bug.cgi?id=202465
+
+ Reviewed by Dewei Zhu.
+
+ * Scripts/test-webkitpy-python3: Add webkitpy.common.net.bugzilla to the test list.
+ * Scripts/webkitpy/common/net/bugzilla/__init__.py: Use full import paths.
+ * Scripts/webkitpy/common/net/bugzilla/attachment_unittest.py: Ditto.
+ * Scripts/webkitpy/common/net/bugzilla/bug.py: Ditto.
+ (Bug.commit_revision): Dictionary sorting is no longer well defined in Python 3.
+ * Scripts/webkitpy/common/net/bugzilla/bug_unittest.py: Use full import paths.
+ * Scripts/webkitpy/common/net/bugzilla/bugzilla.py: Ditto.
+ (EditUsersParser.login_userid_pairs_from_edit_user_results): Return list, not an iterator.
+ (EditUsersParser.user_dict_from_edit_user_page): HTML_ENTITIES is logically connected to BeautifulSoup, not
+ BeautifulStoneSoup.
+ (BugzillaQueries._parse_attachment_ids_request_query): Explicitly cast date_tag to unicode.
+ (Bugzilla._parse_attachment_element): Legacy BeautifulSoup and bs4 Elements have different function calls.
+ (Bugzilla._file_object_for_upload): Use BytesIO instead of StringIO.
+ (Bugzilla.create_bug):
+ (reopen_bug): Access map as list, not iterator.
+ * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py: Use full import paths.
+ * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py: Ditto.
+ * Scripts/webkitpy/common/net/bugzilla/test_expectation_updater.py:
+ (TestExpectationUpdater._update_from_generic_attachment): Use items over iteritems().
+ (TestExpectationUpdater._update_from_platform_specific_attachment): Ditto.
+ (TestExpectationUpdater.do_update): Ditto.
+ * Scripts/webkitpy/common/net/buildbot/buildbot.py: Use Python 2/3 compatible urllib.
+ * Scripts/webkitpy/common/net/credentials.py: Use import statement to force autoinstall logic.
+ * Scripts/webkitpy/common/net/file_uploader.py: Use Python 2/3 compatible urllib.
+ (FileUploader._upload_data.callback):
+ (FileUploader):
+ * Scripts/webkitpy/common/net/resultsjsonparser.py:
+ (JSONTestResult._tokenize): Force map() call to return a list.
+ * Scripts/webkitpy/common/net/web_mock.py: Use Python 2/3 compatible urllib.
+ * Scripts/webkitpy/common/system/autoinstall.py:
+ (AutoInstaller.install): Return True if install was required, otherwise, return False.
+ * Scripts/webkitpy/common/system/filesystem_mock.py:
+ (MockFileSystem.__init__): Encode strings before writing them to the filesystem.
+ * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py: Use Python 2/3 compatible urllib.
+ * Scripts/webkitpy/layout_tests/models/test_expectations.py:
+ (TestExpectationParser): Use items over iteritems().
+ * Scripts/webkitpy/layout_tests/models/test_failures.py: Use Python 3 compatible Pickle library.
+ * Scripts/webkitpy/layout_tests/models/test_results.py: Ditto.
+ * Scripts/webkitpy/port/darwin_testcase.py:
+ (DarwinTest.test_crashlog_path): Use dictionary for files.
+ * Scripts/webkitpy/test/main.py:
+ (Tester._run_tests): Update directories excluded from testing.
+ * Scripts/webkitpy/thirdparty/BeautifulSoup.py: Replaced.
+ (BeautifulSoup):Wrapper class making bs4 BeautifulSoup compatible with legacy BeautifulSoup calls.
+ (BeautifulStoneSoup): Ditto.
+ * Scripts/webkitpy/thirdparty/BeautifulSoup_legacy.py: Copied from Tools/Scripts/webkitpy/thirdparty/BeautifulSoup.py.
+ * Scripts/webkitpy/thirdparty/__init__.py:
+ (AutoinstallImportHook.__init__): Accept executive.
+ (AutoinstallImportHook.find_module): Add beautifulsoup.
+ (AutoinstallImportHook._install_beautifulsoup): bs4 must be converted via 2to3 before use in Python 3.
+ (AutoinstallImportHook._install): Return result from installer.
+ * Scripts/webkitpy/thirdparty/mock.py:
+ (Mock.reset_mock): Use values() instead of itervalues().
+
2019-10-14 Matt Mokary <mmokary@apple.com>
FindController::findString always updates foundStringMatchIndex even if match is the same as before
diff --git a/Tools/Scripts/test-webkitpy-python3 b/Tools/Scripts/test-webkitpy-python3
index ff497fa..eda3a82 100755
--- a/Tools/Scripts/test-webkitpy-python3
+++ b/Tools/Scripts/test-webkitpy-python3
@@ -34,6 +34,7 @@
PYTHON3_COMPATIBLE_DIRECTORIES = [
'webkitpy.common.system',
'webkitpy.common.thread',
+ 'webkitpy.common.net.bugzilla',
]
diff --git a/Tools/Scripts/webkitpy/common/net/bugzilla/__init__.py b/Tools/Scripts/webkitpy/common/net/bugzilla/__init__.py
index c427b18..d28b712 100644
--- a/Tools/Scripts/webkitpy/common/net/bugzilla/__init__.py
+++ b/Tools/Scripts/webkitpy/common/net/bugzilla/__init__.py
@@ -1,7 +1,7 @@
# Required for Python to search this directory for module files
# We only export public API here.
-from .bugzilla import Bugzilla
+from webkitpy.common.net.bugzilla.bugzilla import Bugzilla
# Unclear if Bug and Attachment need to be public classes.
-from .bug import Bug
-from .attachment import Attachment
+from webkitpy.common.net.bugzilla.bug import Bug
+from webkitpy.common.net.bugzilla.attachment import Attachment
diff --git a/Tools/Scripts/webkitpy/common/net/bugzilla/attachment_unittest.py b/Tools/Scripts/webkitpy/common/net/bugzilla/attachment_unittest.py
index f259cfe..8be51f2 100644
--- a/Tools/Scripts/webkitpy/common/net/bugzilla/attachment_unittest.py
+++ b/Tools/Scripts/webkitpy/common/net/bugzilla/attachment_unittest.py
@@ -25,7 +25,7 @@
from datetime import datetime
from webkitpy.common.net.bugzilla.constants import BUGZILLA_DATE_FORMAT
-from .attachment import Attachment
+from webkitpy.common.net.bugzilla.attachment import Attachment
class AttachmentTest(unittest.TestCase):
diff --git a/Tools/Scripts/webkitpy/common/net/bugzilla/bug.py b/Tools/Scripts/webkitpy/common/net/bugzilla/bug.py
index 2cbcba4..d8e8cc1 100644
--- a/Tools/Scripts/webkitpy/common/net/bugzilla/bug.py
+++ b/Tools/Scripts/webkitpy/common/net/bugzilla/bug.py
@@ -30,7 +30,7 @@
import re
-from .attachment import Attachment
+from webkitpy.common.net.bugzilla.attachment import Attachment
class Bug(object):
@@ -133,7 +133,7 @@
def commit_revision(self):
# Sort the comments in reverse order as we want the latest committed revision.
r = re.compile("Committed r(?P<svn_revision>\d+)")
- for comment in sorted(self.comments(), reverse=True):
+ for comment in sorted(self.comments(), key=lambda comment: comment['text'], reverse=True):
rev = r.search(comment['text'])
if rev:
return int(rev.group('svn_revision'))
diff --git a/Tools/Scripts/webkitpy/common/net/bugzilla/bug_unittest.py b/Tools/Scripts/webkitpy/common/net/bugzilla/bug_unittest.py
index 28a1d9a..9358bc0 100644
--- a/Tools/Scripts/webkitpy/common/net/bugzilla/bug_unittest.py
+++ b/Tools/Scripts/webkitpy/common/net/bugzilla/bug_unittest.py
@@ -28,7 +28,7 @@
import unittest
-from .bug import Bug
+from webkitpy.common.net.bugzilla.bug import Bug
class BugTest(unittest.TestCase):
diff --git a/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla.py b/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla.py
index 85e1ed3..9237ac5 100644
--- a/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla.py
+++ b/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla.py
@@ -34,14 +34,15 @@
import logging
import mimetypes
import re
-import StringIO
import socket
+import sys
import urllib
from datetime import datetime # used in timestamp()
-from .attachment import Attachment
-from .bug import Bug
+from webkitpy.common.unicode_compatibility import BytesIO, StringIO, unicode
+from webkitpy.common.net.bugzilla.attachment import Attachment
+from webkitpy.common.net.bugzilla.bug import Bug
from webkitpy.common.config import committers
import webkitpy.common.config.urls as config_urls
@@ -73,11 +74,11 @@
return (login, user_id)
def login_userid_pairs_from_edit_user_results(self, results_page):
- soup = BeautifulSoup(results_page, convertEntities=BeautifulStoneSoup.HTML_ENTITIES)
+ soup = BeautifulSoup(results_page, convertEntities=BeautifulSoup.HTML_ENTITIES)
results_table = soup.find(id="admin_table")
login_userid_pairs = [self._login_and_uid_from_row(row) for row in results_table('tr')]
# Filter out None from the logins.
- return filter(lambda pair: bool(pair), login_userid_pairs)
+ return list(filter(lambda pair: bool(pair), login_userid_pairs))
def _group_name_and_string_from_row(self, row):
label_element = row.find('label')
@@ -86,7 +87,7 @@
return (group_name, group_string)
def user_dict_from_edit_user_page(self, page):
- soup = BeautifulSoup(page, convertEntities=BeautifulStoneSoup.HTML_ENTITIES)
+ soup = BeautifulSoup(page, convertEntities=BeautifulSoup.HTML_ENTITIES)
user_table = soup.find("table", {'class': 'main'})
user_dict = {}
for row in user_table('tr'):
@@ -209,7 +210,7 @@
continue
patch_id = int(digits.search(patch_tag["href"]).group(0))
date_tag = row.find("td", text=date_format)
- if date_tag and datetime.strptime(date_format.search(date_tag).group(0), "%Y-%m-%d %H:%M") < since:
+ if date_tag and datetime.strptime(date_format.search(unicode(date_tag)).group(0), "%Y-%m-%d %H:%M") < since:
continue
patch_ids.append(patch_id)
return patch_ids
@@ -418,8 +419,12 @@
def _parse_attachment_element(self, element, bug_id):
attachment = {}
attachment['bug_id'] = bug_id
- attachment['is_obsolete'] = (element.has_key('isobsolete') and element['isobsolete'] == "1")
- attachment['is_patch'] = (element.has_key('ispatch') and element['ispatch'] == "1")
+ if sys.version_info > (3, 0):
+ attachment['is_obsolete'] = (element.has_attr('isobsolete') and element['isobsolete'] == "1")
+ attachment['is_patch'] = (element.has_attr('ispatch') and element['ispatch'] == "1")
+ else:
+ attachment['is_obsolete'] = (element.has_key('isobsolete') and element['isobsolete'] == "1")
+ attachment['is_patch'] = (element.has_key('ispatch') and element['ispatch'] == "1")
attachment['id'] = int(element.find('attachid').string)
# FIXME: No need to parse out the url here.
attachment['url'] = self.attachment_url_for_id(attachment['id'])
@@ -637,7 +642,7 @@
# Only if file_or_string is not already encoded do we want to encode it.
if isinstance(file_or_string, unicode):
file_or_string = file_or_string.encode('utf-8')
- return StringIO.StringIO(file_or_string)
+ return BytesIO(file_or_string)
# timestamp argument is just for unittests.
def _filename_for_upload(self, file_object, bug_id, extension="txt", timestamp=timestamp):
@@ -759,7 +764,7 @@
# _fill_attachment_form expects a file-like object
# Patch files are already binary, so no encoding needed.
assert(isinstance(diff, str))
- patch_file_object = StringIO.StringIO(diff)
+ patch_file_object = StringIO(diff)
commit_flag = CommitQueueFlag.mark_for_nothing
if mark_for_commit_queue:
commit_flag = CommitQueueFlag.mark_for_commit_queue
@@ -923,7 +928,7 @@
# This is a hack around the fact that ClientForm.ListControl seems to
# have no simpler way to ask if a control has an item named "REOPENED"
# without using exceptions for control flow.
- possible_bug_statuses = map(lambda item: item.name, bug_status.items)
+ possible_bug_statuses = list(map(lambda item: item.name, bug_status.items))
if "REOPENED" in possible_bug_statuses:
bug_status.value = ["REOPENED"]
# If the bug was never confirmed it will not have a "REOPENED"
diff --git a/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py b/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py
index 3ee4d39..d67a86f 100644
--- a/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py
+++ b/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py
@@ -30,9 +30,9 @@
import datetime
import logging
-from .bug import Bug
-from .bugzilla import Bugzilla
-from .attachment import Attachment
+from webkitpy.common.net.bugzilla.bug import Bug
+from webkitpy.common.net.bugzilla.bugzilla import Bugzilla
+from webkitpy.common.net.bugzilla.attachment import Attachment
from webkitpy.common.config.committers import CommitterList, Reviewer
_log = logging.getLogger(__name__)
diff --git a/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py b/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py
index b113367..8033cbc 100644
--- a/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py
+++ b/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py
@@ -26,18 +26,18 @@
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-import StringIO
import datetime
+import sys
import unittest
-from .bugzilla import Bugzilla, BugzillaQueries, CommitQueueFlag, EditUsersParser
-
from webkitpy.common.config import urls
from webkitpy.common.config.committers import Reviewer, Committer, Contributor, CommitterList
-from webkitpy.common.system.outputcapture import OutputCapture
from webkitpy.common.net.web_mock import MockBrowser
-from webkitpy.thirdparty.mock import Mock
+from webkitpy.common.net.bugzilla.bugzilla import Bugzilla, BugzillaQueries, CommitQueueFlag, EditUsersParser
+from webkitpy.common.system.outputcapture import OutputCapture
+from webkitpy.common.unicode_compatibility import StringIO
from webkitpy.thirdparty.BeautifulSoup import BeautifulSoup
+from webkitpy.thirdparty.mock import Mock
class BugzillaTest(unittest.TestCase):
@@ -67,7 +67,7 @@
</attachment>
'''
_expected_example_attachment_parsing = {
- 'attach_date': datetime.datetime(2009, 07, 29, 10, 23),
+ 'attach_date': datetime.datetime(2009, 7, 29, 10, 23),
'bug_id': 100,
'is_obsolete': True,
'is_patch': True,
@@ -212,7 +212,7 @@
# FIXME: This should move to a central location and be shared by more unit tests.
def _assert_dictionaries_equal(self, actual, expected):
# Make sure we aren't parsing more or less than we expect
- self.assertItemsEqual(actual.keys(), expected.keys())
+ self.assertEqual(sorted(actual.keys()), sorted(expected.keys()))
for key, expected_value in expected.items():
self.assertEqual(actual[key], expected_value, ("Failure for key: %s: Actual='%s' Expected='%s'" % (key, actual[key], expected_value)))
@@ -303,7 +303,7 @@
def test_file_object_for_upload(self):
bugzilla = Bugzilla()
- file_object = StringIO.StringIO()
+ file_object = StringIO()
unicode_tor = u"WebKit \u2661 Tor Arne Vestb\u00F8!"
utf8_tor = unicode_tor.encode("utf-8")
self.assertEqual(bugzilla._file_object_for_upload(file_object), file_object)
@@ -316,7 +316,7 @@
mock_file.name = "foo"
self.assertEqual(bugzilla._filename_for_upload(mock_file, 1234), 'foo')
mock_timestamp = lambda: "now"
- filename = bugzilla._filename_for_upload(StringIO.StringIO(), 1234, extension="patch", timestamp=mock_timestamp)
+ filename = bugzilla._filename_for_upload(StringIO(), 1234, extension="patch", timestamp=mock_timestamp)
self.assertEqual(filename, "bug-1234-now.patch")
def test_commit_queue_flag(self):
diff --git a/Tools/Scripts/webkitpy/common/net/bugzilla/test_expectation_updater.py b/Tools/Scripts/webkitpy/common/net/bugzilla/test_expectation_updater.py
index 3bc55bf..a39a157 100644
--- a/Tools/Scripts/webkitpy/common/net/bugzilla/test_expectation_updater.py
+++ b/Tools/Scripts/webkitpy/common/net/bugzilla/test_expectation_updater.py
@@ -113,7 +113,7 @@
# FIXME: Investigate the possibility to align what is done there with what single_test_runner is doing.
# In particular the fact of not overwriting the file if content is the same.
def _update_from_generic_attachment(self):
- for test_name, expected_content in self._tests_to_update(self.generic_attachment).iteritems():
+ for test_name, expected_content in self._tests_to_update(self.generic_attachment).items():
expected_filename = self.filesystem.join(self.layout_test_repository, TestResultWriter.expected_filename(test_name, self.filesystem))
if expected_content != self._file_content_if_exists(expected_filename):
_log.info("Updating " + test_name + " (" + expected_filename + ")")
@@ -122,7 +122,7 @@
# FIXME: Investigate the possibility to align what is done there with what single_test_runner is doing.
# In particular the ability to remove a new specific expectation if it is the same as the generic one.
def _update_from_platform_specific_attachment(self, attachment, bot_type):
- for test_name, expected_content in self._tests_to_update(attachment, bot_type).iteritems():
+ for test_name, expected_content in self._tests_to_update(attachment, bot_type).items():
expected_filename = self.filesystem.join(self.layout_test_repository, TestResultWriter.expected_filename(test_name, self.filesystem, bot_type))
generic_expected_filename = self.filesystem.join(self.layout_test_repository, TestResultWriter.expected_filename(test_name, self.filesystem))
if expected_content != self._file_content_if_exists(generic_expected_filename):
@@ -138,7 +138,7 @@
return
if self.generic_attachment:
self._update_from_generic_attachment()
- for bot_type, attachment in self.platform_specific_attachments.iteritems():
+ for bot_type, attachment in self.platform_specific_attachments.items():
self._update_from_platform_specific_attachment(attachment, bot_type)
diff --git a/Tools/Scripts/webkitpy/common/net/buildbot/buildbot.py b/Tools/Scripts/webkitpy/common/net/buildbot/buildbot.py
index 614d3e4..3cf3da1 100644
--- a/Tools/Scripts/webkitpy/common/net/buildbot/buildbot.py
+++ b/Tools/Scripts/webkitpy/common/net/buildbot/buildbot.py
@@ -29,8 +29,8 @@
import json
import operator
import re
+import sys
import urllib
-import urllib2
import webkitpy.common.config.urls as config_urls
from webkitpy.common.memoized import memoized
@@ -39,8 +39,16 @@
from webkitpy.common.net.networktransaction import NetworkTransaction
from webkitpy.common.net.regressionwindow import RegressionWindow
from webkitpy.common.system.logutils import get_logger
-from webkitpy.thirdparty.BeautifulSoup import BeautifulSoup
from webkitpy.thirdparty.autoinstalled.mechanize import Browser
+from webkitpy.thirdparty.BeautifulSoup import BeautifulSoup
+
+if sys.version_info > (3, 0):
+ from urllib.error import HTTPError, URLError
+ from urllib.parse import quote
+ from urllib.request import urlopen
+else:
+ from urllib2 import HTTPError, quote, URLError, urlopen
+
_log = get_logger(__file__)
@@ -74,7 +82,7 @@
def _fetch_file_from_results(self, results_url, file_name):
# It seems this can return None if the url redirects and then returns 404.
- result = urllib2.urlopen("%s/%s" % (results_url, file_name))
+ result = urlopen("%s/%s" % (results_url, file_name))
if not result:
return None
# urlopen returns a file-like object which sometimes works fine with str()
@@ -161,7 +169,7 @@
print("Loading revision/build list from %s." % self.results_url())
print("This may take a while...")
result_files = self._buildbot._fetch_twisted_directory_listing(self.results_url())
- except urllib2.HTTPError as error:
+ except HTTPError as error:
if error.code != 404:
raise
_log.debug("Revision/build list failed to load.")
@@ -363,8 +371,8 @@
# find a way to reduce the response size further.
json_url = "%s/json/builders/%s/builds/%s?filter=1" % (self.buildbot_url, urllib.quote(builder.name()), build_number)
try:
- return json.load(urllib2.urlopen(json_url))
- except urllib2.URLError as err:
+ return json.load(urlopen(json_url))
+ except URLError as err:
build_url = Build.build_url(builder, build_number)
_log.error("Error fetching data for %s build %s (%s, json: %s): %s" % (builder.name(), build_number, build_url, json_url, err))
return None
@@ -375,7 +383,7 @@
def _fetch_one_box_per_builder(self):
build_status_url = "%s/one_box_per_builder" % self.buildbot_url
- return urllib2.urlopen(build_status_url)
+ return urlopen(build_status_url)
def _file_cell_text(self, file_cell):
"""Traverses down through firstChild elements until one containing a string is found, then returns that string"""
@@ -402,7 +410,7 @@
# FIXME: There should be a better way to get this information directly from twisted.
def _fetch_twisted_directory_listing(self, url):
- return self._parse_twisted_directory_listing(urllib2.urlopen(url))
+ return self._parse_twisted_directory_listing(urlopen(url))
def builders(self):
return [self.builder_with_name(status["name"]) for status in self.builder_statuses()]
@@ -444,8 +452,8 @@
build = build.previous_build()
def _fetch_builder_page(self, builder):
- builder_page_url = "%s/builders/%s?numbuilds=100" % (self.buildbot_url, urllib2.quote(builder.name()))
- return urllib2.urlopen(builder_page_url)
+ builder_page_url = "%s/builders/%s?numbuilds=100" % (self.buildbot_url, quote(builder.name()))
+ return urlopen(builder_page_url)
def _revisions_for_builder(self, builder):
soup = BeautifulSoup(self._fetch_builder_page(builder))
diff --git a/Tools/Scripts/webkitpy/common/net/credentials.py b/Tools/Scripts/webkitpy/common/net/credentials.py
index d981478..cdeb4a8 100644
--- a/Tools/Scripts/webkitpy/common/net/credentials.py
+++ b/Tools/Scripts/webkitpy/common/net/credentials.py
@@ -37,7 +37,7 @@
from webkitpy.common.checkout.scm import Git
from webkitpy.common.system.executive import Executive, ScriptError
from webkitpy.common.system.user import User
-from webkitpy.thirdparty.autoinstalled import keyring
+import webkitpy.thirdparty.autoinstalled.keyring as keyring
_log = logging.getLogger(__name__)
diff --git a/Tools/Scripts/webkitpy/common/net/file_uploader.py b/Tools/Scripts/webkitpy/common/net/file_uploader.py
index 871295b..886653f 100644
--- a/Tools/Scripts/webkitpy/common/net/file_uploader.py
+++ b/Tools/Scripts/webkitpy/common/net/file_uploader.py
@@ -28,10 +28,14 @@
import mimetypes
import time
-import urllib2
+import sys
from webkitpy.common.net.networktransaction import NetworkTransaction, NetworkTimeout
+if sys.version_info > (3, 0):
+ from urllib.request import Request, urlopen
+else:
+ from urllib2 import Request, urlopen
def get_mime_type(filename):
return mimetypes.guess_type(filename)[0] or 'application/octet-stream'
@@ -101,7 +105,7 @@
# FIXME: Setting a timeout, either globally using socket.setdefaulttimeout()
# or in urlopen(), doesn't appear to work on Mac 10.5 with Python 2.7.
# For now we will ignore the timeout value and hope for the best.
- request = urllib2.Request(self._url, data, {"Content-Type": content_type})
- return urllib2.urlopen(request)
+ request = Request(self._url, data, {"Content-Type": content_type})
+ return urlopen(request)
return NetworkTransaction(timeout_seconds=self._timeout_seconds).run(callback)
diff --git a/Tools/Scripts/webkitpy/common/net/resultsjsonparser.py b/Tools/Scripts/webkitpy/common/net/resultsjsonparser.py
index 32b22c9..3980357 100644
--- a/Tools/Scripts/webkitpy/common/net/resultsjsonparser.py
+++ b/Tools/Scripts/webkitpy/common/net/resultsjsonparser.py
@@ -82,7 +82,7 @@
return True
def _tokenize(self, results_string):
- tokens = map(TestExpectations.expectation_from_string, results_string.split(' '))
+ tokens = list(map(TestExpectations.expectation_from_string, results_string.split(' ')))
if None in tokens:
_log.warning("Unrecognized result in %s" % results_string)
return set(tokens)
diff --git a/Tools/Scripts/webkitpy/common/net/web_mock.py b/Tools/Scripts/webkitpy/common/net/web_mock.py
index e8b8405..0c7d5c5 100644
--- a/Tools/Scripts/webkitpy/common/net/web_mock.py
+++ b/Tools/Scripts/webkitpy/common/net/web_mock.py
@@ -27,8 +27,13 @@
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-import StringIO
-import urllib2
+from webkitpy.common.unicode_compatibility import StringIO
+
+import sys
+if sys.version_info > (3, 0):
+ from urllib.error import HTTPError
+else:
+ from urllib2 import HTTPError
class MockWeb(object):
def __init__(self, urls=None, responses=[]):
@@ -53,7 +58,7 @@
self.body = values.get('body', '')
if int(self.status_code) >= 400:
- raise urllib2.HTTPError(
+ raise HTTPError(
url=self.url,
code=self.status_code,
msg='Received error status code: {}'.format(self.status_code),
@@ -83,7 +88,7 @@
return self.params.get(key)
def submit(self):
- return StringIO.StringIO()
+ return StringIO()
def set_handle_robots(self, value):
pass
diff --git a/Tools/Scripts/webkitpy/common/system/autoinstall.py b/Tools/Scripts/webkitpy/common/system/autoinstall.py
index 570933b..f2017a8 100644
--- a/Tools/Scripts/webkitpy/common/system/autoinstall.py
+++ b/Tools/Scripts/webkitpy/common/system/autoinstall.py
@@ -483,7 +483,7 @@
def install(self, url, should_refresh=False, target_name=None,
url_subpath=None, files_to_remove=None):
- """Install a python package from an URL.
+ """Install a python package from an URL, return False if already installed.
Args:
url: The URL from which to download the package.
@@ -513,7 +513,7 @@
target_path = os.path.join(self._target_dir, target_name.replace('/', os.sep))
if not should_refresh and self._is_downloaded(target_name, url) and os.path.exists(target_path):
- return
+ return False
files_to_remove = files_to_remove or []
package_name = target_name.replace(os.sep, '.')
@@ -542,3 +542,4 @@
shutil.rmtree(scratch_dir, ignore_errors=True)
_log.debug('Auto-installed %s to:' % url)
_log.debug(' "%s"' % target_path)
+ return True
diff --git a/Tools/Scripts/webkitpy/common/system/filesystem_mock.py b/Tools/Scripts/webkitpy/common/system/filesystem_mock.py
index 3a8c935..3ea39c6 100644
--- a/Tools/Scripts/webkitpy/common/system/filesystem_mock.py
+++ b/Tools/Scripts/webkitpy/common/system/filesystem_mock.py
@@ -48,7 +48,7 @@
value of None is used to indicate that the file should
not exist.
"""
- self.files = files or {}
+ self.files = {name: unicode_compatibility.encode_if_necessary(contents) for name, contents in (files or {}).items()}
self.written_files = {}
self.last_tmpdir = None
self.current_tmpno = 0
diff --git a/Tools/Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py b/Tools/Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py
index 3b5abbc..85aeed8 100644
--- a/Tools/Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py
+++ b/Tools/Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py
@@ -31,9 +31,15 @@
import subprocess
import sys
import time
-import urllib2
import xml.dom.minidom
+if sys.version_info > (3, 0):
+ from urllib.error import HTTPError, URLError
+ from urllib.parse import quote
+ from urllib.request import urlopen
+else:
+ from urllib2 import HTTPError, quote, URLError, urlopen
+
from webkitpy.common.checkout.scm.detection import SCMDetector
from webkitpy.common.net.file_uploader import FileUploader
@@ -465,23 +471,23 @@
return {}, None
results_file_url = (self.URL_FOR_TEST_LIST_JSON %
- (urllib2.quote(self._test_results_servers[server_index]),
- urllib2.quote(self._builder_name),
+ (quote(self._test_results_servers[server_index]),
+ quote(self._builder_name),
self.RESULTS_FILENAME,
- urllib2.quote(self._test_type),
- urllib2.quote(self._master_name)))
+ quote(self._test_type),
+ quote(self._master_name)))
try:
# FIXME: We should talk to the network via a Host object.
- results_file = urllib2.urlopen(results_file_url)
+ results_file = urlopen(results_file_url)
info = results_file.info()
old_results = results_file.read()
- except urllib2.HTTPError as http_error:
+ except HTTPError as http_error:
# A non-4xx status code means the bot is hosed for some reason
# and we can't grab the results.json file off of it.
if (http_error.code < 400 and http_error.code >= 500):
error = http_error
- except urllib2.URLError as url_error:
+ except URLError as url_error:
error = url_error
if old_results:
diff --git a/Tools/Scripts/webkitpy/layout_tests/models/test_expectations.py b/Tools/Scripts/webkitpy/layout_tests/models/test_expectations.py
index 69e3362..e7c5813 100644
--- a/Tools/Scripts/webkitpy/layout_tests/models/test_expectations.py
+++ b/Tools/Scripts/webkitpy/layout_tests/models/test_expectations.py
@@ -235,7 +235,7 @@
]
_configuration_tokens = dict((token, token.upper()) for token in _configuration_tokens_list)
- _inverted_configuration_tokens = dict((value, name) for name, value in _configuration_tokens.iteritems())
+ _inverted_configuration_tokens = dict((value, name) for name, value in _configuration_tokens.items())
# FIXME: Update the original modifiers list and remove this once the old syntax is gone.
_expectation_tokens = {
@@ -252,7 +252,7 @@
'WontFix': 'WONTFIX',
}
- _inverted_expectation_tokens = dict([(value, name) for name, value in _expectation_tokens.iteritems()] +
+ _inverted_expectation_tokens = dict([(value, name) for name, value in _expectation_tokens.items()] +
[('TEXT', 'Failure'), ('IMAGE+TEXT', 'Failure'), ('AUDIO', 'Failure')])
# FIXME: Seems like these should be classmethods on TestExpectationLine instead of TestExpectationParser.
diff --git a/Tools/Scripts/webkitpy/layout_tests/models/test_failures.py b/Tools/Scripts/webkitpy/layout_tests/models/test_failures.py
index d95da02..d0aa366 100644
--- a/Tools/Scripts/webkitpy/layout_tests/models/test_failures.py
+++ b/Tools/Scripts/webkitpy/layout_tests/models/test_failures.py
@@ -26,8 +26,14 @@
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-import cPickle
import logging
+import sys
+
+if sys.version_info > (3, 0):
+ import pickle
+else:
+ import cPickle as pickle
+
from webkitpy.layout_tests.models import test_expectations
@@ -91,7 +97,7 @@
@staticmethod
def loads(s):
"""Creates a TestFailure object from the specified string."""
- return cPickle.loads(s)
+ return pickle.loads(s)
def message(self):
"""Returns a string describing the failure in more detail."""
@@ -108,7 +114,7 @@
def dumps(self):
"""Returns the string/JSON representation of a TestFailure."""
- return cPickle.dumps(self)
+ return pickle.dumps(self)
def driver_needs_restart(self):
"""Returns True if we should kill DumpRenderTree/WebKitTestRunner before the next test."""
diff --git a/Tools/Scripts/webkitpy/layout_tests/models/test_results.py b/Tools/Scripts/webkitpy/layout_tests/models/test_results.py
index 947f707..770840f 100644
--- a/Tools/Scripts/webkitpy/layout_tests/models/test_results.py
+++ b/Tools/Scripts/webkitpy/layout_tests/models/test_results.py
@@ -26,7 +26,11 @@
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-import cPickle
+import sys
+if sys.version_info > (3, 0):
+ import pickle
+else:
+ import cPickle as pickle
from webkitpy.layout_tests.models import test_failures
@@ -36,7 +40,7 @@
@staticmethod
def loads(string):
- return cPickle.loads(string)
+ return pickle.loads(string)
def __init__(self, test_name, failures=None, test_run_time=None, has_stderr=False, reftest_type=None, pid=None, references=None):
self.test_name = test_name
@@ -89,4 +93,4 @@
return False
def dumps(self):
- return cPickle.dumps(self)
+ return pickle.dumps(self)
diff --git a/Tools/Scripts/webkitpy/port/darwin_testcase.py b/Tools/Scripts/webkitpy/port/darwin_testcase.py
index ef4386b..9b5caa4 100644
--- a/Tools/Scripts/webkitpy/port/darwin_testcase.py
+++ b/Tools/Scripts/webkitpy/port/darwin_testcase.py
@@ -94,7 +94,7 @@
port = self.make_port()
self.assertEqual(port.path_to_crash_logs(), '/Users/mock/Library/Logs/CrashReporter')
- host = MockSystemHost(filesystem=MockFileSystem(files=['/Users/mock/Library/Logs/DiagnosticReports/crashlog.crash']))
+ host = MockSystemHost(filesystem=MockFileSystem(files={'/Users/mock/Library/Logs/DiagnosticReports/crashlog.crash': None}))
port = self.make_port(host)
self.assertEqual(port.path_to_crash_logs(), '/Users/mock/Library/Logs/DiagnosticReports')
diff --git a/Tools/Scripts/webkitpy/test/main.py b/Tools/Scripts/webkitpy/test/main.py
index 5145c57..7162d33 100644
--- a/Tools/Scripts/webkitpy/test/main.py
+++ b/Tools/Scripts/webkitpy/test/main.py
@@ -221,7 +221,12 @@
self._options.child_processes = 1
import webkitpy.thirdparty.autoinstalled.coverage as coverage
- cov = coverage.coverage(omit=["/usr/*", "*/webkitpy/thirdparty/autoinstalled/*", "*/webkitpy/thirdparty/BeautifulSoup.py"])
+ cov = coverage.coverage(omit=[
+ "/usr/*",
+ "*/webkitpy/thirdparty/autoinstalled/*",
+ "*/webkitpy/thirdparty/BeautifulSoup.py",
+ "*/webkitpy/thirdparty/BeautifulSoup_legacy.py",
+ ])
cov.start()
self.printer.write_update("Checking imports ...")
diff --git a/Tools/Scripts/webkitpy/thirdparty/BeautifulSoup.py b/Tools/Scripts/webkitpy/thirdparty/BeautifulSoup.py
index 4b17b85..a48ba63 100644
--- a/Tools/Scripts/webkitpy/thirdparty/BeautifulSoup.py
+++ b/Tools/Scripts/webkitpy/thirdparty/BeautifulSoup.py
@@ -1,2014 +1,58 @@
-"""Beautiful Soup
-Elixir and Tonic
-"The Screen-Scraper's Friend"
-http://www.crummy.com/software/BeautifulSoup/
-
-Beautiful Soup parses a (possibly invalid) XML or HTML document into a
-tree representation. It provides methods and Pythonic idioms that make
-it easy to navigate, search, and modify the tree.
-
-A well-formed XML/HTML document yields a well-formed data
-structure. An ill-formed XML/HTML document yields a correspondingly
-ill-formed data structure. If your document is only locally
-well-formed, you can use this library to find and process the
-well-formed part of it.
-
-Beautiful Soup works with Python 2.2 and up. It has no external
-dependencies, but you'll have more success at converting data to UTF-8
-if you also install these three packages:
-
-* chardet, for auto-detecting character encodings
- http://chardet.feedparser.org/
-* cjkcodecs and iconv_codec, which add more encodings to the ones supported
- by stock Python.
- http://cjkpython.i18n.org/
-
-Beautiful Soup defines classes for two main parsing strategies:
-
- * BeautifulStoneSoup, for parsing XML, SGML, or your domain-specific
- language that kind of looks like XML.
-
- * BeautifulSoup, for parsing run-of-the-mill HTML code, be it valid
- or invalid. This class has web browser-like heuristics for
- obtaining a sensible parse tree in the face of common HTML errors.
-
-Beautiful Soup also defines a class (UnicodeDammit) for autodetecting
-the encoding of an HTML or XML document, and converting it to
-Unicode. Much of this code is taken from Mark Pilgrim's Universal Feed Parser.
-
-For more than you ever wanted to know about Beautiful Soup, see the
-documentation:
-http://www.crummy.com/software/BeautifulSoup/documentation.html
-
-Here, have some legalese:
-
-Copyright (c) 2004-2010, Leonard Richardson
-
-All rights reserved.
-
-Redistribution and use in source and binary forms, with or without
-modification, are permitted provided that the following conditions are
-met:
-
- * Redistributions of source code must retain the above copyright
- notice, this list of conditions and the following disclaimer.
-
- * Redistributions in binary form must reproduce the above
- copyright notice, this list of conditions and the following
- disclaimer in the documentation and/or other materials provided
- with the distribution.
-
- * Neither the name of the the Beautiful Soup Consortium and All
- Night Kosher Bakery nor the names of its contributors may be
- used to endorse or promote products derived from this software
- without specific prior written permission.
-
-THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
-A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
-CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
-EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
-LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
-NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
-SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE, DAMMIT.
-
-"""
-from __future__ import generators
-
-__author__ = "Leonard Richardson (leonardr@segfault.org)"
-__version__ = "3.2.0"
-__copyright__ = "Copyright (c) 2004-2010 Leonard Richardson"
-__license__ = "New-style BSD"
-
-from sgmllib import SGMLParser, SGMLParseError
-import codecs
-import markupbase
-import types
-import re
-import sgmllib
-try:
- from htmlentitydefs import name2codepoint
-except ImportError:
- name2codepoint = {}
-try:
- set
-except NameError:
- from sets import Set as set
-
-#These hacks make Beautiful Soup able to parse XML with namespaces
-sgmllib.tagfind = re.compile('[a-zA-Z][-_.:a-zA-Z0-9]*')
-markupbase._declname_match = re.compile(r'[a-zA-Z][-_.:a-zA-Z0-9]*\s*').match
-
-DEFAULT_OUTPUT_ENCODING = "utf-8"
-
-def _match_css_class(str):
- """Build a RE to match the given CSS class."""
- return re.compile(r"(^|.*\s)%s($|\s)" % str)
-
-# First, the classes that represent markup elements.
-
-class PageElement(object):
- """Contains the navigational information for some part of the page
- (either a tag or a piece of text)"""
-
- def setup(self, parent=None, previous=None):
- """Sets up the initial relations between this element and
- other elements."""
- self.parent = parent
- self.previous = previous
- self.next = None
- self.previousSibling = None
- self.nextSibling = None
- if self.parent and self.parent.contents:
- self.previousSibling = self.parent.contents[-1]
- self.previousSibling.nextSibling = self
-
- def replaceWith(self, replaceWith):
- oldParent = self.parent
- myIndex = self.parent.index(self)
- if hasattr(replaceWith, "parent")\
- and replaceWith.parent is self.parent:
- # We're replacing this element with one of its siblings.
- index = replaceWith.parent.index(replaceWith)
- if index and index < myIndex:
- # Furthermore, it comes before this element. That
- # means that when we extract it, the index of this
- # element will change.
- myIndex = myIndex - 1
- self.extract()
- oldParent.insert(myIndex, replaceWith)
-
- def replaceWithChildren(self):
- myParent = self.parent
- myIndex = self.parent.index(self)
- self.extract()
- reversedChildren = list(self.contents)
- reversedChildren.reverse()
- for child in reversedChildren:
- myParent.insert(myIndex, child)
-
- def extract(self):
- """Destructively rips this element out of the tree."""
- if self.parent:
- try:
- del self.parent.contents[self.parent.index(self)]
- except ValueError:
- pass
-
- #Find the two elements that would be next to each other if
- #this element (and any children) hadn't been parsed. Connect
- #the two.
- lastChild = self._lastRecursiveChild()
- nextElement = lastChild.next
-
- if self.previous:
- self.previous.next = nextElement
- if nextElement:
- nextElement.previous = self.previous
- self.previous = None
- lastChild.next = None
-
- self.parent = None
- if self.previousSibling:
- self.previousSibling.nextSibling = self.nextSibling
- if self.nextSibling:
- self.nextSibling.previousSibling = self.previousSibling
- self.previousSibling = self.nextSibling = None
- return self
-
- def _lastRecursiveChild(self):
- "Finds the last element beneath this object to be parsed."
- lastChild = self
- while hasattr(lastChild, 'contents') and lastChild.contents:
- lastChild = lastChild.contents[-1]
- return lastChild
-
- def insert(self, position, newChild):
- if isinstance(newChild, basestring) \
- and not isinstance(newChild, NavigableString):
- newChild = NavigableString(newChild)
-
- position = min(position, len(self.contents))
- if hasattr(newChild, 'parent') and newChild.parent is not None:
- # We're 'inserting' an element that's already one
- # of this object's children.
- if newChild.parent is self:
- index = self.index(newChild)
- if index > position:
- # Furthermore we're moving it further down the
- # list of this object's children. That means that
- # when we extract this element, our target index
- # will jump down one.
- position = position - 1
- newChild.extract()
-
- newChild.parent = self
- previousChild = None
- if position == 0:
- newChild.previousSibling = None
- newChild.previous = self
- else:
- previousChild = self.contents[position-1]
- newChild.previousSibling = previousChild
- newChild.previousSibling.nextSibling = newChild
- newChild.previous = previousChild._lastRecursiveChild()
- if newChild.previous:
- newChild.previous.next = newChild
-
- newChildsLastElement = newChild._lastRecursiveChild()
-
- if position >= len(self.contents):
- newChild.nextSibling = None
-
- parent = self
- parentsNextSibling = None
- while not parentsNextSibling:
- parentsNextSibling = parent.nextSibling
- parent = parent.parent
- if not parent: # This is the last element in the document.
- break
- if parentsNextSibling:
- newChildsLastElement.next = parentsNextSibling
- else:
- newChildsLastElement.next = None
- else:
- nextChild = self.contents[position]
- newChild.nextSibling = nextChild
- if newChild.nextSibling:
- newChild.nextSibling.previousSibling = newChild
- newChildsLastElement.next = nextChild
-
- if newChildsLastElement.next:
- newChildsLastElement.next.previous = newChildsLastElement
- self.contents.insert(position, newChild)
-
- def append(self, tag):
- """Appends the given tag to the contents of this tag."""
- self.insert(len(self.contents), tag)
-
- def findNext(self, name=None, attrs={}, text=None, **kwargs):
- """Returns the first item that matches the given criteria and
- appears after this Tag in the document."""
- return self._findOne(self.findAllNext, name, attrs, text, **kwargs)
-
- def findAllNext(self, name=None, attrs={}, text=None, limit=None,
- **kwargs):
- """Returns all items that match the given criteria and appear
- after this Tag in the document."""
- return self._findAll(name, attrs, text, limit, self.nextGenerator,
- **kwargs)
-
- def findNextSibling(self, name=None, attrs={}, text=None, **kwargs):
- """Returns the closest sibling to this Tag that matches the
- given criteria and appears after this Tag in the document."""
- return self._findOne(self.findNextSiblings, name, attrs, text,
- **kwargs)
-
- def findNextSiblings(self, name=None, attrs={}, text=None, limit=None,
- **kwargs):
- """Returns the siblings of this Tag that match the given
- criteria and appear after this Tag in the document."""
- return self._findAll(name, attrs, text, limit,
- self.nextSiblingGenerator, **kwargs)
- fetchNextSiblings = findNextSiblings # Compatibility with pre-3.x
-
- def findPrevious(self, name=None, attrs={}, text=None, **kwargs):
- """Returns the first item that matches the given criteria and
- appears before this Tag in the document."""
- return self._findOne(self.findAllPrevious, name, attrs, text, **kwargs)
-
- def findAllPrevious(self, name=None, attrs={}, text=None, limit=None,
- **kwargs):
- """Returns all items that match the given criteria and appear
- before this Tag in the document."""
- return self._findAll(name, attrs, text, limit, self.previousGenerator,
- **kwargs)
- fetchPrevious = findAllPrevious # Compatibility with pre-3.x
-
- def findPreviousSibling(self, name=None, attrs={}, text=None, **kwargs):
- """Returns the closest sibling to this Tag that matches the
- given criteria and appears before this Tag in the document."""
- return self._findOne(self.findPreviousSiblings, name, attrs, text,
- **kwargs)
-
- def findPreviousSiblings(self, name=None, attrs={}, text=None,
- limit=None, **kwargs):
- """Returns the siblings of this Tag that match the given
- criteria and appear before this Tag in the document."""
- return self._findAll(name, attrs, text, limit,
- self.previousSiblingGenerator, **kwargs)
- fetchPreviousSiblings = findPreviousSiblings # Compatibility with pre-3.x
-
- def findParent(self, name=None, attrs={}, **kwargs):
- """Returns the closest parent of this Tag that matches the given
- criteria."""
- # NOTE: We can't use _findOne because findParents takes a different
- # set of arguments.
- r = None
- l = self.findParents(name, attrs, 1)
- if l:
- r = l[0]
- return r
-
- def findParents(self, name=None, attrs={}, limit=None, **kwargs):
- """Returns the parents of this Tag that match the given
- criteria."""
-
- return self._findAll(name, attrs, None, limit, self.parentGenerator,
- **kwargs)
- fetchParents = findParents # Compatibility with pre-3.x
-
- #These methods do the real heavy lifting.
-
- def _findOne(self, method, name, attrs, text, **kwargs):
- r = None
- l = method(name, attrs, text, 1, **kwargs)
- if l:
- r = l[0]
- return r
-
- def _findAll(self, name, attrs, text, limit, generator, **kwargs):
- "Iterates over a generator looking for things that match."
-
- if isinstance(name, SoupStrainer):
- strainer = name
- # (Possibly) special case some findAll*(...) searches
- elif text is None and not limit and not attrs and not kwargs:
- # findAll*(True)
- if name is True:
- return [element for element in generator()
- if isinstance(element, Tag)]
- # findAll*('tag-name')
- elif isinstance(name, basestring):
- return [element for element in generator()
- if isinstance(element, Tag) and
- element.name == name]
- else:
- strainer = SoupStrainer(name, attrs, text, **kwargs)
- # Build a SoupStrainer
- else:
- strainer = SoupStrainer(name, attrs, text, **kwargs)
- results = ResultSet(strainer)
- g = generator()
- while True:
- try:
- i = g.next()
- except StopIteration:
- break
- if i:
- found = strainer.search(i)
- if found:
- results.append(found)
- if limit and len(results) >= limit:
- break
- return results
-
- #These Generators can be used to navigate starting from both
- #NavigableStrings and Tags.
- def nextGenerator(self):
- i = self
- while i is not None:
- i = i.next
- yield i
-
- def nextSiblingGenerator(self):
- i = self
- while i is not None:
- i = i.nextSibling
- yield i
-
- def previousGenerator(self):
- i = self
- while i is not None:
- i = i.previous
- yield i
-
- def previousSiblingGenerator(self):
- i = self
- while i is not None:
- i = i.previousSibling
- yield i
-
- def parentGenerator(self):
- i = self
- while i is not None:
- i = i.parent
- yield i
-
- # Utility methods
- def substituteEncoding(self, str, encoding=None):
- encoding = encoding or "utf-8"
- return str.replace("%SOUP-ENCODING%", encoding)
-
- def toEncoding(self, s, encoding=None):
- """Encodes an object to a string in some encoding, or to Unicode.
- ."""
- if isinstance(s, unicode):
- if encoding:
- s = s.encode(encoding)
- elif isinstance(s, str):
- if encoding:
- s = s.encode(encoding)
- else:
- s = unicode(s)
- else:
- if encoding:
- s = self.toEncoding(str(s), encoding)
- else:
- s = unicode(s)
- return s
-
-class NavigableString(unicode, PageElement):
-
- def __new__(cls, value):
- """Create a new NavigableString.
-
- When unpickling a NavigableString, this method is called with
- the string in DEFAULT_OUTPUT_ENCODING. That encoding needs to be
- passed in to the superclass's __new__ or the superclass won't know
- how to handle non-ASCII characters.
- """
- if isinstance(value, unicode):
- return unicode.__new__(cls, value)
- return unicode.__new__(cls, value, DEFAULT_OUTPUT_ENCODING)
-
- def __getnewargs__(self):
- return (NavigableString.__str__(self),)
-
- def __getattr__(self, attr):
- """text.string gives you text. This is for backwards
- compatibility for Navigable*String, but for CData* it lets you
- get the string without the CData wrapper."""
- if attr == 'string':
- return self
- else:
- raise AttributeError, "'%s' object has no attribute '%s'" % (self.__class__.__name__, attr)
-
- def __unicode__(self):
- return str(self).decode(DEFAULT_OUTPUT_ENCODING)
-
- def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
- if encoding:
- return self.encode(encoding)
- else:
- return self
-
-class CData(NavigableString):
-
- def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
- return "<![CDATA[%s]]>" % NavigableString.__str__(self, encoding)
-
-class ProcessingInstruction(NavigableString):
- def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
- output = self
- if "%SOUP-ENCODING%" in output:
- output = self.substituteEncoding(output, encoding)
- return "<?%s?>" % self.toEncoding(output, encoding)
-
-class Comment(NavigableString):
- def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
- return "<!--%s-->" % NavigableString.__str__(self, encoding)
-
-class Declaration(NavigableString):
- def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
- return "<!%s>" % NavigableString.__str__(self, encoding)
-
-class Tag(PageElement):
-
- """Represents a found HTML tag with its attributes and contents."""
-
- def _invert(h):
- "Cheap function to invert a hash."
- i = {}
- for k,v in h.items():
- i[v] = k
- return i
-
- XML_ENTITIES_TO_SPECIAL_CHARS = { "apos" : "'",
- "quot" : '"',
- "amp" : "&",
- "lt" : "<",
- "gt" : ">" }
-
- XML_SPECIAL_CHARS_TO_ENTITIES = _invert(XML_ENTITIES_TO_SPECIAL_CHARS)
-
- def _convertEntities(self, match):
- """Used in a call to re.sub to replace HTML, XML, and numeric
- entities with the appropriate Unicode characters. If HTML
- entities are being converted, any unrecognized entities are
- escaped."""
- x = match.group(1)
- if self.convertHTMLEntities and x in name2codepoint:
- return unichr(name2codepoint[x])
- elif x in self.XML_ENTITIES_TO_SPECIAL_CHARS:
- if self.convertXMLEntities:
- return self.XML_ENTITIES_TO_SPECIAL_CHARS[x]
- else:
- return u'&%s;' % x
- elif len(x) > 0 and x[0] == '#':
- # Handle numeric entities
- if len(x) > 1 and x[1] == 'x':
- return unichr(int(x[2:], 16))
- else:
- return unichr(int(x[1:]))
-
- elif self.escapeUnrecognizedEntities:
- return u'&%s;' % x
- else:
- return u'&%s;' % x
-
- def __init__(self, parser, name, attrs=None, parent=None,
- previous=None):
- "Basic constructor."
-
- # We don't actually store the parser object: that lets extracted
- # chunks be garbage-collected
- self.parserClass = parser.__class__
- self.isSelfClosing = parser.isSelfClosingTag(name)
- self.name = name
- if attrs is None:
- attrs = []
- elif isinstance(attrs, dict):
- attrs = attrs.items()
- self.attrs = attrs
- self.contents = []
- self.setup(parent, previous)
- self.hidden = False
- self.containsSubstitutions = False
- self.convertHTMLEntities = parser.convertHTMLEntities
- self.convertXMLEntities = parser.convertXMLEntities
- self.escapeUnrecognizedEntities = parser.escapeUnrecognizedEntities
-
- # Convert any HTML, XML, or numeric entities in the attribute values.
- convert = lambda(k, val): (k,
- re.sub("&(#\d+|#x[0-9a-fA-F]+|\w+);",
- self._convertEntities,
- val))
- self.attrs = map(convert, self.attrs)
-
- def getString(self):
- if (len(self.contents) == 1
- and isinstance(self.contents[0], NavigableString)):
- return self.contents[0]
-
- def setString(self, string):
- """Replace the contents of the tag with a string"""
- self.clear()
- self.append(string)
-
- string = property(getString, setString)
-
- def getText(self, separator=u""):
- if not len(self.contents):
- return u""
- stopNode = self._lastRecursiveChild().next
- strings = []
- current = self.contents[0]
- while current is not stopNode:
- if isinstance(current, NavigableString):
- strings.append(current.strip())
- current = current.next
- return separator.join(strings)
-
- text = property(getText)
-
- def get(self, key, default=None):
- """Returns the value of the 'key' attribute for the tag, or
- the value given for 'default' if it doesn't have that
- attribute."""
- return self._getAttrMap().get(key, default)
-
- def clear(self):
- """Extract all children."""
- for child in self.contents[:]:
- child.extract()
-
- def index(self, element):
- for i, child in enumerate(self.contents):
- if child is element:
- return i
- raise ValueError("Tag.index: element not in tag")
-
- def has_key(self, key):
- return self._getAttrMap().has_key(key)
-
- def __getitem__(self, key):
- """tag[key] returns the value of the 'key' attribute for the tag,
- and throws an exception if it's not there."""
- return self._getAttrMap()[key]
-
- def __iter__(self):
- "Iterating over a tag iterates over its contents."
- return iter(self.contents)
-
- def __len__(self):
- "The length of a tag is the length of its list of contents."
- return len(self.contents)
-
- def __contains__(self, x):
- return x in self.contents
-
- def __nonzero__(self):
- "A tag is non-None even if it has no contents."
- return True
-
- def __setitem__(self, key, value):
- """Setting tag[key] sets the value of the 'key' attribute for the
- tag."""
- self._getAttrMap()
- self.attrMap[key] = value
- found = False
- for i in range(0, len(self.attrs)):
- if self.attrs[i][0] == key:
- self.attrs[i] = (key, value)
- found = True
- if not found:
- self.attrs.append((key, value))
- self._getAttrMap()[key] = value
-
- def __delitem__(self, key):
- "Deleting tag[key] deletes all 'key' attributes for the tag."
- for item in self.attrs:
- if item[0] == key:
- self.attrs.remove(item)
- #We don't break because bad HTML can define the same
- #attribute multiple times.
- self._getAttrMap()
- if self.attrMap.has_key(key):
- del self.attrMap[key]
-
- def __call__(self, *args, **kwargs):
- """Calling a tag like a function is the same as calling its
- findAll() method. Eg. tag('a') returns a list of all the A tags
- found within this tag."""
- return apply(self.findAll, args, kwargs)
-
- def __getattr__(self, tag):
- #print "Getattr %s.%s" % (self.__class__, tag)
- if len(tag) > 3 and tag.rfind('Tag') == len(tag)-3:
- return self.find(tag[:-3])
- elif tag.find('__') != 0:
- return self.find(tag)
- raise AttributeError, "'%s' object has no attribute '%s'" % (self.__class__, tag)
-
- def __eq__(self, other):
- """Returns true iff this tag has the same name, the same attributes,
- and the same contents (recursively) as the given tag.
-
- NOTE: right now this will return false if two tags have the
- same attributes in a different order. Should this be fixed?"""
- if other is self:
- return True
- if not hasattr(other, 'name') or not hasattr(other, 'attrs') or not hasattr(other, 'contents') or self.name != other.name or self.attrs != other.attrs or len(self) != len(other):
- return False
- for i in range(0, len(self.contents)):
- if self.contents[i] != other.contents[i]:
- return False
- return True
-
- def __ne__(self, other):
- """Returns true iff this tag is not identical to the other tag,
- as defined in __eq__."""
- return not self == other
-
- def __repr__(self, encoding=DEFAULT_OUTPUT_ENCODING):
- """Renders this tag as a string."""
- return self.__str__(encoding)
-
- def __unicode__(self):
- return self.__str__(None)
-
- BARE_AMPERSAND_OR_BRACKET = re.compile("([<>]|"
- + "&(?!#\d+;|#x[0-9a-fA-F]+;|\w+;)"
- + ")")
-
- def _sub_entity(self, x):
- """Used with a regular expression to substitute the
- appropriate XML entity for an XML special character."""
- return "&" + self.XML_SPECIAL_CHARS_TO_ENTITIES[x.group(0)[0]] + ";"
-
- def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING,
- prettyPrint=False, indentLevel=0):
- """Returns a string or Unicode representation of this tag and
- its contents. To get Unicode, pass None for encoding.
-
- NOTE: since Python's HTML parser consumes whitespace, this
- method is not certain to reproduce the whitespace present in
- the original string."""
-
- encodedName = self.toEncoding(self.name, encoding)
-
- attrs = []
- if self.attrs:
- for key, val in self.attrs:
- fmt = '%s="%s"'
- if isinstance(val, basestring):
- if self.containsSubstitutions and '%SOUP-ENCODING%' in val:
- val = self.substituteEncoding(val, encoding)
-
- # The attribute value either:
- #
- # * Contains no embedded double quotes or single quotes.
- # No problem: we enclose it in double quotes.
- # * Contains embedded single quotes. No problem:
- # double quotes work here too.
- # * Contains embedded double quotes. No problem:
- # we enclose it in single quotes.
- # * Embeds both single _and_ double quotes. This
- # can't happen naturally, but it can happen if
- # you modify an attribute value after parsing
- # the document. Now we have a bit of a
- # problem. We solve it by enclosing the
- # attribute in single quotes, and escaping any
- # embedded single quotes to XML entities.
- if '"' in val:
- fmt = "%s='%s'"
- if "'" in val:
- # TODO: replace with apos when
- # appropriate.
- val = val.replace("'", "&squot;")
-
- # Now we're okay w/r/t quotes. But the attribute
- # value might also contain angle brackets, or
- # ampersands that aren't part of entities. We need
- # to escape those to XML entities too.
- val = self.BARE_AMPERSAND_OR_BRACKET.sub(self._sub_entity, val)
-
- attrs.append(fmt % (self.toEncoding(key, encoding),
- self.toEncoding(val, encoding)))
- close = ''
- closeTag = ''
- if self.isSelfClosing:
- close = ' /'
- else:
- closeTag = '</%s>' % encodedName
-
- indentTag, indentContents = 0, 0
- if prettyPrint:
- indentTag = indentLevel
- space = (' ' * (indentTag-1))
- indentContents = indentTag + 1
- contents = self.renderContents(encoding, prettyPrint, indentContents)
- if self.hidden:
- s = contents
- else:
- s = []
- attributeString = ''
- if attrs:
- attributeString = ' ' + ' '.join(attrs)
- if prettyPrint:
- s.append(space)
- s.append('<%s%s%s>' % (encodedName, attributeString, close))
- if prettyPrint:
- s.append("\n")
- s.append(contents)
- if prettyPrint and contents and contents[-1] != "\n":
- s.append("\n")
- if prettyPrint and closeTag:
- s.append(space)
- s.append(closeTag)
- if prettyPrint and closeTag and self.nextSibling:
- s.append("\n")
- s = ''.join(s)
- return s
-
- def decompose(self):
- """Recursively destroys the contents of this tree."""
- self.extract()
- if len(self.contents) == 0:
- return
- current = self.contents[0]
- while current is not None:
- next = current.next
- if isinstance(current, Tag):
- del current.contents[:]
- current.parent = None
- current.previous = None
- current.previousSibling = None
- current.next = None
- current.nextSibling = None
- current = next
-
- def prettify(self, encoding=DEFAULT_OUTPUT_ENCODING):
- return self.__str__(encoding, True)
-
- def renderContents(self, encoding=DEFAULT_OUTPUT_ENCODING,
- prettyPrint=False, indentLevel=0):
- """Renders the contents of this tag as a string in the given
- encoding. If encoding is None, returns a Unicode string.."""
- s=[]
- for c in self:
- text = None
- if isinstance(c, NavigableString):
- text = c.__str__(encoding)
- elif isinstance(c, Tag):
- s.append(c.__str__(encoding, prettyPrint, indentLevel))
- if text and prettyPrint:
- text = text.strip()
- if text:
- if prettyPrint:
- s.append(" " * (indentLevel-1))
- s.append(text)
- if prettyPrint:
- s.append("\n")
- return ''.join(s)
-
- #Soup methods
-
- def find(self, name=None, attrs={}, recursive=True, text=None,
- **kwargs):
- """Return only the first child of this Tag matching the given
- criteria."""
- r = None
- l = self.findAll(name, attrs, recursive, text, 1, **kwargs)
- if l:
- r = l[0]
- return r
- findChild = find
-
- def findAll(self, name=None, attrs={}, recursive=True, text=None,
- limit=None, **kwargs):
- """Extracts a list of Tag objects that match the given
- criteria. You can specify the name of the Tag and any
- attributes you want the Tag to have.
-
- The value of a key-value pair in the 'attrs' map can be a
- string, a list of strings, a regular expression object, or a
- callable that takes a string and returns whether or not the
- string matches for some custom definition of 'matches'. The
- same is true of the tag name."""
- generator = self.recursiveChildGenerator
- if not recursive:
- generator = self.childGenerator
- return self._findAll(name, attrs, text, limit, generator, **kwargs)
- findChildren = findAll
-
- # Pre-3.x compatibility methods
- first = find
- fetch = findAll
-
- def fetchText(self, text=None, recursive=True, limit=None):
- return self.findAll(text=text, recursive=recursive, limit=limit)
-
- def firstText(self, text=None, recursive=True):
- return self.find(text=text, recursive=recursive)
-
- #Private methods
-
- def _getAttrMap(self):
- """Initializes a map representation of this tag's attributes,
- if not already initialized."""
- if not getattr(self, 'attrMap'):
- self.attrMap = {}
- for (key, value) in self.attrs:
- self.attrMap[key] = value
- return self.attrMap
-
- #Generator methods
- def childGenerator(self):
- # Just use the iterator from the contents
- return iter(self.contents)
-
- def recursiveChildGenerator(self):
- if not len(self.contents):
- raise StopIteration
- stopNode = self._lastRecursiveChild().next
- current = self.contents[0]
- while current is not stopNode:
- yield current
- current = current.next
-
-
-# Next, a couple classes to represent queries and their results.
-class SoupStrainer:
- """Encapsulates a number of ways of matching a markup element (tag or
- text)."""
-
- def __init__(self, name=None, attrs={}, text=None, **kwargs):
- self.name = name
- if isinstance(attrs, basestring):
- kwargs['class'] = _match_css_class(attrs)
- attrs = None
- if kwargs:
- if attrs:
- attrs = attrs.copy()
- attrs.update(kwargs)
- else:
- attrs = kwargs
- self.attrs = attrs
- self.text = text
-
- def __str__(self):
- if self.text:
- return self.text
- else:
- return "%s|%s" % (self.name, self.attrs)
-
- def searchTag(self, markupName=None, markupAttrs={}):
- found = None
- markup = None
- if isinstance(markupName, Tag):
- markup = markupName
- markupAttrs = markup
- callFunctionWithTagData = callable(self.name) \
- and not isinstance(markupName, Tag)
-
- if (not self.name) \
- or callFunctionWithTagData \
- or (markup and self._matches(markup, self.name)) \
- or (not markup and self._matches(markupName, self.name)):
- if callFunctionWithTagData:
- match = self.name(markupName, markupAttrs)
- else:
- match = True
- markupAttrMap = None
- for attr, matchAgainst in self.attrs.items():
- if not markupAttrMap:
- if hasattr(markupAttrs, 'get'):
- markupAttrMap = markupAttrs
- else:
- markupAttrMap = {}
- for k,v in markupAttrs:
- markupAttrMap[k] = v
- attrValue = markupAttrMap.get(attr)
- if not self._matches(attrValue, matchAgainst):
- match = False
- break
- if match:
- if markup:
- found = markup
- else:
- found = markupName
- return found
-
- def search(self, markup):
- #print 'looking for %s in %s' % (self, markup)
- found = None
- # If given a list of items, scan it for a text element that
- # matches.
- if hasattr(markup, "__iter__") \
- and not isinstance(markup, Tag):
- for element in markup:
- if isinstance(element, NavigableString) \
- and self.search(element):
- found = element
- break
- # If it's a Tag, make sure its name or attributes match.
- # Don't bother with Tags if we're searching for text.
- elif isinstance(markup, Tag):
- if not self.text:
- found = self.searchTag(markup)
- # If it's text, make sure the text matches.
- elif isinstance(markup, NavigableString) or \
- isinstance(markup, basestring):
- if self._matches(markup, self.text):
- found = markup
- else:
- raise Exception, "I don't know how to match against a %s" \
- % markup.__class__
- return found
-
- def _matches(self, markup, matchAgainst):
- #print "Matching %s against %s" % (markup, matchAgainst)
- result = False
- if matchAgainst is True:
- result = markup is not None
- elif callable(matchAgainst):
- result = matchAgainst(markup)
- else:
- #Custom match methods take the tag as an argument, but all
- #other ways of matching match the tag name as a string.
- if isinstance(markup, Tag):
- markup = markup.name
- if markup and not isinstance(markup, basestring):
- markup = unicode(markup)
- #Now we know that chunk is either a string, or None.
- if hasattr(matchAgainst, 'match'):
- # It's a regexp object.
- result = markup and matchAgainst.search(markup)
- elif hasattr(matchAgainst, '__iter__'): # list-like
- result = markup in matchAgainst
- elif hasattr(matchAgainst, 'items'):
- result = markup.has_key(matchAgainst)
- elif matchAgainst and isinstance(markup, basestring):
- if isinstance(markup, unicode):
- matchAgainst = unicode(matchAgainst)
- else:
- matchAgainst = str(matchAgainst)
-
- if not result:
- result = matchAgainst == markup
- return result
-
-class ResultSet(list):
- """A ResultSet is just a list that keeps track of the SoupStrainer
- that created it."""
- def __init__(self, source):
- list.__init__([])
- self.source = source
-
-# Now, some helper functions.
-
-def buildTagMap(default, *args):
- """Turns a list of maps, lists, or scalars into a single map.
- Used to build the SELF_CLOSING_TAGS, NESTABLE_TAGS, and
- NESTING_RESET_TAGS maps out of lists and partial maps."""
- built = {}
- for portion in args:
- if hasattr(portion, 'items'):
- #It's a map. Merge it.
- for k,v in portion.items():
- built[k] = v
- elif hasattr(portion, '__iter__'): # is a list
- #It's a list. Map each item to the default.
- for k in portion:
- built[k] = default
- else:
- #It's a scalar. Map it to the default.
- built[portion] = default
- return built
-
-# Now, the parser classes.
-
-class BeautifulStoneSoup(Tag, SGMLParser):
-
- """This class contains the basic parser and search code. It defines
- a parser that knows nothing about tag behavior except for the
- following:
-
- You can't close a tag without closing all the tags it encloses.
- That is, "<foo><bar></foo>" actually means
- "<foo><bar></bar></foo>".
-
- [Another possible explanation is "<foo><bar /></foo>", but since
- this class defines no SELF_CLOSING_TAGS, it will never use that
- explanation.]
-
- This class is useful for parsing XML or made-up markup languages,
- or when BeautifulSoup makes an assumption counter to what you were
- expecting."""
-
- SELF_CLOSING_TAGS = {}
- NESTABLE_TAGS = {}
- RESET_NESTING_TAGS = {}
- QUOTE_TAGS = {}
- PRESERVE_WHITESPACE_TAGS = []
-
- MARKUP_MASSAGE = [(re.compile('(<[^<>]*)/>'),
- lambda x: x.group(1) + ' />'),
- (re.compile('<!\s+([^<>]*)>'),
- lambda x: '<!' + x.group(1) + '>')
- ]
-
- ROOT_TAG_NAME = u'[document]'
-
- HTML_ENTITIES = "html"
- XML_ENTITIES = "xml"
- XHTML_ENTITIES = "xhtml"
- # TODO: This only exists for backwards-compatibility
- ALL_ENTITIES = XHTML_ENTITIES
-
- # Used when determining whether a text node is all whitespace and
- # can be replaced with a single space. A text node that contains
- # fancy Unicode spaces (usually non-breaking) should be left
- # alone.
- STRIP_ASCII_SPACES = { 9: None, 10: None, 12: None, 13: None, 32: None, }
-
- def __init__(self, markup="", parseOnlyThese=None, fromEncoding=None,
- markupMassage=True, smartQuotesTo=XML_ENTITIES,
- convertEntities=None, selfClosingTags=None, isHTML=False):
- """The Soup object is initialized as the 'root tag', and the
- provided markup (which can be a string or a file-like object)
- is fed into the underlying parser.
-
- sgmllib will process most bad HTML, and the BeautifulSoup
- class has some tricks for dealing with some HTML that kills
- sgmllib, but Beautiful Soup can nonetheless choke or lose data
- if your data uses self-closing tags or declarations
- incorrectly.
-
- By default, Beautiful Soup uses regexes to sanitize input,
- avoiding the vast majority of these problems. If the problems
- don't apply to you, pass in False for markupMassage, and
- you'll get better performance.
-
- The default parser massage techniques fix the two most common
- instances of invalid HTML that choke sgmllib:
-
- <br/> (No space between name of closing tag and tag close)
- <! --Comment--> (Extraneous whitespace in declaration)
-
- You can pass in a custom list of (RE object, replace method)
- tuples to get Beautiful Soup to scrub your input the way you
- want."""
-
- self.parseOnlyThese = parseOnlyThese
- self.fromEncoding = fromEncoding
- self.smartQuotesTo = smartQuotesTo
- self.convertEntities = convertEntities
- # Set the rules for how we'll deal with the entities we
- # encounter
- if self.convertEntities:
- # It doesn't make sense to convert encoded characters to
- # entities even while you're converting entities to Unicode.
- # Just convert it all to Unicode.
- self.smartQuotesTo = None
- if convertEntities == self.HTML_ENTITIES:
- self.convertXMLEntities = False
- self.convertHTMLEntities = True
- self.escapeUnrecognizedEntities = True
- elif convertEntities == self.XHTML_ENTITIES:
- self.convertXMLEntities = True
- self.convertHTMLEntities = True
- self.escapeUnrecognizedEntities = False
- elif convertEntities == self.XML_ENTITIES:
- self.convertXMLEntities = True
- self.convertHTMLEntities = False
- self.escapeUnrecognizedEntities = False
- else:
- self.convertXMLEntities = False
- self.convertHTMLEntities = False
- self.escapeUnrecognizedEntities = False
-
- self.instanceSelfClosingTags = buildTagMap(None, selfClosingTags)
- SGMLParser.__init__(self)
-
- if hasattr(markup, 'read'): # It's a file-type object.
- markup = markup.read()
- self.markup = markup
- self.markupMassage = markupMassage
- try:
- self._feed(isHTML=isHTML)
- except StopParsing:
- pass
- self.markup = None # The markup can now be GCed
-
- def convert_charref(self, name):
- """This method fixes a bug in Python's SGMLParser."""
- try:
- n = int(name)
- except ValueError:
- return
- if not 0 <= n <= 127 : # ASCII ends at 127, not 255
- return
- return self.convert_codepoint(n)
-
- def _feed(self, inDocumentEncoding=None, isHTML=False):
- # Convert the document to Unicode.
- markup = self.markup
- if isinstance(markup, unicode):
- if not hasattr(self, 'originalEncoding'):
- self.originalEncoding = None
- else:
- dammit = UnicodeDammit\
- (markup, [self.fromEncoding, inDocumentEncoding],
- smartQuotesTo=self.smartQuotesTo, isHTML=isHTML)
- markup = dammit.unicode
- self.originalEncoding = dammit.originalEncoding
- self.declaredHTMLEncoding = dammit.declaredHTMLEncoding
- if markup:
- if self.markupMassage:
- if not hasattr(self.markupMassage, "__iter__"):
- self.markupMassage = self.MARKUP_MASSAGE
- for fix, m in self.markupMassage:
- markup = fix.sub(m, markup)
- # TODO: We get rid of markupMassage so that the
- # soup object can be deepcopied later on. Some
- # Python installations can't copy regexes. If anyone
- # was relying on the existence of markupMassage, this
- # might cause problems.
- del(self.markupMassage)
- self.reset()
-
- SGMLParser.feed(self, markup)
- # Close out any unfinished strings and close all the open tags.
- self.endData()
- while self.currentTag.name != self.ROOT_TAG_NAME:
- self.popTag()
-
- def __getattr__(self, methodName):
- """This method routes method call requests to either the SGMLParser
- superclass or the Tag superclass, depending on the method name."""
- #print "__getattr__ called on %s.%s" % (self.__class__, methodName)
-
- if methodName.startswith('start_') or methodName.startswith('end_') \
- or methodName.startswith('do_'):
- return SGMLParser.__getattr__(self, methodName)
- elif not methodName.startswith('__'):
- return Tag.__getattr__(self, methodName)
- else:
- raise AttributeError
-
- def isSelfClosingTag(self, name):
- """Returns true iff the given string is the name of a
- self-closing tag according to this parser."""
- return self.SELF_CLOSING_TAGS.has_key(name) \
- or self.instanceSelfClosingTags.has_key(name)
-
- def reset(self):
- Tag.__init__(self, self, self.ROOT_TAG_NAME)
- self.hidden = 1
- SGMLParser.reset(self)
- self.currentData = []
- self.currentTag = None
- self.tagStack = []
- self.quoteStack = []
- self.pushTag(self)
-
- def popTag(self):
- tag = self.tagStack.pop()
-
- #print "Pop", tag.name
- if self.tagStack:
- self.currentTag = self.tagStack[-1]
- return self.currentTag
-
- def pushTag(self, tag):
- #print "Push", tag.name
- if self.currentTag:
- self.currentTag.contents.append(tag)
- self.tagStack.append(tag)
- self.currentTag = self.tagStack[-1]
-
- def endData(self, containerClass=NavigableString):
- if self.currentData:
- currentData = u''.join(self.currentData)
- if (currentData.translate(self.STRIP_ASCII_SPACES) == '' and
- not set([tag.name for tag in self.tagStack]).intersection(
- self.PRESERVE_WHITESPACE_TAGS)):
- if '\n' in currentData:
- currentData = '\n'
- else:
- currentData = ' '
- self.currentData = []
- if self.parseOnlyThese and len(self.tagStack) <= 1 and \
- (not self.parseOnlyThese.text or \
- not self.parseOnlyThese.search(currentData)):
- return
- o = containerClass(currentData)
- o.setup(self.currentTag, self.previous)
- if self.previous:
- self.previous.next = o
- self.previous = o
- self.currentTag.contents.append(o)
-
-
- def _popToTag(self, name, inclusivePop=True):
- """Pops the tag stack up to and including the most recent
- instance of the given tag. If inclusivePop is false, pops the tag
- stack up to but *not* including the most recent instqance of
- the given tag."""
- #print "Popping to %s" % name
- if name == self.ROOT_TAG_NAME:
- return
-
- numPops = 0
- mostRecentTag = None
- for i in range(len(self.tagStack)-1, 0, -1):
- if name == self.tagStack[i].name:
- numPops = len(self.tagStack)-i
- break
- if not inclusivePop:
- numPops = numPops - 1
-
- for i in range(0, numPops):
- mostRecentTag = self.popTag()
- return mostRecentTag
-
- def _smartPop(self, name):
-
- """We need to pop up to the previous tag of this type, unless
- one of this tag's nesting reset triggers comes between this
- tag and the previous tag of this type, OR unless this tag is a
- generic nesting trigger and another generic nesting trigger
- comes between this tag and the previous tag of this type.
-
- Examples:
- <p>Foo<b>Bar *<p>* should pop to 'p', not 'b'.
- <p>Foo<table>Bar *<p>* should pop to 'table', not 'p'.
- <p>Foo<table><tr>Bar *<p>* should pop to 'tr', not 'p'.
-
- <li><ul><li> *<li>* should pop to 'ul', not the first 'li'.
- <tr><table><tr> *<tr>* should pop to 'table', not the first 'tr'
- <td><tr><td> *<td>* should pop to 'tr', not the first 'td'
- """
-
- nestingResetTriggers = self.NESTABLE_TAGS.get(name)
- isNestable = nestingResetTriggers != None
- isResetNesting = self.RESET_NESTING_TAGS.has_key(name)
- popTo = None
- inclusive = True
- for i in range(len(self.tagStack)-1, 0, -1):
- p = self.tagStack[i]
- if (not p or p.name == name) and not isNestable:
- #Non-nestable tags get popped to the top or to their
- #last occurance.
- popTo = name
- break
- if (nestingResetTriggers is not None
- and p.name in nestingResetTriggers) \
- or (nestingResetTriggers is None and isResetNesting
- and self.RESET_NESTING_TAGS.has_key(p.name)):
-
- #If we encounter one of the nesting reset triggers
- #peculiar to this tag, or we encounter another tag
- #that causes nesting to reset, pop up to but not
- #including that tag.
- popTo = p.name
- inclusive = False
- break
- p = p.parent
- if popTo:
- self._popToTag(popTo, inclusive)
-
- def unknown_starttag(self, name, attrs, selfClosing=0):
- #print "Start tag %s: %s" % (name, attrs)
- if self.quoteStack:
- #This is not a real tag.
- #print "<%s> is not real!" % name
- attrs = ''.join([' %s="%s"' % (x, y) for x, y in attrs])
- self.handle_data('<%s%s>' % (name, attrs))
- return
- self.endData()
-
- if not self.isSelfClosingTag(name) and not selfClosing:
- self._smartPop(name)
-
- if self.parseOnlyThese and len(self.tagStack) <= 1 \
- and (self.parseOnlyThese.text or not self.parseOnlyThese.searchTag(name, attrs)):
- return
-
- tag = Tag(self, name, attrs, self.currentTag, self.previous)
- if self.previous:
- self.previous.next = tag
- self.previous = tag
- self.pushTag(tag)
- if selfClosing or self.isSelfClosingTag(name):
- self.popTag()
- if name in self.QUOTE_TAGS:
- #print "Beginning quote (%s)" % name
- self.quoteStack.append(name)
- self.literal = 1
- return tag
-
- def unknown_endtag(self, name):
- #print "End tag %s" % name
- if self.quoteStack and self.quoteStack[-1] != name:
- #This is not a real end tag.
- #print "</%s> is not real!" % name
- self.handle_data('</%s>' % name)
- return
- self.endData()
- self._popToTag(name)
- if self.quoteStack and self.quoteStack[-1] == name:
- self.quoteStack.pop()
- self.literal = (len(self.quoteStack) > 0)
-
- def handle_data(self, data):
- self.currentData.append(data)
-
- def _toStringSubclass(self, text, subclass):
- """Adds a certain piece of text to the tree as a NavigableString
- subclass."""
- self.endData()
- self.handle_data(text)
- self.endData(subclass)
-
- def handle_pi(self, text):
- """Handle a processing instruction as a ProcessingInstruction
- object, possibly one with a %SOUP-ENCODING% slot into which an
- encoding will be plugged later."""
- if text[:3] == "xml":
- text = u"xml version='1.0' encoding='%SOUP-ENCODING%'"
- self._toStringSubclass(text, ProcessingInstruction)
-
- def handle_comment(self, text):
- "Handle comments as Comment objects."
- self._toStringSubclass(text, Comment)
-
- def handle_charref(self, ref):
- "Handle character references as data."
- if self.convertEntities:
- data = unichr(int(ref))
- else:
- data = '&#%s;' % ref
- self.handle_data(data)
-
- def handle_entityref(self, ref):
- """Handle entity references as data, possibly converting known
- HTML and/or XML entity references to the corresponding Unicode
- characters."""
- data = None
- if self.convertHTMLEntities:
- try:
- data = unichr(name2codepoint[ref])
- except KeyError:
- pass
-
- if not data and self.convertXMLEntities:
- data = self.XML_ENTITIES_TO_SPECIAL_CHARS.get(ref)
-
- if not data and self.convertHTMLEntities and \
- not self.XML_ENTITIES_TO_SPECIAL_CHARS.get(ref):
- # TODO: We've got a problem here. We're told this is
- # an entity reference, but it's not an XML entity
- # reference or an HTML entity reference. Nonetheless,
- # the logical thing to do is to pass it through as an
- # unrecognized entity reference.
- #
- # Except: when the input is "&carol;" this function
- # will be called with input "carol". When the input is
- # "AT&T", this function will be called with input
- # "T". We have no way of knowing whether a semicolon
- # was present originally, so we don't know whether
- # this is an unknown entity or just a misplaced
- # ampersand.
- #
- # The more common case is a misplaced ampersand, so I
- # escape the ampersand and omit the trailing semicolon.
- data = "&%s" % ref
- if not data:
- # This case is different from the one above, because we
- # haven't already gone through a supposedly comprehensive
- # mapping of entities to Unicode characters. We might not
- # have gone through any mapping at all. So the chances are
- # very high that this is a real entity, and not a
- # misplaced ampersand.
- data = "&%s;" % ref
- self.handle_data(data)
-
- def handle_decl(self, data):
- "Handle DOCTYPEs and the like as Declaration objects."
- self._toStringSubclass(data, Declaration)
-
- def parse_declaration(self, i):
- """Treat a bogus SGML declaration as raw data. Treat a CDATA
- declaration as a CData object."""
- j = None
- if self.rawdata[i:i+9] == '<![CDATA[':
- k = self.rawdata.find(']]>', i)
- if k == -1:
- k = len(self.rawdata)
- data = self.rawdata[i+9:k]
- j = k+3
- self._toStringSubclass(data, CData)
- else:
- try:
- j = SGMLParser.parse_declaration(self, i)
- except SGMLParseError:
- toHandle = self.rawdata[i:]
- self.handle_data(toHandle)
- j = i + len(toHandle)
- return j
-
-class BeautifulSoup(BeautifulStoneSoup):
-
- """This parser knows the following facts about HTML:
-
- * Some tags have no closing tag and should be interpreted as being
- closed as soon as they are encountered.
-
- * The text inside some tags (ie. 'script') may contain tags which
- are not really part of the document and which should be parsed
- as text, not tags. If you want to parse the text as tags, you can
- always fetch it and parse it explicitly.
-
- * Tag nesting rules:
-
- Most tags can't be nested at all. For instance, the occurance of
- a <p> tag should implicitly close the previous <p> tag.
-
- <p>Para1<p>Para2
- should be transformed into:
- <p>Para1</p><p>Para2
-
- Some tags can be nested arbitrarily. For instance, the occurance
- of a <blockquote> tag should _not_ implicitly close the previous
- <blockquote> tag.
-
- Alice said: <blockquote>Bob said: <blockquote>Blah
- should NOT be transformed into:
- Alice said: <blockquote>Bob said: </blockquote><blockquote>Blah
-
- Some tags can be nested, but the nesting is reset by the
- interposition of other tags. For instance, a <tr> tag should
- implicitly close the previous <tr> tag within the same <table>,
- but not close a <tr> tag in another table.
-
- <table><tr>Blah<tr>Blah
- should be transformed into:
- <table><tr>Blah</tr><tr>Blah
- but,
- <tr>Blah<table><tr>Blah
- should NOT be transformed into
- <tr>Blah<table></tr><tr>Blah
-
- Differing assumptions about tag nesting rules are a major source
- of problems with the BeautifulSoup class. If BeautifulSoup is not
- treating as nestable a tag your page author treats as nestable,
- try ICantBelieveItsBeautifulSoup, MinimalSoup, or
- BeautifulStoneSoup before writing your own subclass."""
-
- def __init__(self, *args, **kwargs):
- if not kwargs.has_key('smartQuotesTo'):
- kwargs['smartQuotesTo'] = self.HTML_ENTITIES
- kwargs['isHTML'] = True
- BeautifulStoneSoup.__init__(self, *args, **kwargs)
-
- SELF_CLOSING_TAGS = buildTagMap(None,
- ('br' , 'hr', 'input', 'img', 'meta',
- 'spacer', 'link', 'frame', 'base', 'col'))
-
- PRESERVE_WHITESPACE_TAGS = set(['pre', 'textarea'])
-
- QUOTE_TAGS = {'script' : None, 'textarea' : None}
-
- #According to the HTML standard, each of these inline tags can
- #contain another tag of the same type. Furthermore, it's common
- #to actually use these tags this way.
- NESTABLE_INLINE_TAGS = ('span', 'font', 'q', 'object', 'bdo', 'sub', 'sup',
- 'center')
-
- #According to the HTML standard, these block tags can contain
- #another tag of the same type. Furthermore, it's common
- #to actually use these tags this way.
- NESTABLE_BLOCK_TAGS = ('blockquote', 'div', 'fieldset', 'ins', 'del')
-
- #Lists can contain other lists, but there are restrictions.
- NESTABLE_LIST_TAGS = { 'ol' : [],
- 'ul' : [],
- 'li' : ['ul', 'ol'],
- 'dl' : [],
- 'dd' : ['dl'],
- 'dt' : ['dl'] }
-
- #Tables can contain other tables, but there are restrictions.
- NESTABLE_TABLE_TAGS = {'table' : [],
- 'tr' : ['table', 'tbody', 'tfoot', 'thead'],
- 'td' : ['tr'],
- 'th' : ['tr'],
- 'thead' : ['table'],
- 'tbody' : ['table'],
- 'tfoot' : ['table'],
- }
-
- NON_NESTABLE_BLOCK_TAGS = ('address', 'form', 'p', 'pre')
-
- #If one of these tags is encountered, all tags up to the next tag of
- #this type are popped.
- RESET_NESTING_TAGS = buildTagMap(None, NESTABLE_BLOCK_TAGS, 'noscript',
- NON_NESTABLE_BLOCK_TAGS,
- NESTABLE_LIST_TAGS,
- NESTABLE_TABLE_TAGS)
-
- NESTABLE_TAGS = buildTagMap([], NESTABLE_INLINE_TAGS, NESTABLE_BLOCK_TAGS,
- NESTABLE_LIST_TAGS, NESTABLE_TABLE_TAGS)
-
- # Used to detect the charset in a META tag; see start_meta
- CHARSET_RE = re.compile("((^|;)\s*charset=)([^;]*)", re.M)
-
- def start_meta(self, attrs):
- """Beautiful Soup can detect a charset included in a META tag,
- try to convert the document to that charset, and re-parse the
- document from the beginning."""
- httpEquiv = None
- contentType = None
- contentTypeIndex = None
- tagNeedsEncodingSubstitution = False
-
- for i in range(0, len(attrs)):
- key, value = attrs[i]
- key = key.lower()
- if key == 'http-equiv':
- httpEquiv = value
- elif key == 'content':
- contentType = value
- contentTypeIndex = i
-
- if httpEquiv and contentType: # It's an interesting meta tag.
- match = self.CHARSET_RE.search(contentType)
- if match:
- if (self.declaredHTMLEncoding is not None or
- self.originalEncoding == self.fromEncoding):
- # An HTML encoding was sniffed while converting
- # the document to Unicode, or an HTML encoding was
- # sniffed during a previous pass through the
- # document, or an encoding was specified
- # explicitly and it worked. Rewrite the meta tag.
- def rewrite(match):
- return match.group(1) + "%SOUP-ENCODING%"
- newAttr = self.CHARSET_RE.sub(rewrite, contentType)
- attrs[contentTypeIndex] = (attrs[contentTypeIndex][0],
- newAttr)
- tagNeedsEncodingSubstitution = True
- else:
- # This is our first pass through the document.
- # Go through it again with the encoding information.
- newCharset = match.group(3)
- if newCharset and newCharset != self.originalEncoding:
- self.declaredHTMLEncoding = newCharset
- self._feed(self.declaredHTMLEncoding)
- raise StopParsing
- pass
- tag = self.unknown_starttag("meta", attrs)
- if tag and tagNeedsEncodingSubstitution:
- tag.containsSubstitutions = True
-
-class StopParsing(Exception):
- pass
-
-class ICantBelieveItsBeautifulSoup(BeautifulSoup):
-
- """The BeautifulSoup class is oriented towards skipping over
- common HTML errors like unclosed tags. However, sometimes it makes
- errors of its own. For instance, consider this fragment:
-
- <b>Foo<b>Bar</b></b>
-
- This is perfectly valid (if bizarre) HTML. However, the
- BeautifulSoup class will implicitly close the first b tag when it
- encounters the second 'b'. It will think the author wrote
- "<b>Foo<b>Bar", and didn't close the first 'b' tag, because
- there's no real-world reason to bold something that's already
- bold. When it encounters '</b></b>' it will close two more 'b'
- tags, for a grand total of three tags closed instead of two. This
- can throw off the rest of your document structure. The same is
- true of a number of other tags, listed below.
-
- It's much more common for someone to forget to close a 'b' tag
- than to actually use nested 'b' tags, and the BeautifulSoup class
- handles the common case. This class handles the not-co-common
- case: where you can't believe someone wrote what they did, but
- it's valid HTML and BeautifulSoup screwed up by assuming it
- wouldn't be."""
-
- I_CANT_BELIEVE_THEYRE_NESTABLE_INLINE_TAGS = \
- ('em', 'big', 'i', 'small', 'tt', 'abbr', 'acronym', 'strong',
- 'cite', 'code', 'dfn', 'kbd', 'samp', 'strong', 'var', 'b',
- 'big')
-
- I_CANT_BELIEVE_THEYRE_NESTABLE_BLOCK_TAGS = ('noscript',)
-
- NESTABLE_TAGS = buildTagMap([], BeautifulSoup.NESTABLE_TAGS,
- I_CANT_BELIEVE_THEYRE_NESTABLE_BLOCK_TAGS,
- I_CANT_BELIEVE_THEYRE_NESTABLE_INLINE_TAGS)
-
-class MinimalSoup(BeautifulSoup):
- """The MinimalSoup class is for parsing HTML that contains
- pathologically bad markup. It makes no assumptions about tag
- nesting, but it does know which tags are self-closing, that
- <script> tags contain Javascript and should not be parsed, that
- META tags may contain encoding information, and so on.
-
- This also makes it better for subclassing than BeautifulStoneSoup
- or BeautifulSoup."""
-
- RESET_NESTING_TAGS = buildTagMap('noscript')
- NESTABLE_TAGS = {}
-
-class BeautifulSOAP(BeautifulStoneSoup):
- """This class will push a tag with only a single string child into
- the tag's parent as an attribute. The attribute's name is the tag
- name, and the value is the string child. An example should give
- the flavor of the change:
-
- <foo><bar>baz</bar></foo>
- =>
- <foo bar="baz"><bar>baz</bar></foo>
-
- You can then access fooTag['bar'] instead of fooTag.barTag.string.
-
- This is, of course, useful for scraping structures that tend to
- use subelements instead of attributes, such as SOAP messages. Note
- that it modifies its input, so don't print the modified version
- out.
-
- I'm not sure how many people really want to use this class; let me
- know if you do. Mainly I like the name."""
-
- def popTag(self):
- if len(self.tagStack) > 1:
- tag = self.tagStack[-1]
- parent = self.tagStack[-2]
- parent._getAttrMap()
- if (isinstance(tag, Tag) and len(tag.contents) == 1 and
- isinstance(tag.contents[0], NavigableString) and
- not parent.attrMap.has_key(tag.name)):
- parent[tag.name] = tag.contents[0]
- BeautifulStoneSoup.popTag(self)
-
-#Enterprise class names! It has come to our attention that some people
-#think the names of the Beautiful Soup parser classes are too silly
-#and "unprofessional" for use in enterprise screen-scraping. We feel
-#your pain! For such-minded folk, the Beautiful Soup Consortium And
-#All-Night Kosher Bakery recommends renaming this file to
-#"RobustParser.py" (or, in cases of extreme enterprisiness,
-#"RobustParserBeanInterface.class") and using the following
-#enterprise-friendly class aliases:
-class RobustXMLParser(BeautifulStoneSoup):
- pass
-class RobustHTMLParser(BeautifulSoup):
- pass
-class RobustWackAssHTMLParser(ICantBelieveItsBeautifulSoup):
- pass
-class RobustInsanelyWackAssHTMLParser(MinimalSoup):
- pass
-class SimplifyingSOAPParser(BeautifulSOAP):
- pass
-
-######################################################
+# Copyright (C) 2019 Apple Inc. All rights reserved.
#
-# Bonus library: Unicode, Dammit
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+# 1. Redistributions of source code must retain the above copyright
+# notice, this list of conditions and the following disclaimer.
+# 2. Redistributions in binary form must reproduce the above copyright
+# notice, this list of conditions and the following disclaimer in the
+# documentation and/or other materials provided with the distribution.
#
-# This class forces XML data into a standard format (usually to UTF-8
-# or Unicode). It is heavily based on code from Mark Pilgrim's
-# Universal Feed Parser. It does not rewrite the XML or HTML to
-# reflect a new encoding: that happens in BeautifulStoneSoup.handle_pi
-# (XML) and BeautifulSoup.start_meta (HTML).
+# THIS SOFTWARE IS PROVIDED BY APPLE INC. AND ITS CONTRIBUTORS ``AS IS'' AND
+# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+# DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR ITS CONTRIBUTORS BE LIABLE FOR
+# ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-# Autodetects character encodings.
-# Download from http://chardet.feedparser.org/
-try:
- import chardet
-# import chardet.constants
-# chardet.constants._debug = 1
-except ImportError:
- chardet = None
+import sys
-# cjkcodecs and iconv_codec make Python know about more character encodings.
-# Both are available from http://cjkpython.i18n.org/
-# They're built in if you use Python 2.4.
-try:
- import cjkcodecs.aliases
-except ImportError:
- pass
-try:
- import iconv_codec
-except ImportError:
- pass
+if sys.version_info > (3, 0):
+ import html5lib
+ import webkitpy.thirdparty.autoinstalled.bs4 as bs4
+ from webkitpy.thirdparty.autoinstalled.bs4.builder import builder_registry, TreeBuilder
+ from webkitpy.thirdparty.autoinstalled.bs4.builder._htmlparser import HTMLParserTreeBuilder
-class UnicodeDammit:
- """A class for detecting the encoding of a *ML document and
- converting it to a Unicode string. If the source encoding is
- windows-1252, can replace MS smart quotes with their HTML or XML
- equivalents."""
+ class BeautifulSoup(bs4.BeautifulSoup):
+ HTML_ENTITIES = 'html'
- # This dictionary maps commonly seen values for "charset" in HTML
- # meta tags to the corresponding Python codec names. It only covers
- # values that aren't in Python's aliases and can't be determined
- # by the heuristics in find_codec.
- CHARSET_ALIASES = { "macintosh" : "mac-roman",
- "x-sjis" : "shift-jis" }
+ def __init__(self, *args, **kwargs):
+ if 'convertEntities' in kwargs:
+ del kwargs['convertEntities']
+ if 'parseOnlyThese' in kwargs:
+ kwargs['parse_only'] = kwargs['parseOnlyThese']
+ del kwargs['parseOnlyThese']
+ kwargs['features'] = self.HTML_ENTITIES
+ kwargs['builder'] = HTMLParserTreeBuilder
+ super(BeautifulSoup, self).__init__(*args, **kwargs)
- def __init__(self, markup, overrideEncodings=[],
- smartQuotesTo='xml', isHTML=False):
- self.declaredHTMLEncoding = None
- self.markup, documentEncoding, sniffedEncoding = \
- self._detectEncoding(markup, isHTML)
- self.smartQuotesTo = smartQuotesTo
- self.triedEncodings = []
- if markup == '' or isinstance(markup, unicode):
- self.originalEncoding = None
- self.unicode = unicode(markup)
- return
+ class BeautifulStoneSoup(bs4.BeautifulSoup):
+ XML_ENTITIES = 'xml'
- u = None
- for proposedEncoding in overrideEncodings:
- u = self._convertFrom(proposedEncoding)
- if u: break
- if not u:
- for proposedEncoding in (documentEncoding, sniffedEncoding):
- u = self._convertFrom(proposedEncoding)
- if u: break
+ def __init__(self, *args, **kwargs):
+ if 'convertEntities' in kwargs:
+ del kwargs['convertEntities']
+ kwargs['features'] = self.XML_ENTITIES
- # If no luck and we have auto-detection library, try that:
- if not u and chardet and not isinstance(self.markup, unicode):
- u = self._convertFrom(chardet.detect(self.markup)['encoding'])
+ # FIXME: This hack isn't strictly accurate, but getting lxml to work with the autoinstaller is a non-trivial task
+ kwargs['builder'] = HTMLParserTreeBuilder
+ super(BeautifulStoneSoup, self).__init__(*args, **kwargs)
- # As a last resort, try utf-8 and windows-1252:
- if not u:
- for proposed_encoding in ("utf-8", "windows-1252"):
- u = self._convertFrom(proposed_encoding)
- if u: break
-
- self.unicode = u
- if not u: self.originalEncoding = None
-
- def _subMSChar(self, orig):
- """Changes a MS smart quote character to an XML or HTML
- entity."""
- sub = self.MS_CHARS.get(orig)
- if isinstance(sub, tuple):
- if self.smartQuotesTo == 'xml':
- sub = '&#x%s;' % sub[1]
- else:
- sub = '&%s;' % sub[0]
- return sub
-
- def _convertFrom(self, proposed):
- proposed = self.find_codec(proposed)
- if not proposed or proposed in self.triedEncodings:
- return None
- self.triedEncodings.append(proposed)
- markup = self.markup
-
- # Convert smart quotes to HTML if coming from an encoding
- # that might have them.
- if self.smartQuotesTo and proposed.lower() in("windows-1252",
- "iso-8859-1",
- "iso-8859-2"):
- markup = re.compile("([\x80-\x9f])").sub \
- (lambda(x): self._subMSChar(x.group(1)),
- markup)
-
- try:
- # print "Trying to convert document to %s" % proposed
- u = self._toUnicode(markup, proposed)
- self.markup = u
- self.originalEncoding = proposed
- except Exception, e:
- # print "That didn't work!"
- # print e
- return None
- #print "Correct encoding: %s" % proposed
- return self.markup
-
- def _toUnicode(self, data, encoding):
- '''Given a string and its encoding, decodes the string into Unicode.
- %encoding is a string recognized by encodings.aliases'''
-
- # strip Byte Order Mark (if present)
- if (len(data) >= 4) and (data[:2] == '\xfe\xff') \
- and (data[2:4] != '\x00\x00'):
- encoding = 'utf-16be'
- data = data[2:]
- elif (len(data) >= 4) and (data[:2] == '\xff\xfe') \
- and (data[2:4] != '\x00\x00'):
- encoding = 'utf-16le'
- data = data[2:]
- elif data[:3] == '\xef\xbb\xbf':
- encoding = 'utf-8'
- data = data[3:]
- elif data[:4] == '\x00\x00\xfe\xff':
- encoding = 'utf-32be'
- data = data[4:]
- elif data[:4] == '\xff\xfe\x00\x00':
- encoding = 'utf-32le'
- data = data[4:]
- newdata = unicode(data, encoding)
- return newdata
-
- def _detectEncoding(self, xml_data, isHTML=False):
- """Given a document, tries to detect its XML encoding."""
- xml_encoding = sniffed_xml_encoding = None
- try:
- if xml_data[:4] == '\x4c\x6f\xa7\x94':
- # EBCDIC
- xml_data = self._ebcdic_to_ascii(xml_data)
- elif xml_data[:4] == '\x00\x3c\x00\x3f':
- # UTF-16BE
- sniffed_xml_encoding = 'utf-16be'
- xml_data = unicode(xml_data, 'utf-16be').encode('utf-8')
- elif (len(xml_data) >= 4) and (xml_data[:2] == '\xfe\xff') \
- and (xml_data[2:4] != '\x00\x00'):
- # UTF-16BE with BOM
- sniffed_xml_encoding = 'utf-16be'
- xml_data = unicode(xml_data[2:], 'utf-16be').encode('utf-8')
- elif xml_data[:4] == '\x3c\x00\x3f\x00':
- # UTF-16LE
- sniffed_xml_encoding = 'utf-16le'
- xml_data = unicode(xml_data, 'utf-16le').encode('utf-8')
- elif (len(xml_data) >= 4) and (xml_data[:2] == '\xff\xfe') and \
- (xml_data[2:4] != '\x00\x00'):
- # UTF-16LE with BOM
- sniffed_xml_encoding = 'utf-16le'
- xml_data = unicode(xml_data[2:], 'utf-16le').encode('utf-8')
- elif xml_data[:4] == '\x00\x00\x00\x3c':
- # UTF-32BE
- sniffed_xml_encoding = 'utf-32be'
- xml_data = unicode(xml_data, 'utf-32be').encode('utf-8')
- elif xml_data[:4] == '\x3c\x00\x00\x00':
- # UTF-32LE
- sniffed_xml_encoding = 'utf-32le'
- xml_data = unicode(xml_data, 'utf-32le').encode('utf-8')
- elif xml_data[:4] == '\x00\x00\xfe\xff':
- # UTF-32BE with BOM
- sniffed_xml_encoding = 'utf-32be'
- xml_data = unicode(xml_data[4:], 'utf-32be').encode('utf-8')
- elif xml_data[:4] == '\xff\xfe\x00\x00':
- # UTF-32LE with BOM
- sniffed_xml_encoding = 'utf-32le'
- xml_data = unicode(xml_data[4:], 'utf-32le').encode('utf-8')
- elif xml_data[:3] == '\xef\xbb\xbf':
- # UTF-8 with BOM
- sniffed_xml_encoding = 'utf-8'
- xml_data = unicode(xml_data[3:], 'utf-8').encode('utf-8')
- else:
- sniffed_xml_encoding = 'ascii'
- pass
- except:
- xml_encoding_match = None
- xml_encoding_match = re.compile(
- '^<\?.*encoding=[\'"](.*?)[\'"].*\?>').match(xml_data)
- if not xml_encoding_match and isHTML:
- regexp = re.compile('<\s*meta[^>]+charset=([^>]*?)[;\'">]', re.I)
- xml_encoding_match = regexp.search(xml_data)
- if xml_encoding_match is not None:
- xml_encoding = xml_encoding_match.groups()[0].lower()
- if isHTML:
- self.declaredHTMLEncoding = xml_encoding
- if sniffed_xml_encoding and \
- (xml_encoding in ('iso-10646-ucs-2', 'ucs-2', 'csunicode',
- 'iso-10646-ucs-4', 'ucs-4', 'csucs4',
- 'utf-16', 'utf-32', 'utf_16', 'utf_32',
- 'utf16', 'u16')):
- xml_encoding = sniffed_xml_encoding
- return xml_data, xml_encoding, sniffed_xml_encoding
-
-
- def find_codec(self, charset):
- return self._codec(self.CHARSET_ALIASES.get(charset, charset)) \
- or (charset and self._codec(charset.replace("-", ""))) \
- or (charset and self._codec(charset.replace("-", "_"))) \
- or charset
-
- def _codec(self, charset):
- if not charset: return charset
- codec = None
- try:
- codecs.lookup(charset)
- codec = charset
- except (LookupError, ValueError):
- pass
- return codec
-
- EBCDIC_TO_ASCII_MAP = None
- def _ebcdic_to_ascii(self, s):
- c = self.__class__
- if not c.EBCDIC_TO_ASCII_MAP:
- emap = (0,1,2,3,156,9,134,127,151,141,142,11,12,13,14,15,
- 16,17,18,19,157,133,8,135,24,25,146,143,28,29,30,31,
- 128,129,130,131,132,10,23,27,136,137,138,139,140,5,6,7,
- 144,145,22,147,148,149,150,4,152,153,154,155,20,21,158,26,
- 32,160,161,162,163,164,165,166,167,168,91,46,60,40,43,33,
- 38,169,170,171,172,173,174,175,176,177,93,36,42,41,59,94,
- 45,47,178,179,180,181,182,183,184,185,124,44,37,95,62,63,
- 186,187,188,189,190,191,192,193,194,96,58,35,64,39,61,34,
- 195,97,98,99,100,101,102,103,104,105,196,197,198,199,200,
- 201,202,106,107,108,109,110,111,112,113,114,203,204,205,
- 206,207,208,209,126,115,116,117,118,119,120,121,122,210,
- 211,212,213,214,215,216,217,218,219,220,221,222,223,224,
- 225,226,227,228,229,230,231,123,65,66,67,68,69,70,71,72,
- 73,232,233,234,235,236,237,125,74,75,76,77,78,79,80,81,
- 82,238,239,240,241,242,243,92,159,83,84,85,86,87,88,89,
- 90,244,245,246,247,248,249,48,49,50,51,52,53,54,55,56,57,
- 250,251,252,253,254,255)
- import string
- c.EBCDIC_TO_ASCII_MAP = string.maketrans( \
- ''.join(map(chr, range(256))), ''.join(map(chr, emap)))
- return s.translate(c.EBCDIC_TO_ASCII_MAP)
-
- MS_CHARS = { '\x80' : ('euro', '20AC'),
- '\x81' : ' ',
- '\x82' : ('sbquo', '201A'),
- '\x83' : ('fnof', '192'),
- '\x84' : ('bdquo', '201E'),
- '\x85' : ('hellip', '2026'),
- '\x86' : ('dagger', '2020'),
- '\x87' : ('Dagger', '2021'),
- '\x88' : ('circ', '2C6'),
- '\x89' : ('permil', '2030'),
- '\x8A' : ('Scaron', '160'),
- '\x8B' : ('lsaquo', '2039'),
- '\x8C' : ('OElig', '152'),
- '\x8D' : '?',
- '\x8E' : ('#x17D', '17D'),
- '\x8F' : '?',
- '\x90' : '?',
- '\x91' : ('lsquo', '2018'),
- '\x92' : ('rsquo', '2019'),
- '\x93' : ('ldquo', '201C'),
- '\x94' : ('rdquo', '201D'),
- '\x95' : ('bull', '2022'),
- '\x96' : ('ndash', '2013'),
- '\x97' : ('mdash', '2014'),
- '\x98' : ('tilde', '2DC'),
- '\x99' : ('trade', '2122'),
- '\x9a' : ('scaron', '161'),
- '\x9b' : ('rsaquo', '203A'),
- '\x9c' : ('oelig', '153'),
- '\x9d' : '?',
- '\x9e' : ('#x17E', '17E'),
- '\x9f' : ('Yuml', ''),}
-
-#######################################################################
-
-
-#By default, act as an HTML pretty-printer.
-if __name__ == '__main__':
- import sys
- soup = BeautifulSoup(sys.stdin)
- print soup.prettify()
+ SoupStrainer = bs4.SoupStrainer
+else:
+ from webkitpy.thirdparty.BeautifulSoup_legacy import BeautifulSoup, BeautifulStoneSoup, SoupStrainer
diff --git a/Tools/Scripts/webkitpy/thirdparty/BeautifulSoup_legacy.py b/Tools/Scripts/webkitpy/thirdparty/BeautifulSoup_legacy.py
new file mode 100644
index 0000000..4b17b85
--- /dev/null
+++ b/Tools/Scripts/webkitpy/thirdparty/BeautifulSoup_legacy.py
@@ -0,0 +1,2014 @@
+"""Beautiful Soup
+Elixir and Tonic
+"The Screen-Scraper's Friend"
+http://www.crummy.com/software/BeautifulSoup/
+
+Beautiful Soup parses a (possibly invalid) XML or HTML document into a
+tree representation. It provides methods and Pythonic idioms that make
+it easy to navigate, search, and modify the tree.
+
+A well-formed XML/HTML document yields a well-formed data
+structure. An ill-formed XML/HTML document yields a correspondingly
+ill-formed data structure. If your document is only locally
+well-formed, you can use this library to find and process the
+well-formed part of it.
+
+Beautiful Soup works with Python 2.2 and up. It has no external
+dependencies, but you'll have more success at converting data to UTF-8
+if you also install these three packages:
+
+* chardet, for auto-detecting character encodings
+ http://chardet.feedparser.org/
+* cjkcodecs and iconv_codec, which add more encodings to the ones supported
+ by stock Python.
+ http://cjkpython.i18n.org/
+
+Beautiful Soup defines classes for two main parsing strategies:
+
+ * BeautifulStoneSoup, for parsing XML, SGML, or your domain-specific
+ language that kind of looks like XML.
+
+ * BeautifulSoup, for parsing run-of-the-mill HTML code, be it valid
+ or invalid. This class has web browser-like heuristics for
+ obtaining a sensible parse tree in the face of common HTML errors.
+
+Beautiful Soup also defines a class (UnicodeDammit) for autodetecting
+the encoding of an HTML or XML document, and converting it to
+Unicode. Much of this code is taken from Mark Pilgrim's Universal Feed Parser.
+
+For more than you ever wanted to know about Beautiful Soup, see the
+documentation:
+http://www.crummy.com/software/BeautifulSoup/documentation.html
+
+Here, have some legalese:
+
+Copyright (c) 2004-2010, Leonard Richardson
+
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+ * Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+
+ * Redistributions in binary form must reproduce the above
+ copyright notice, this list of conditions and the following
+ disclaimer in the documentation and/or other materials provided
+ with the distribution.
+
+ * Neither the name of the the Beautiful Soup Consortium and All
+ Night Kosher Bakery nor the names of its contributors may be
+ used to endorse or promote products derived from this software
+ without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE, DAMMIT.
+
+"""
+from __future__ import generators
+
+__author__ = "Leonard Richardson (leonardr@segfault.org)"
+__version__ = "3.2.0"
+__copyright__ = "Copyright (c) 2004-2010 Leonard Richardson"
+__license__ = "New-style BSD"
+
+from sgmllib import SGMLParser, SGMLParseError
+import codecs
+import markupbase
+import types
+import re
+import sgmllib
+try:
+ from htmlentitydefs import name2codepoint
+except ImportError:
+ name2codepoint = {}
+try:
+ set
+except NameError:
+ from sets import Set as set
+
+#These hacks make Beautiful Soup able to parse XML with namespaces
+sgmllib.tagfind = re.compile('[a-zA-Z][-_.:a-zA-Z0-9]*')
+markupbase._declname_match = re.compile(r'[a-zA-Z][-_.:a-zA-Z0-9]*\s*').match
+
+DEFAULT_OUTPUT_ENCODING = "utf-8"
+
+def _match_css_class(str):
+ """Build a RE to match the given CSS class."""
+ return re.compile(r"(^|.*\s)%s($|\s)" % str)
+
+# First, the classes that represent markup elements.
+
+class PageElement(object):
+ """Contains the navigational information for some part of the page
+ (either a tag or a piece of text)"""
+
+ def setup(self, parent=None, previous=None):
+ """Sets up the initial relations between this element and
+ other elements."""
+ self.parent = parent
+ self.previous = previous
+ self.next = None
+ self.previousSibling = None
+ self.nextSibling = None
+ if self.parent and self.parent.contents:
+ self.previousSibling = self.parent.contents[-1]
+ self.previousSibling.nextSibling = self
+
+ def replaceWith(self, replaceWith):
+ oldParent = self.parent
+ myIndex = self.parent.index(self)
+ if hasattr(replaceWith, "parent")\
+ and replaceWith.parent is self.parent:
+ # We're replacing this element with one of its siblings.
+ index = replaceWith.parent.index(replaceWith)
+ if index and index < myIndex:
+ # Furthermore, it comes before this element. That
+ # means that when we extract it, the index of this
+ # element will change.
+ myIndex = myIndex - 1
+ self.extract()
+ oldParent.insert(myIndex, replaceWith)
+
+ def replaceWithChildren(self):
+ myParent = self.parent
+ myIndex = self.parent.index(self)
+ self.extract()
+ reversedChildren = list(self.contents)
+ reversedChildren.reverse()
+ for child in reversedChildren:
+ myParent.insert(myIndex, child)
+
+ def extract(self):
+ """Destructively rips this element out of the tree."""
+ if self.parent:
+ try:
+ del self.parent.contents[self.parent.index(self)]
+ except ValueError:
+ pass
+
+ #Find the two elements that would be next to each other if
+ #this element (and any children) hadn't been parsed. Connect
+ #the two.
+ lastChild = self._lastRecursiveChild()
+ nextElement = lastChild.next
+
+ if self.previous:
+ self.previous.next = nextElement
+ if nextElement:
+ nextElement.previous = self.previous
+ self.previous = None
+ lastChild.next = None
+
+ self.parent = None
+ if self.previousSibling:
+ self.previousSibling.nextSibling = self.nextSibling
+ if self.nextSibling:
+ self.nextSibling.previousSibling = self.previousSibling
+ self.previousSibling = self.nextSibling = None
+ return self
+
+ def _lastRecursiveChild(self):
+ "Finds the last element beneath this object to be parsed."
+ lastChild = self
+ while hasattr(lastChild, 'contents') and lastChild.contents:
+ lastChild = lastChild.contents[-1]
+ return lastChild
+
+ def insert(self, position, newChild):
+ if isinstance(newChild, basestring) \
+ and not isinstance(newChild, NavigableString):
+ newChild = NavigableString(newChild)
+
+ position = min(position, len(self.contents))
+ if hasattr(newChild, 'parent') and newChild.parent is not None:
+ # We're 'inserting' an element that's already one
+ # of this object's children.
+ if newChild.parent is self:
+ index = self.index(newChild)
+ if index > position:
+ # Furthermore we're moving it further down the
+ # list of this object's children. That means that
+ # when we extract this element, our target index
+ # will jump down one.
+ position = position - 1
+ newChild.extract()
+
+ newChild.parent = self
+ previousChild = None
+ if position == 0:
+ newChild.previousSibling = None
+ newChild.previous = self
+ else:
+ previousChild = self.contents[position-1]
+ newChild.previousSibling = previousChild
+ newChild.previousSibling.nextSibling = newChild
+ newChild.previous = previousChild._lastRecursiveChild()
+ if newChild.previous:
+ newChild.previous.next = newChild
+
+ newChildsLastElement = newChild._lastRecursiveChild()
+
+ if position >= len(self.contents):
+ newChild.nextSibling = None
+
+ parent = self
+ parentsNextSibling = None
+ while not parentsNextSibling:
+ parentsNextSibling = parent.nextSibling
+ parent = parent.parent
+ if not parent: # This is the last element in the document.
+ break
+ if parentsNextSibling:
+ newChildsLastElement.next = parentsNextSibling
+ else:
+ newChildsLastElement.next = None
+ else:
+ nextChild = self.contents[position]
+ newChild.nextSibling = nextChild
+ if newChild.nextSibling:
+ newChild.nextSibling.previousSibling = newChild
+ newChildsLastElement.next = nextChild
+
+ if newChildsLastElement.next:
+ newChildsLastElement.next.previous = newChildsLastElement
+ self.contents.insert(position, newChild)
+
+ def append(self, tag):
+ """Appends the given tag to the contents of this tag."""
+ self.insert(len(self.contents), tag)
+
+ def findNext(self, name=None, attrs={}, text=None, **kwargs):
+ """Returns the first item that matches the given criteria and
+ appears after this Tag in the document."""
+ return self._findOne(self.findAllNext, name, attrs, text, **kwargs)
+
+ def findAllNext(self, name=None, attrs={}, text=None, limit=None,
+ **kwargs):
+ """Returns all items that match the given criteria and appear
+ after this Tag in the document."""
+ return self._findAll(name, attrs, text, limit, self.nextGenerator,
+ **kwargs)
+
+ def findNextSibling(self, name=None, attrs={}, text=None, **kwargs):
+ """Returns the closest sibling to this Tag that matches the
+ given criteria and appears after this Tag in the document."""
+ return self._findOne(self.findNextSiblings, name, attrs, text,
+ **kwargs)
+
+ def findNextSiblings(self, name=None, attrs={}, text=None, limit=None,
+ **kwargs):
+ """Returns the siblings of this Tag that match the given
+ criteria and appear after this Tag in the document."""
+ return self._findAll(name, attrs, text, limit,
+ self.nextSiblingGenerator, **kwargs)
+ fetchNextSiblings = findNextSiblings # Compatibility with pre-3.x
+
+ def findPrevious(self, name=None, attrs={}, text=None, **kwargs):
+ """Returns the first item that matches the given criteria and
+ appears before this Tag in the document."""
+ return self._findOne(self.findAllPrevious, name, attrs, text, **kwargs)
+
+ def findAllPrevious(self, name=None, attrs={}, text=None, limit=None,
+ **kwargs):
+ """Returns all items that match the given criteria and appear
+ before this Tag in the document."""
+ return self._findAll(name, attrs, text, limit, self.previousGenerator,
+ **kwargs)
+ fetchPrevious = findAllPrevious # Compatibility with pre-3.x
+
+ def findPreviousSibling(self, name=None, attrs={}, text=None, **kwargs):
+ """Returns the closest sibling to this Tag that matches the
+ given criteria and appears before this Tag in the document."""
+ return self._findOne(self.findPreviousSiblings, name, attrs, text,
+ **kwargs)
+
+ def findPreviousSiblings(self, name=None, attrs={}, text=None,
+ limit=None, **kwargs):
+ """Returns the siblings of this Tag that match the given
+ criteria and appear before this Tag in the document."""
+ return self._findAll(name, attrs, text, limit,
+ self.previousSiblingGenerator, **kwargs)
+ fetchPreviousSiblings = findPreviousSiblings # Compatibility with pre-3.x
+
+ def findParent(self, name=None, attrs={}, **kwargs):
+ """Returns the closest parent of this Tag that matches the given
+ criteria."""
+ # NOTE: We can't use _findOne because findParents takes a different
+ # set of arguments.
+ r = None
+ l = self.findParents(name, attrs, 1)
+ if l:
+ r = l[0]
+ return r
+
+ def findParents(self, name=None, attrs={}, limit=None, **kwargs):
+ """Returns the parents of this Tag that match the given
+ criteria."""
+
+ return self._findAll(name, attrs, None, limit, self.parentGenerator,
+ **kwargs)
+ fetchParents = findParents # Compatibility with pre-3.x
+
+ #These methods do the real heavy lifting.
+
+ def _findOne(self, method, name, attrs, text, **kwargs):
+ r = None
+ l = method(name, attrs, text, 1, **kwargs)
+ if l:
+ r = l[0]
+ return r
+
+ def _findAll(self, name, attrs, text, limit, generator, **kwargs):
+ "Iterates over a generator looking for things that match."
+
+ if isinstance(name, SoupStrainer):
+ strainer = name
+ # (Possibly) special case some findAll*(...) searches
+ elif text is None and not limit and not attrs and not kwargs:
+ # findAll*(True)
+ if name is True:
+ return [element for element in generator()
+ if isinstance(element, Tag)]
+ # findAll*('tag-name')
+ elif isinstance(name, basestring):
+ return [element for element in generator()
+ if isinstance(element, Tag) and
+ element.name == name]
+ else:
+ strainer = SoupStrainer(name, attrs, text, **kwargs)
+ # Build a SoupStrainer
+ else:
+ strainer = SoupStrainer(name, attrs, text, **kwargs)
+ results = ResultSet(strainer)
+ g = generator()
+ while True:
+ try:
+ i = g.next()
+ except StopIteration:
+ break
+ if i:
+ found = strainer.search(i)
+ if found:
+ results.append(found)
+ if limit and len(results) >= limit:
+ break
+ return results
+
+ #These Generators can be used to navigate starting from both
+ #NavigableStrings and Tags.
+ def nextGenerator(self):
+ i = self
+ while i is not None:
+ i = i.next
+ yield i
+
+ def nextSiblingGenerator(self):
+ i = self
+ while i is not None:
+ i = i.nextSibling
+ yield i
+
+ def previousGenerator(self):
+ i = self
+ while i is not None:
+ i = i.previous
+ yield i
+
+ def previousSiblingGenerator(self):
+ i = self
+ while i is not None:
+ i = i.previousSibling
+ yield i
+
+ def parentGenerator(self):
+ i = self
+ while i is not None:
+ i = i.parent
+ yield i
+
+ # Utility methods
+ def substituteEncoding(self, str, encoding=None):
+ encoding = encoding or "utf-8"
+ return str.replace("%SOUP-ENCODING%", encoding)
+
+ def toEncoding(self, s, encoding=None):
+ """Encodes an object to a string in some encoding, or to Unicode.
+ ."""
+ if isinstance(s, unicode):
+ if encoding:
+ s = s.encode(encoding)
+ elif isinstance(s, str):
+ if encoding:
+ s = s.encode(encoding)
+ else:
+ s = unicode(s)
+ else:
+ if encoding:
+ s = self.toEncoding(str(s), encoding)
+ else:
+ s = unicode(s)
+ return s
+
+class NavigableString(unicode, PageElement):
+
+ def __new__(cls, value):
+ """Create a new NavigableString.
+
+ When unpickling a NavigableString, this method is called with
+ the string in DEFAULT_OUTPUT_ENCODING. That encoding needs to be
+ passed in to the superclass's __new__ or the superclass won't know
+ how to handle non-ASCII characters.
+ """
+ if isinstance(value, unicode):
+ return unicode.__new__(cls, value)
+ return unicode.__new__(cls, value, DEFAULT_OUTPUT_ENCODING)
+
+ def __getnewargs__(self):
+ return (NavigableString.__str__(self),)
+
+ def __getattr__(self, attr):
+ """text.string gives you text. This is for backwards
+ compatibility for Navigable*String, but for CData* it lets you
+ get the string without the CData wrapper."""
+ if attr == 'string':
+ return self
+ else:
+ raise AttributeError, "'%s' object has no attribute '%s'" % (self.__class__.__name__, attr)
+
+ def __unicode__(self):
+ return str(self).decode(DEFAULT_OUTPUT_ENCODING)
+
+ def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ if encoding:
+ return self.encode(encoding)
+ else:
+ return self
+
+class CData(NavigableString):
+
+ def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ return "<![CDATA[%s]]>" % NavigableString.__str__(self, encoding)
+
+class ProcessingInstruction(NavigableString):
+ def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ output = self
+ if "%SOUP-ENCODING%" in output:
+ output = self.substituteEncoding(output, encoding)
+ return "<?%s?>" % self.toEncoding(output, encoding)
+
+class Comment(NavigableString):
+ def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ return "<!--%s-->" % NavigableString.__str__(self, encoding)
+
+class Declaration(NavigableString):
+ def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ return "<!%s>" % NavigableString.__str__(self, encoding)
+
+class Tag(PageElement):
+
+ """Represents a found HTML tag with its attributes and contents."""
+
+ def _invert(h):
+ "Cheap function to invert a hash."
+ i = {}
+ for k,v in h.items():
+ i[v] = k
+ return i
+
+ XML_ENTITIES_TO_SPECIAL_CHARS = { "apos" : "'",
+ "quot" : '"',
+ "amp" : "&",
+ "lt" : "<",
+ "gt" : ">" }
+
+ XML_SPECIAL_CHARS_TO_ENTITIES = _invert(XML_ENTITIES_TO_SPECIAL_CHARS)
+
+ def _convertEntities(self, match):
+ """Used in a call to re.sub to replace HTML, XML, and numeric
+ entities with the appropriate Unicode characters. If HTML
+ entities are being converted, any unrecognized entities are
+ escaped."""
+ x = match.group(1)
+ if self.convertHTMLEntities and x in name2codepoint:
+ return unichr(name2codepoint[x])
+ elif x in self.XML_ENTITIES_TO_SPECIAL_CHARS:
+ if self.convertXMLEntities:
+ return self.XML_ENTITIES_TO_SPECIAL_CHARS[x]
+ else:
+ return u'&%s;' % x
+ elif len(x) > 0 and x[0] == '#':
+ # Handle numeric entities
+ if len(x) > 1 and x[1] == 'x':
+ return unichr(int(x[2:], 16))
+ else:
+ return unichr(int(x[1:]))
+
+ elif self.escapeUnrecognizedEntities:
+ return u'&%s;' % x
+ else:
+ return u'&%s;' % x
+
+ def __init__(self, parser, name, attrs=None, parent=None,
+ previous=None):
+ "Basic constructor."
+
+ # We don't actually store the parser object: that lets extracted
+ # chunks be garbage-collected
+ self.parserClass = parser.__class__
+ self.isSelfClosing = parser.isSelfClosingTag(name)
+ self.name = name
+ if attrs is None:
+ attrs = []
+ elif isinstance(attrs, dict):
+ attrs = attrs.items()
+ self.attrs = attrs
+ self.contents = []
+ self.setup(parent, previous)
+ self.hidden = False
+ self.containsSubstitutions = False
+ self.convertHTMLEntities = parser.convertHTMLEntities
+ self.convertXMLEntities = parser.convertXMLEntities
+ self.escapeUnrecognizedEntities = parser.escapeUnrecognizedEntities
+
+ # Convert any HTML, XML, or numeric entities in the attribute values.
+ convert = lambda(k, val): (k,
+ re.sub("&(#\d+|#x[0-9a-fA-F]+|\w+);",
+ self._convertEntities,
+ val))
+ self.attrs = map(convert, self.attrs)
+
+ def getString(self):
+ if (len(self.contents) == 1
+ and isinstance(self.contents[0], NavigableString)):
+ return self.contents[0]
+
+ def setString(self, string):
+ """Replace the contents of the tag with a string"""
+ self.clear()
+ self.append(string)
+
+ string = property(getString, setString)
+
+ def getText(self, separator=u""):
+ if not len(self.contents):
+ return u""
+ stopNode = self._lastRecursiveChild().next
+ strings = []
+ current = self.contents[0]
+ while current is not stopNode:
+ if isinstance(current, NavigableString):
+ strings.append(current.strip())
+ current = current.next
+ return separator.join(strings)
+
+ text = property(getText)
+
+ def get(self, key, default=None):
+ """Returns the value of the 'key' attribute for the tag, or
+ the value given for 'default' if it doesn't have that
+ attribute."""
+ return self._getAttrMap().get(key, default)
+
+ def clear(self):
+ """Extract all children."""
+ for child in self.contents[:]:
+ child.extract()
+
+ def index(self, element):
+ for i, child in enumerate(self.contents):
+ if child is element:
+ return i
+ raise ValueError("Tag.index: element not in tag")
+
+ def has_key(self, key):
+ return self._getAttrMap().has_key(key)
+
+ def __getitem__(self, key):
+ """tag[key] returns the value of the 'key' attribute for the tag,
+ and throws an exception if it's not there."""
+ return self._getAttrMap()[key]
+
+ def __iter__(self):
+ "Iterating over a tag iterates over its contents."
+ return iter(self.contents)
+
+ def __len__(self):
+ "The length of a tag is the length of its list of contents."
+ return len(self.contents)
+
+ def __contains__(self, x):
+ return x in self.contents
+
+ def __nonzero__(self):
+ "A tag is non-None even if it has no contents."
+ return True
+
+ def __setitem__(self, key, value):
+ """Setting tag[key] sets the value of the 'key' attribute for the
+ tag."""
+ self._getAttrMap()
+ self.attrMap[key] = value
+ found = False
+ for i in range(0, len(self.attrs)):
+ if self.attrs[i][0] == key:
+ self.attrs[i] = (key, value)
+ found = True
+ if not found:
+ self.attrs.append((key, value))
+ self._getAttrMap()[key] = value
+
+ def __delitem__(self, key):
+ "Deleting tag[key] deletes all 'key' attributes for the tag."
+ for item in self.attrs:
+ if item[0] == key:
+ self.attrs.remove(item)
+ #We don't break because bad HTML can define the same
+ #attribute multiple times.
+ self._getAttrMap()
+ if self.attrMap.has_key(key):
+ del self.attrMap[key]
+
+ def __call__(self, *args, **kwargs):
+ """Calling a tag like a function is the same as calling its
+ findAll() method. Eg. tag('a') returns a list of all the A tags
+ found within this tag."""
+ return apply(self.findAll, args, kwargs)
+
+ def __getattr__(self, tag):
+ #print "Getattr %s.%s" % (self.__class__, tag)
+ if len(tag) > 3 and tag.rfind('Tag') == len(tag)-3:
+ return self.find(tag[:-3])
+ elif tag.find('__') != 0:
+ return self.find(tag)
+ raise AttributeError, "'%s' object has no attribute '%s'" % (self.__class__, tag)
+
+ def __eq__(self, other):
+ """Returns true iff this tag has the same name, the same attributes,
+ and the same contents (recursively) as the given tag.
+
+ NOTE: right now this will return false if two tags have the
+ same attributes in a different order. Should this be fixed?"""
+ if other is self:
+ return True
+ if not hasattr(other, 'name') or not hasattr(other, 'attrs') or not hasattr(other, 'contents') or self.name != other.name or self.attrs != other.attrs or len(self) != len(other):
+ return False
+ for i in range(0, len(self.contents)):
+ if self.contents[i] != other.contents[i]:
+ return False
+ return True
+
+ def __ne__(self, other):
+ """Returns true iff this tag is not identical to the other tag,
+ as defined in __eq__."""
+ return not self == other
+
+ def __repr__(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ """Renders this tag as a string."""
+ return self.__str__(encoding)
+
+ def __unicode__(self):
+ return self.__str__(None)
+
+ BARE_AMPERSAND_OR_BRACKET = re.compile("([<>]|"
+ + "&(?!#\d+;|#x[0-9a-fA-F]+;|\w+;)"
+ + ")")
+
+ def _sub_entity(self, x):
+ """Used with a regular expression to substitute the
+ appropriate XML entity for an XML special character."""
+ return "&" + self.XML_SPECIAL_CHARS_TO_ENTITIES[x.group(0)[0]] + ";"
+
+ def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING,
+ prettyPrint=False, indentLevel=0):
+ """Returns a string or Unicode representation of this tag and
+ its contents. To get Unicode, pass None for encoding.
+
+ NOTE: since Python's HTML parser consumes whitespace, this
+ method is not certain to reproduce the whitespace present in
+ the original string."""
+
+ encodedName = self.toEncoding(self.name, encoding)
+
+ attrs = []
+ if self.attrs:
+ for key, val in self.attrs:
+ fmt = '%s="%s"'
+ if isinstance(val, basestring):
+ if self.containsSubstitutions and '%SOUP-ENCODING%' in val:
+ val = self.substituteEncoding(val, encoding)
+
+ # The attribute value either:
+ #
+ # * Contains no embedded double quotes or single quotes.
+ # No problem: we enclose it in double quotes.
+ # * Contains embedded single quotes. No problem:
+ # double quotes work here too.
+ # * Contains embedded double quotes. No problem:
+ # we enclose it in single quotes.
+ # * Embeds both single _and_ double quotes. This
+ # can't happen naturally, but it can happen if
+ # you modify an attribute value after parsing
+ # the document. Now we have a bit of a
+ # problem. We solve it by enclosing the
+ # attribute in single quotes, and escaping any
+ # embedded single quotes to XML entities.
+ if '"' in val:
+ fmt = "%s='%s'"
+ if "'" in val:
+ # TODO: replace with apos when
+ # appropriate.
+ val = val.replace("'", "&squot;")
+
+ # Now we're okay w/r/t quotes. But the attribute
+ # value might also contain angle brackets, or
+ # ampersands that aren't part of entities. We need
+ # to escape those to XML entities too.
+ val = self.BARE_AMPERSAND_OR_BRACKET.sub(self._sub_entity, val)
+
+ attrs.append(fmt % (self.toEncoding(key, encoding),
+ self.toEncoding(val, encoding)))
+ close = ''
+ closeTag = ''
+ if self.isSelfClosing:
+ close = ' /'
+ else:
+ closeTag = '</%s>' % encodedName
+
+ indentTag, indentContents = 0, 0
+ if prettyPrint:
+ indentTag = indentLevel
+ space = (' ' * (indentTag-1))
+ indentContents = indentTag + 1
+ contents = self.renderContents(encoding, prettyPrint, indentContents)
+ if self.hidden:
+ s = contents
+ else:
+ s = []
+ attributeString = ''
+ if attrs:
+ attributeString = ' ' + ' '.join(attrs)
+ if prettyPrint:
+ s.append(space)
+ s.append('<%s%s%s>' % (encodedName, attributeString, close))
+ if prettyPrint:
+ s.append("\n")
+ s.append(contents)
+ if prettyPrint and contents and contents[-1] != "\n":
+ s.append("\n")
+ if prettyPrint and closeTag:
+ s.append(space)
+ s.append(closeTag)
+ if prettyPrint and closeTag and self.nextSibling:
+ s.append("\n")
+ s = ''.join(s)
+ return s
+
+ def decompose(self):
+ """Recursively destroys the contents of this tree."""
+ self.extract()
+ if len(self.contents) == 0:
+ return
+ current = self.contents[0]
+ while current is not None:
+ next = current.next
+ if isinstance(current, Tag):
+ del current.contents[:]
+ current.parent = None
+ current.previous = None
+ current.previousSibling = None
+ current.next = None
+ current.nextSibling = None
+ current = next
+
+ def prettify(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ return self.__str__(encoding, True)
+
+ def renderContents(self, encoding=DEFAULT_OUTPUT_ENCODING,
+ prettyPrint=False, indentLevel=0):
+ """Renders the contents of this tag as a string in the given
+ encoding. If encoding is None, returns a Unicode string.."""
+ s=[]
+ for c in self:
+ text = None
+ if isinstance(c, NavigableString):
+ text = c.__str__(encoding)
+ elif isinstance(c, Tag):
+ s.append(c.__str__(encoding, prettyPrint, indentLevel))
+ if text and prettyPrint:
+ text = text.strip()
+ if text:
+ if prettyPrint:
+ s.append(" " * (indentLevel-1))
+ s.append(text)
+ if prettyPrint:
+ s.append("\n")
+ return ''.join(s)
+
+ #Soup methods
+
+ def find(self, name=None, attrs={}, recursive=True, text=None,
+ **kwargs):
+ """Return only the first child of this Tag matching the given
+ criteria."""
+ r = None
+ l = self.findAll(name, attrs, recursive, text, 1, **kwargs)
+ if l:
+ r = l[0]
+ return r
+ findChild = find
+
+ def findAll(self, name=None, attrs={}, recursive=True, text=None,
+ limit=None, **kwargs):
+ """Extracts a list of Tag objects that match the given
+ criteria. You can specify the name of the Tag and any
+ attributes you want the Tag to have.
+
+ The value of a key-value pair in the 'attrs' map can be a
+ string, a list of strings, a regular expression object, or a
+ callable that takes a string and returns whether or not the
+ string matches for some custom definition of 'matches'. The
+ same is true of the tag name."""
+ generator = self.recursiveChildGenerator
+ if not recursive:
+ generator = self.childGenerator
+ return self._findAll(name, attrs, text, limit, generator, **kwargs)
+ findChildren = findAll
+
+ # Pre-3.x compatibility methods
+ first = find
+ fetch = findAll
+
+ def fetchText(self, text=None, recursive=True, limit=None):
+ return self.findAll(text=text, recursive=recursive, limit=limit)
+
+ def firstText(self, text=None, recursive=True):
+ return self.find(text=text, recursive=recursive)
+
+ #Private methods
+
+ def _getAttrMap(self):
+ """Initializes a map representation of this tag's attributes,
+ if not already initialized."""
+ if not getattr(self, 'attrMap'):
+ self.attrMap = {}
+ for (key, value) in self.attrs:
+ self.attrMap[key] = value
+ return self.attrMap
+
+ #Generator methods
+ def childGenerator(self):
+ # Just use the iterator from the contents
+ return iter(self.contents)
+
+ def recursiveChildGenerator(self):
+ if not len(self.contents):
+ raise StopIteration
+ stopNode = self._lastRecursiveChild().next
+ current = self.contents[0]
+ while current is not stopNode:
+ yield current
+ current = current.next
+
+
+# Next, a couple classes to represent queries and their results.
+class SoupStrainer:
+ """Encapsulates a number of ways of matching a markup element (tag or
+ text)."""
+
+ def __init__(self, name=None, attrs={}, text=None, **kwargs):
+ self.name = name
+ if isinstance(attrs, basestring):
+ kwargs['class'] = _match_css_class(attrs)
+ attrs = None
+ if kwargs:
+ if attrs:
+ attrs = attrs.copy()
+ attrs.update(kwargs)
+ else:
+ attrs = kwargs
+ self.attrs = attrs
+ self.text = text
+
+ def __str__(self):
+ if self.text:
+ return self.text
+ else:
+ return "%s|%s" % (self.name, self.attrs)
+
+ def searchTag(self, markupName=None, markupAttrs={}):
+ found = None
+ markup = None
+ if isinstance(markupName, Tag):
+ markup = markupName
+ markupAttrs = markup
+ callFunctionWithTagData = callable(self.name) \
+ and not isinstance(markupName, Tag)
+
+ if (not self.name) \
+ or callFunctionWithTagData \
+ or (markup and self._matches(markup, self.name)) \
+ or (not markup and self._matches(markupName, self.name)):
+ if callFunctionWithTagData:
+ match = self.name(markupName, markupAttrs)
+ else:
+ match = True
+ markupAttrMap = None
+ for attr, matchAgainst in self.attrs.items():
+ if not markupAttrMap:
+ if hasattr(markupAttrs, 'get'):
+ markupAttrMap = markupAttrs
+ else:
+ markupAttrMap = {}
+ for k,v in markupAttrs:
+ markupAttrMap[k] = v
+ attrValue = markupAttrMap.get(attr)
+ if not self._matches(attrValue, matchAgainst):
+ match = False
+ break
+ if match:
+ if markup:
+ found = markup
+ else:
+ found = markupName
+ return found
+
+ def search(self, markup):
+ #print 'looking for %s in %s' % (self, markup)
+ found = None
+ # If given a list of items, scan it for a text element that
+ # matches.
+ if hasattr(markup, "__iter__") \
+ and not isinstance(markup, Tag):
+ for element in markup:
+ if isinstance(element, NavigableString) \
+ and self.search(element):
+ found = element
+ break
+ # If it's a Tag, make sure its name or attributes match.
+ # Don't bother with Tags if we're searching for text.
+ elif isinstance(markup, Tag):
+ if not self.text:
+ found = self.searchTag(markup)
+ # If it's text, make sure the text matches.
+ elif isinstance(markup, NavigableString) or \
+ isinstance(markup, basestring):
+ if self._matches(markup, self.text):
+ found = markup
+ else:
+ raise Exception, "I don't know how to match against a %s" \
+ % markup.__class__
+ return found
+
+ def _matches(self, markup, matchAgainst):
+ #print "Matching %s against %s" % (markup, matchAgainst)
+ result = False
+ if matchAgainst is True:
+ result = markup is not None
+ elif callable(matchAgainst):
+ result = matchAgainst(markup)
+ else:
+ #Custom match methods take the tag as an argument, but all
+ #other ways of matching match the tag name as a string.
+ if isinstance(markup, Tag):
+ markup = markup.name
+ if markup and not isinstance(markup, basestring):
+ markup = unicode(markup)
+ #Now we know that chunk is either a string, or None.
+ if hasattr(matchAgainst, 'match'):
+ # It's a regexp object.
+ result = markup and matchAgainst.search(markup)
+ elif hasattr(matchAgainst, '__iter__'): # list-like
+ result = markup in matchAgainst
+ elif hasattr(matchAgainst, 'items'):
+ result = markup.has_key(matchAgainst)
+ elif matchAgainst and isinstance(markup, basestring):
+ if isinstance(markup, unicode):
+ matchAgainst = unicode(matchAgainst)
+ else:
+ matchAgainst = str(matchAgainst)
+
+ if not result:
+ result = matchAgainst == markup
+ return result
+
+class ResultSet(list):
+ """A ResultSet is just a list that keeps track of the SoupStrainer
+ that created it."""
+ def __init__(self, source):
+ list.__init__([])
+ self.source = source
+
+# Now, some helper functions.
+
+def buildTagMap(default, *args):
+ """Turns a list of maps, lists, or scalars into a single map.
+ Used to build the SELF_CLOSING_TAGS, NESTABLE_TAGS, and
+ NESTING_RESET_TAGS maps out of lists and partial maps."""
+ built = {}
+ for portion in args:
+ if hasattr(portion, 'items'):
+ #It's a map. Merge it.
+ for k,v in portion.items():
+ built[k] = v
+ elif hasattr(portion, '__iter__'): # is a list
+ #It's a list. Map each item to the default.
+ for k in portion:
+ built[k] = default
+ else:
+ #It's a scalar. Map it to the default.
+ built[portion] = default
+ return built
+
+# Now, the parser classes.
+
+class BeautifulStoneSoup(Tag, SGMLParser):
+
+ """This class contains the basic parser and search code. It defines
+ a parser that knows nothing about tag behavior except for the
+ following:
+
+ You can't close a tag without closing all the tags it encloses.
+ That is, "<foo><bar></foo>" actually means
+ "<foo><bar></bar></foo>".
+
+ [Another possible explanation is "<foo><bar /></foo>", but since
+ this class defines no SELF_CLOSING_TAGS, it will never use that
+ explanation.]
+
+ This class is useful for parsing XML or made-up markup languages,
+ or when BeautifulSoup makes an assumption counter to what you were
+ expecting."""
+
+ SELF_CLOSING_TAGS = {}
+ NESTABLE_TAGS = {}
+ RESET_NESTING_TAGS = {}
+ QUOTE_TAGS = {}
+ PRESERVE_WHITESPACE_TAGS = []
+
+ MARKUP_MASSAGE = [(re.compile('(<[^<>]*)/>'),
+ lambda x: x.group(1) + ' />'),
+ (re.compile('<!\s+([^<>]*)>'),
+ lambda x: '<!' + x.group(1) + '>')
+ ]
+
+ ROOT_TAG_NAME = u'[document]'
+
+ HTML_ENTITIES = "html"
+ XML_ENTITIES = "xml"
+ XHTML_ENTITIES = "xhtml"
+ # TODO: This only exists for backwards-compatibility
+ ALL_ENTITIES = XHTML_ENTITIES
+
+ # Used when determining whether a text node is all whitespace and
+ # can be replaced with a single space. A text node that contains
+ # fancy Unicode spaces (usually non-breaking) should be left
+ # alone.
+ STRIP_ASCII_SPACES = { 9: None, 10: None, 12: None, 13: None, 32: None, }
+
+ def __init__(self, markup="", parseOnlyThese=None, fromEncoding=None,
+ markupMassage=True, smartQuotesTo=XML_ENTITIES,
+ convertEntities=None, selfClosingTags=None, isHTML=False):
+ """The Soup object is initialized as the 'root tag', and the
+ provided markup (which can be a string or a file-like object)
+ is fed into the underlying parser.
+
+ sgmllib will process most bad HTML, and the BeautifulSoup
+ class has some tricks for dealing with some HTML that kills
+ sgmllib, but Beautiful Soup can nonetheless choke or lose data
+ if your data uses self-closing tags or declarations
+ incorrectly.
+
+ By default, Beautiful Soup uses regexes to sanitize input,
+ avoiding the vast majority of these problems. If the problems
+ don't apply to you, pass in False for markupMassage, and
+ you'll get better performance.
+
+ The default parser massage techniques fix the two most common
+ instances of invalid HTML that choke sgmllib:
+
+ <br/> (No space between name of closing tag and tag close)
+ <! --Comment--> (Extraneous whitespace in declaration)
+
+ You can pass in a custom list of (RE object, replace method)
+ tuples to get Beautiful Soup to scrub your input the way you
+ want."""
+
+ self.parseOnlyThese = parseOnlyThese
+ self.fromEncoding = fromEncoding
+ self.smartQuotesTo = smartQuotesTo
+ self.convertEntities = convertEntities
+ # Set the rules for how we'll deal with the entities we
+ # encounter
+ if self.convertEntities:
+ # It doesn't make sense to convert encoded characters to
+ # entities even while you're converting entities to Unicode.
+ # Just convert it all to Unicode.
+ self.smartQuotesTo = None
+ if convertEntities == self.HTML_ENTITIES:
+ self.convertXMLEntities = False
+ self.convertHTMLEntities = True
+ self.escapeUnrecognizedEntities = True
+ elif convertEntities == self.XHTML_ENTITIES:
+ self.convertXMLEntities = True
+ self.convertHTMLEntities = True
+ self.escapeUnrecognizedEntities = False
+ elif convertEntities == self.XML_ENTITIES:
+ self.convertXMLEntities = True
+ self.convertHTMLEntities = False
+ self.escapeUnrecognizedEntities = False
+ else:
+ self.convertXMLEntities = False
+ self.convertHTMLEntities = False
+ self.escapeUnrecognizedEntities = False
+
+ self.instanceSelfClosingTags = buildTagMap(None, selfClosingTags)
+ SGMLParser.__init__(self)
+
+ if hasattr(markup, 'read'): # It's a file-type object.
+ markup = markup.read()
+ self.markup = markup
+ self.markupMassage = markupMassage
+ try:
+ self._feed(isHTML=isHTML)
+ except StopParsing:
+ pass
+ self.markup = None # The markup can now be GCed
+
+ def convert_charref(self, name):
+ """This method fixes a bug in Python's SGMLParser."""
+ try:
+ n = int(name)
+ except ValueError:
+ return
+ if not 0 <= n <= 127 : # ASCII ends at 127, not 255
+ return
+ return self.convert_codepoint(n)
+
+ def _feed(self, inDocumentEncoding=None, isHTML=False):
+ # Convert the document to Unicode.
+ markup = self.markup
+ if isinstance(markup, unicode):
+ if not hasattr(self, 'originalEncoding'):
+ self.originalEncoding = None
+ else:
+ dammit = UnicodeDammit\
+ (markup, [self.fromEncoding, inDocumentEncoding],
+ smartQuotesTo=self.smartQuotesTo, isHTML=isHTML)
+ markup = dammit.unicode
+ self.originalEncoding = dammit.originalEncoding
+ self.declaredHTMLEncoding = dammit.declaredHTMLEncoding
+ if markup:
+ if self.markupMassage:
+ if not hasattr(self.markupMassage, "__iter__"):
+ self.markupMassage = self.MARKUP_MASSAGE
+ for fix, m in self.markupMassage:
+ markup = fix.sub(m, markup)
+ # TODO: We get rid of markupMassage so that the
+ # soup object can be deepcopied later on. Some
+ # Python installations can't copy regexes. If anyone
+ # was relying on the existence of markupMassage, this
+ # might cause problems.
+ del(self.markupMassage)
+ self.reset()
+
+ SGMLParser.feed(self, markup)
+ # Close out any unfinished strings and close all the open tags.
+ self.endData()
+ while self.currentTag.name != self.ROOT_TAG_NAME:
+ self.popTag()
+
+ def __getattr__(self, methodName):
+ """This method routes method call requests to either the SGMLParser
+ superclass or the Tag superclass, depending on the method name."""
+ #print "__getattr__ called on %s.%s" % (self.__class__, methodName)
+
+ if methodName.startswith('start_') or methodName.startswith('end_') \
+ or methodName.startswith('do_'):
+ return SGMLParser.__getattr__(self, methodName)
+ elif not methodName.startswith('__'):
+ return Tag.__getattr__(self, methodName)
+ else:
+ raise AttributeError
+
+ def isSelfClosingTag(self, name):
+ """Returns true iff the given string is the name of a
+ self-closing tag according to this parser."""
+ return self.SELF_CLOSING_TAGS.has_key(name) \
+ or self.instanceSelfClosingTags.has_key(name)
+
+ def reset(self):
+ Tag.__init__(self, self, self.ROOT_TAG_NAME)
+ self.hidden = 1
+ SGMLParser.reset(self)
+ self.currentData = []
+ self.currentTag = None
+ self.tagStack = []
+ self.quoteStack = []
+ self.pushTag(self)
+
+ def popTag(self):
+ tag = self.tagStack.pop()
+
+ #print "Pop", tag.name
+ if self.tagStack:
+ self.currentTag = self.tagStack[-1]
+ return self.currentTag
+
+ def pushTag(self, tag):
+ #print "Push", tag.name
+ if self.currentTag:
+ self.currentTag.contents.append(tag)
+ self.tagStack.append(tag)
+ self.currentTag = self.tagStack[-1]
+
+ def endData(self, containerClass=NavigableString):
+ if self.currentData:
+ currentData = u''.join(self.currentData)
+ if (currentData.translate(self.STRIP_ASCII_SPACES) == '' and
+ not set([tag.name for tag in self.tagStack]).intersection(
+ self.PRESERVE_WHITESPACE_TAGS)):
+ if '\n' in currentData:
+ currentData = '\n'
+ else:
+ currentData = ' '
+ self.currentData = []
+ if self.parseOnlyThese and len(self.tagStack) <= 1 and \
+ (not self.parseOnlyThese.text or \
+ not self.parseOnlyThese.search(currentData)):
+ return
+ o = containerClass(currentData)
+ o.setup(self.currentTag, self.previous)
+ if self.previous:
+ self.previous.next = o
+ self.previous = o
+ self.currentTag.contents.append(o)
+
+
+ def _popToTag(self, name, inclusivePop=True):
+ """Pops the tag stack up to and including the most recent
+ instance of the given tag. If inclusivePop is false, pops the tag
+ stack up to but *not* including the most recent instqance of
+ the given tag."""
+ #print "Popping to %s" % name
+ if name == self.ROOT_TAG_NAME:
+ return
+
+ numPops = 0
+ mostRecentTag = None
+ for i in range(len(self.tagStack)-1, 0, -1):
+ if name == self.tagStack[i].name:
+ numPops = len(self.tagStack)-i
+ break
+ if not inclusivePop:
+ numPops = numPops - 1
+
+ for i in range(0, numPops):
+ mostRecentTag = self.popTag()
+ return mostRecentTag
+
+ def _smartPop(self, name):
+
+ """We need to pop up to the previous tag of this type, unless
+ one of this tag's nesting reset triggers comes between this
+ tag and the previous tag of this type, OR unless this tag is a
+ generic nesting trigger and another generic nesting trigger
+ comes between this tag and the previous tag of this type.
+
+ Examples:
+ <p>Foo<b>Bar *<p>* should pop to 'p', not 'b'.
+ <p>Foo<table>Bar *<p>* should pop to 'table', not 'p'.
+ <p>Foo<table><tr>Bar *<p>* should pop to 'tr', not 'p'.
+
+ <li><ul><li> *<li>* should pop to 'ul', not the first 'li'.
+ <tr><table><tr> *<tr>* should pop to 'table', not the first 'tr'
+ <td><tr><td> *<td>* should pop to 'tr', not the first 'td'
+ """
+
+ nestingResetTriggers = self.NESTABLE_TAGS.get(name)
+ isNestable = nestingResetTriggers != None
+ isResetNesting = self.RESET_NESTING_TAGS.has_key(name)
+ popTo = None
+ inclusive = True
+ for i in range(len(self.tagStack)-1, 0, -1):
+ p = self.tagStack[i]
+ if (not p or p.name == name) and not isNestable:
+ #Non-nestable tags get popped to the top or to their
+ #last occurance.
+ popTo = name
+ break
+ if (nestingResetTriggers is not None
+ and p.name in nestingResetTriggers) \
+ or (nestingResetTriggers is None and isResetNesting
+ and self.RESET_NESTING_TAGS.has_key(p.name)):
+
+ #If we encounter one of the nesting reset triggers
+ #peculiar to this tag, or we encounter another tag
+ #that causes nesting to reset, pop up to but not
+ #including that tag.
+ popTo = p.name
+ inclusive = False
+ break
+ p = p.parent
+ if popTo:
+ self._popToTag(popTo, inclusive)
+
+ def unknown_starttag(self, name, attrs, selfClosing=0):
+ #print "Start tag %s: %s" % (name, attrs)
+ if self.quoteStack:
+ #This is not a real tag.
+ #print "<%s> is not real!" % name
+ attrs = ''.join([' %s="%s"' % (x, y) for x, y in attrs])
+ self.handle_data('<%s%s>' % (name, attrs))
+ return
+ self.endData()
+
+ if not self.isSelfClosingTag(name) and not selfClosing:
+ self._smartPop(name)
+
+ if self.parseOnlyThese and len(self.tagStack) <= 1 \
+ and (self.parseOnlyThese.text or not self.parseOnlyThese.searchTag(name, attrs)):
+ return
+
+ tag = Tag(self, name, attrs, self.currentTag, self.previous)
+ if self.previous:
+ self.previous.next = tag
+ self.previous = tag
+ self.pushTag(tag)
+ if selfClosing or self.isSelfClosingTag(name):
+ self.popTag()
+ if name in self.QUOTE_TAGS:
+ #print "Beginning quote (%s)" % name
+ self.quoteStack.append(name)
+ self.literal = 1
+ return tag
+
+ def unknown_endtag(self, name):
+ #print "End tag %s" % name
+ if self.quoteStack and self.quoteStack[-1] != name:
+ #This is not a real end tag.
+ #print "</%s> is not real!" % name
+ self.handle_data('</%s>' % name)
+ return
+ self.endData()
+ self._popToTag(name)
+ if self.quoteStack and self.quoteStack[-1] == name:
+ self.quoteStack.pop()
+ self.literal = (len(self.quoteStack) > 0)
+
+ def handle_data(self, data):
+ self.currentData.append(data)
+
+ def _toStringSubclass(self, text, subclass):
+ """Adds a certain piece of text to the tree as a NavigableString
+ subclass."""
+ self.endData()
+ self.handle_data(text)
+ self.endData(subclass)
+
+ def handle_pi(self, text):
+ """Handle a processing instruction as a ProcessingInstruction
+ object, possibly one with a %SOUP-ENCODING% slot into which an
+ encoding will be plugged later."""
+ if text[:3] == "xml":
+ text = u"xml version='1.0' encoding='%SOUP-ENCODING%'"
+ self._toStringSubclass(text, ProcessingInstruction)
+
+ def handle_comment(self, text):
+ "Handle comments as Comment objects."
+ self._toStringSubclass(text, Comment)
+
+ def handle_charref(self, ref):
+ "Handle character references as data."
+ if self.convertEntities:
+ data = unichr(int(ref))
+ else:
+ data = '&#%s;' % ref
+ self.handle_data(data)
+
+ def handle_entityref(self, ref):
+ """Handle entity references as data, possibly converting known
+ HTML and/or XML entity references to the corresponding Unicode
+ characters."""
+ data = None
+ if self.convertHTMLEntities:
+ try:
+ data = unichr(name2codepoint[ref])
+ except KeyError:
+ pass
+
+ if not data and self.convertXMLEntities:
+ data = self.XML_ENTITIES_TO_SPECIAL_CHARS.get(ref)
+
+ if not data and self.convertHTMLEntities and \
+ not self.XML_ENTITIES_TO_SPECIAL_CHARS.get(ref):
+ # TODO: We've got a problem here. We're told this is
+ # an entity reference, but it's not an XML entity
+ # reference or an HTML entity reference. Nonetheless,
+ # the logical thing to do is to pass it through as an
+ # unrecognized entity reference.
+ #
+ # Except: when the input is "&carol;" this function
+ # will be called with input "carol". When the input is
+ # "AT&T", this function will be called with input
+ # "T". We have no way of knowing whether a semicolon
+ # was present originally, so we don't know whether
+ # this is an unknown entity or just a misplaced
+ # ampersand.
+ #
+ # The more common case is a misplaced ampersand, so I
+ # escape the ampersand and omit the trailing semicolon.
+ data = "&%s" % ref
+ if not data:
+ # This case is different from the one above, because we
+ # haven't already gone through a supposedly comprehensive
+ # mapping of entities to Unicode characters. We might not
+ # have gone through any mapping at all. So the chances are
+ # very high that this is a real entity, and not a
+ # misplaced ampersand.
+ data = "&%s;" % ref
+ self.handle_data(data)
+
+ def handle_decl(self, data):
+ "Handle DOCTYPEs and the like as Declaration objects."
+ self._toStringSubclass(data, Declaration)
+
+ def parse_declaration(self, i):
+ """Treat a bogus SGML declaration as raw data. Treat a CDATA
+ declaration as a CData object."""
+ j = None
+ if self.rawdata[i:i+9] == '<![CDATA[':
+ k = self.rawdata.find(']]>', i)
+ if k == -1:
+ k = len(self.rawdata)
+ data = self.rawdata[i+9:k]
+ j = k+3
+ self._toStringSubclass(data, CData)
+ else:
+ try:
+ j = SGMLParser.parse_declaration(self, i)
+ except SGMLParseError:
+ toHandle = self.rawdata[i:]
+ self.handle_data(toHandle)
+ j = i + len(toHandle)
+ return j
+
+class BeautifulSoup(BeautifulStoneSoup):
+
+ """This parser knows the following facts about HTML:
+
+ * Some tags have no closing tag and should be interpreted as being
+ closed as soon as they are encountered.
+
+ * The text inside some tags (ie. 'script') may contain tags which
+ are not really part of the document and which should be parsed
+ as text, not tags. If you want to parse the text as tags, you can
+ always fetch it and parse it explicitly.
+
+ * Tag nesting rules:
+
+ Most tags can't be nested at all. For instance, the occurance of
+ a <p> tag should implicitly close the previous <p> tag.
+
+ <p>Para1<p>Para2
+ should be transformed into:
+ <p>Para1</p><p>Para2
+
+ Some tags can be nested arbitrarily. For instance, the occurance
+ of a <blockquote> tag should _not_ implicitly close the previous
+ <blockquote> tag.
+
+ Alice said: <blockquote>Bob said: <blockquote>Blah
+ should NOT be transformed into:
+ Alice said: <blockquote>Bob said: </blockquote><blockquote>Blah
+
+ Some tags can be nested, but the nesting is reset by the
+ interposition of other tags. For instance, a <tr> tag should
+ implicitly close the previous <tr> tag within the same <table>,
+ but not close a <tr> tag in another table.
+
+ <table><tr>Blah<tr>Blah
+ should be transformed into:
+ <table><tr>Blah</tr><tr>Blah
+ but,
+ <tr>Blah<table><tr>Blah
+ should NOT be transformed into
+ <tr>Blah<table></tr><tr>Blah
+
+ Differing assumptions about tag nesting rules are a major source
+ of problems with the BeautifulSoup class. If BeautifulSoup is not
+ treating as nestable a tag your page author treats as nestable,
+ try ICantBelieveItsBeautifulSoup, MinimalSoup, or
+ BeautifulStoneSoup before writing your own subclass."""
+
+ def __init__(self, *args, **kwargs):
+ if not kwargs.has_key('smartQuotesTo'):
+ kwargs['smartQuotesTo'] = self.HTML_ENTITIES
+ kwargs['isHTML'] = True
+ BeautifulStoneSoup.__init__(self, *args, **kwargs)
+
+ SELF_CLOSING_TAGS = buildTagMap(None,
+ ('br' , 'hr', 'input', 'img', 'meta',
+ 'spacer', 'link', 'frame', 'base', 'col'))
+
+ PRESERVE_WHITESPACE_TAGS = set(['pre', 'textarea'])
+
+ QUOTE_TAGS = {'script' : None, 'textarea' : None}
+
+ #According to the HTML standard, each of these inline tags can
+ #contain another tag of the same type. Furthermore, it's common
+ #to actually use these tags this way.
+ NESTABLE_INLINE_TAGS = ('span', 'font', 'q', 'object', 'bdo', 'sub', 'sup',
+ 'center')
+
+ #According to the HTML standard, these block tags can contain
+ #another tag of the same type. Furthermore, it's common
+ #to actually use these tags this way.
+ NESTABLE_BLOCK_TAGS = ('blockquote', 'div', 'fieldset', 'ins', 'del')
+
+ #Lists can contain other lists, but there are restrictions.
+ NESTABLE_LIST_TAGS = { 'ol' : [],
+ 'ul' : [],
+ 'li' : ['ul', 'ol'],
+ 'dl' : [],
+ 'dd' : ['dl'],
+ 'dt' : ['dl'] }
+
+ #Tables can contain other tables, but there are restrictions.
+ NESTABLE_TABLE_TAGS = {'table' : [],
+ 'tr' : ['table', 'tbody', 'tfoot', 'thead'],
+ 'td' : ['tr'],
+ 'th' : ['tr'],
+ 'thead' : ['table'],
+ 'tbody' : ['table'],
+ 'tfoot' : ['table'],
+ }
+
+ NON_NESTABLE_BLOCK_TAGS = ('address', 'form', 'p', 'pre')
+
+ #If one of these tags is encountered, all tags up to the next tag of
+ #this type are popped.
+ RESET_NESTING_TAGS = buildTagMap(None, NESTABLE_BLOCK_TAGS, 'noscript',
+ NON_NESTABLE_BLOCK_TAGS,
+ NESTABLE_LIST_TAGS,
+ NESTABLE_TABLE_TAGS)
+
+ NESTABLE_TAGS = buildTagMap([], NESTABLE_INLINE_TAGS, NESTABLE_BLOCK_TAGS,
+ NESTABLE_LIST_TAGS, NESTABLE_TABLE_TAGS)
+
+ # Used to detect the charset in a META tag; see start_meta
+ CHARSET_RE = re.compile("((^|;)\s*charset=)([^;]*)", re.M)
+
+ def start_meta(self, attrs):
+ """Beautiful Soup can detect a charset included in a META tag,
+ try to convert the document to that charset, and re-parse the
+ document from the beginning."""
+ httpEquiv = None
+ contentType = None
+ contentTypeIndex = None
+ tagNeedsEncodingSubstitution = False
+
+ for i in range(0, len(attrs)):
+ key, value = attrs[i]
+ key = key.lower()
+ if key == 'http-equiv':
+ httpEquiv = value
+ elif key == 'content':
+ contentType = value
+ contentTypeIndex = i
+
+ if httpEquiv and contentType: # It's an interesting meta tag.
+ match = self.CHARSET_RE.search(contentType)
+ if match:
+ if (self.declaredHTMLEncoding is not None or
+ self.originalEncoding == self.fromEncoding):
+ # An HTML encoding was sniffed while converting
+ # the document to Unicode, or an HTML encoding was
+ # sniffed during a previous pass through the
+ # document, or an encoding was specified
+ # explicitly and it worked. Rewrite the meta tag.
+ def rewrite(match):
+ return match.group(1) + "%SOUP-ENCODING%"
+ newAttr = self.CHARSET_RE.sub(rewrite, contentType)
+ attrs[contentTypeIndex] = (attrs[contentTypeIndex][0],
+ newAttr)
+ tagNeedsEncodingSubstitution = True
+ else:
+ # This is our first pass through the document.
+ # Go through it again with the encoding information.
+ newCharset = match.group(3)
+ if newCharset and newCharset != self.originalEncoding:
+ self.declaredHTMLEncoding = newCharset
+ self._feed(self.declaredHTMLEncoding)
+ raise StopParsing
+ pass
+ tag = self.unknown_starttag("meta", attrs)
+ if tag and tagNeedsEncodingSubstitution:
+ tag.containsSubstitutions = True
+
+class StopParsing(Exception):
+ pass
+
+class ICantBelieveItsBeautifulSoup(BeautifulSoup):
+
+ """The BeautifulSoup class is oriented towards skipping over
+ common HTML errors like unclosed tags. However, sometimes it makes
+ errors of its own. For instance, consider this fragment:
+
+ <b>Foo<b>Bar</b></b>
+
+ This is perfectly valid (if bizarre) HTML. However, the
+ BeautifulSoup class will implicitly close the first b tag when it
+ encounters the second 'b'. It will think the author wrote
+ "<b>Foo<b>Bar", and didn't close the first 'b' tag, because
+ there's no real-world reason to bold something that's already
+ bold. When it encounters '</b></b>' it will close two more 'b'
+ tags, for a grand total of three tags closed instead of two. This
+ can throw off the rest of your document structure. The same is
+ true of a number of other tags, listed below.
+
+ It's much more common for someone to forget to close a 'b' tag
+ than to actually use nested 'b' tags, and the BeautifulSoup class
+ handles the common case. This class handles the not-co-common
+ case: where you can't believe someone wrote what they did, but
+ it's valid HTML and BeautifulSoup screwed up by assuming it
+ wouldn't be."""
+
+ I_CANT_BELIEVE_THEYRE_NESTABLE_INLINE_TAGS = \
+ ('em', 'big', 'i', 'small', 'tt', 'abbr', 'acronym', 'strong',
+ 'cite', 'code', 'dfn', 'kbd', 'samp', 'strong', 'var', 'b',
+ 'big')
+
+ I_CANT_BELIEVE_THEYRE_NESTABLE_BLOCK_TAGS = ('noscript',)
+
+ NESTABLE_TAGS = buildTagMap([], BeautifulSoup.NESTABLE_TAGS,
+ I_CANT_BELIEVE_THEYRE_NESTABLE_BLOCK_TAGS,
+ I_CANT_BELIEVE_THEYRE_NESTABLE_INLINE_TAGS)
+
+class MinimalSoup(BeautifulSoup):
+ """The MinimalSoup class is for parsing HTML that contains
+ pathologically bad markup. It makes no assumptions about tag
+ nesting, but it does know which tags are self-closing, that
+ <script> tags contain Javascript and should not be parsed, that
+ META tags may contain encoding information, and so on.
+
+ This also makes it better for subclassing than BeautifulStoneSoup
+ or BeautifulSoup."""
+
+ RESET_NESTING_TAGS = buildTagMap('noscript')
+ NESTABLE_TAGS = {}
+
+class BeautifulSOAP(BeautifulStoneSoup):
+ """This class will push a tag with only a single string child into
+ the tag's parent as an attribute. The attribute's name is the tag
+ name, and the value is the string child. An example should give
+ the flavor of the change:
+
+ <foo><bar>baz</bar></foo>
+ =>
+ <foo bar="baz"><bar>baz</bar></foo>
+
+ You can then access fooTag['bar'] instead of fooTag.barTag.string.
+
+ This is, of course, useful for scraping structures that tend to
+ use subelements instead of attributes, such as SOAP messages. Note
+ that it modifies its input, so don't print the modified version
+ out.
+
+ I'm not sure how many people really want to use this class; let me
+ know if you do. Mainly I like the name."""
+
+ def popTag(self):
+ if len(self.tagStack) > 1:
+ tag = self.tagStack[-1]
+ parent = self.tagStack[-2]
+ parent._getAttrMap()
+ if (isinstance(tag, Tag) and len(tag.contents) == 1 and
+ isinstance(tag.contents[0], NavigableString) and
+ not parent.attrMap.has_key(tag.name)):
+ parent[tag.name] = tag.contents[0]
+ BeautifulStoneSoup.popTag(self)
+
+#Enterprise class names! It has come to our attention that some people
+#think the names of the Beautiful Soup parser classes are too silly
+#and "unprofessional" for use in enterprise screen-scraping. We feel
+#your pain! For such-minded folk, the Beautiful Soup Consortium And
+#All-Night Kosher Bakery recommends renaming this file to
+#"RobustParser.py" (or, in cases of extreme enterprisiness,
+#"RobustParserBeanInterface.class") and using the following
+#enterprise-friendly class aliases:
+class RobustXMLParser(BeautifulStoneSoup):
+ pass
+class RobustHTMLParser(BeautifulSoup):
+ pass
+class RobustWackAssHTMLParser(ICantBelieveItsBeautifulSoup):
+ pass
+class RobustInsanelyWackAssHTMLParser(MinimalSoup):
+ pass
+class SimplifyingSOAPParser(BeautifulSOAP):
+ pass
+
+######################################################
+#
+# Bonus library: Unicode, Dammit
+#
+# This class forces XML data into a standard format (usually to UTF-8
+# or Unicode). It is heavily based on code from Mark Pilgrim's
+# Universal Feed Parser. It does not rewrite the XML or HTML to
+# reflect a new encoding: that happens in BeautifulStoneSoup.handle_pi
+# (XML) and BeautifulSoup.start_meta (HTML).
+
+# Autodetects character encodings.
+# Download from http://chardet.feedparser.org/
+try:
+ import chardet
+# import chardet.constants
+# chardet.constants._debug = 1
+except ImportError:
+ chardet = None
+
+# cjkcodecs and iconv_codec make Python know about more character encodings.
+# Both are available from http://cjkpython.i18n.org/
+# They're built in if you use Python 2.4.
+try:
+ import cjkcodecs.aliases
+except ImportError:
+ pass
+try:
+ import iconv_codec
+except ImportError:
+ pass
+
+class UnicodeDammit:
+ """A class for detecting the encoding of a *ML document and
+ converting it to a Unicode string. If the source encoding is
+ windows-1252, can replace MS smart quotes with their HTML or XML
+ equivalents."""
+
+ # This dictionary maps commonly seen values for "charset" in HTML
+ # meta tags to the corresponding Python codec names. It only covers
+ # values that aren't in Python's aliases and can't be determined
+ # by the heuristics in find_codec.
+ CHARSET_ALIASES = { "macintosh" : "mac-roman",
+ "x-sjis" : "shift-jis" }
+
+ def __init__(self, markup, overrideEncodings=[],
+ smartQuotesTo='xml', isHTML=False):
+ self.declaredHTMLEncoding = None
+ self.markup, documentEncoding, sniffedEncoding = \
+ self._detectEncoding(markup, isHTML)
+ self.smartQuotesTo = smartQuotesTo
+ self.triedEncodings = []
+ if markup == '' or isinstance(markup, unicode):
+ self.originalEncoding = None
+ self.unicode = unicode(markup)
+ return
+
+ u = None
+ for proposedEncoding in overrideEncodings:
+ u = self._convertFrom(proposedEncoding)
+ if u: break
+ if not u:
+ for proposedEncoding in (documentEncoding, sniffedEncoding):
+ u = self._convertFrom(proposedEncoding)
+ if u: break
+
+ # If no luck and we have auto-detection library, try that:
+ if not u and chardet and not isinstance(self.markup, unicode):
+ u = self._convertFrom(chardet.detect(self.markup)['encoding'])
+
+ # As a last resort, try utf-8 and windows-1252:
+ if not u:
+ for proposed_encoding in ("utf-8", "windows-1252"):
+ u = self._convertFrom(proposed_encoding)
+ if u: break
+
+ self.unicode = u
+ if not u: self.originalEncoding = None
+
+ def _subMSChar(self, orig):
+ """Changes a MS smart quote character to an XML or HTML
+ entity."""
+ sub = self.MS_CHARS.get(orig)
+ if isinstance(sub, tuple):
+ if self.smartQuotesTo == 'xml':
+ sub = '&#x%s;' % sub[1]
+ else:
+ sub = '&%s;' % sub[0]
+ return sub
+
+ def _convertFrom(self, proposed):
+ proposed = self.find_codec(proposed)
+ if not proposed or proposed in self.triedEncodings:
+ return None
+ self.triedEncodings.append(proposed)
+ markup = self.markup
+
+ # Convert smart quotes to HTML if coming from an encoding
+ # that might have them.
+ if self.smartQuotesTo and proposed.lower() in("windows-1252",
+ "iso-8859-1",
+ "iso-8859-2"):
+ markup = re.compile("([\x80-\x9f])").sub \
+ (lambda(x): self._subMSChar(x.group(1)),
+ markup)
+
+ try:
+ # print "Trying to convert document to %s" % proposed
+ u = self._toUnicode(markup, proposed)
+ self.markup = u
+ self.originalEncoding = proposed
+ except Exception, e:
+ # print "That didn't work!"
+ # print e
+ return None
+ #print "Correct encoding: %s" % proposed
+ return self.markup
+
+ def _toUnicode(self, data, encoding):
+ '''Given a string and its encoding, decodes the string into Unicode.
+ %encoding is a string recognized by encodings.aliases'''
+
+ # strip Byte Order Mark (if present)
+ if (len(data) >= 4) and (data[:2] == '\xfe\xff') \
+ and (data[2:4] != '\x00\x00'):
+ encoding = 'utf-16be'
+ data = data[2:]
+ elif (len(data) >= 4) and (data[:2] == '\xff\xfe') \
+ and (data[2:4] != '\x00\x00'):
+ encoding = 'utf-16le'
+ data = data[2:]
+ elif data[:3] == '\xef\xbb\xbf':
+ encoding = 'utf-8'
+ data = data[3:]
+ elif data[:4] == '\x00\x00\xfe\xff':
+ encoding = 'utf-32be'
+ data = data[4:]
+ elif data[:4] == '\xff\xfe\x00\x00':
+ encoding = 'utf-32le'
+ data = data[4:]
+ newdata = unicode(data, encoding)
+ return newdata
+
+ def _detectEncoding(self, xml_data, isHTML=False):
+ """Given a document, tries to detect its XML encoding."""
+ xml_encoding = sniffed_xml_encoding = None
+ try:
+ if xml_data[:4] == '\x4c\x6f\xa7\x94':
+ # EBCDIC
+ xml_data = self._ebcdic_to_ascii(xml_data)
+ elif xml_data[:4] == '\x00\x3c\x00\x3f':
+ # UTF-16BE
+ sniffed_xml_encoding = 'utf-16be'
+ xml_data = unicode(xml_data, 'utf-16be').encode('utf-8')
+ elif (len(xml_data) >= 4) and (xml_data[:2] == '\xfe\xff') \
+ and (xml_data[2:4] != '\x00\x00'):
+ # UTF-16BE with BOM
+ sniffed_xml_encoding = 'utf-16be'
+ xml_data = unicode(xml_data[2:], 'utf-16be').encode('utf-8')
+ elif xml_data[:4] == '\x3c\x00\x3f\x00':
+ # UTF-16LE
+ sniffed_xml_encoding = 'utf-16le'
+ xml_data = unicode(xml_data, 'utf-16le').encode('utf-8')
+ elif (len(xml_data) >= 4) and (xml_data[:2] == '\xff\xfe') and \
+ (xml_data[2:4] != '\x00\x00'):
+ # UTF-16LE with BOM
+ sniffed_xml_encoding = 'utf-16le'
+ xml_data = unicode(xml_data[2:], 'utf-16le').encode('utf-8')
+ elif xml_data[:4] == '\x00\x00\x00\x3c':
+ # UTF-32BE
+ sniffed_xml_encoding = 'utf-32be'
+ xml_data = unicode(xml_data, 'utf-32be').encode('utf-8')
+ elif xml_data[:4] == '\x3c\x00\x00\x00':
+ # UTF-32LE
+ sniffed_xml_encoding = 'utf-32le'
+ xml_data = unicode(xml_data, 'utf-32le').encode('utf-8')
+ elif xml_data[:4] == '\x00\x00\xfe\xff':
+ # UTF-32BE with BOM
+ sniffed_xml_encoding = 'utf-32be'
+ xml_data = unicode(xml_data[4:], 'utf-32be').encode('utf-8')
+ elif xml_data[:4] == '\xff\xfe\x00\x00':
+ # UTF-32LE with BOM
+ sniffed_xml_encoding = 'utf-32le'
+ xml_data = unicode(xml_data[4:], 'utf-32le').encode('utf-8')
+ elif xml_data[:3] == '\xef\xbb\xbf':
+ # UTF-8 with BOM
+ sniffed_xml_encoding = 'utf-8'
+ xml_data = unicode(xml_data[3:], 'utf-8').encode('utf-8')
+ else:
+ sniffed_xml_encoding = 'ascii'
+ pass
+ except:
+ xml_encoding_match = None
+ xml_encoding_match = re.compile(
+ '^<\?.*encoding=[\'"](.*?)[\'"].*\?>').match(xml_data)
+ if not xml_encoding_match and isHTML:
+ regexp = re.compile('<\s*meta[^>]+charset=([^>]*?)[;\'">]', re.I)
+ xml_encoding_match = regexp.search(xml_data)
+ if xml_encoding_match is not None:
+ xml_encoding = xml_encoding_match.groups()[0].lower()
+ if isHTML:
+ self.declaredHTMLEncoding = xml_encoding
+ if sniffed_xml_encoding and \
+ (xml_encoding in ('iso-10646-ucs-2', 'ucs-2', 'csunicode',
+ 'iso-10646-ucs-4', 'ucs-4', 'csucs4',
+ 'utf-16', 'utf-32', 'utf_16', 'utf_32',
+ 'utf16', 'u16')):
+ xml_encoding = sniffed_xml_encoding
+ return xml_data, xml_encoding, sniffed_xml_encoding
+
+
+ def find_codec(self, charset):
+ return self._codec(self.CHARSET_ALIASES.get(charset, charset)) \
+ or (charset and self._codec(charset.replace("-", ""))) \
+ or (charset and self._codec(charset.replace("-", "_"))) \
+ or charset
+
+ def _codec(self, charset):
+ if not charset: return charset
+ codec = None
+ try:
+ codecs.lookup(charset)
+ codec = charset
+ except (LookupError, ValueError):
+ pass
+ return codec
+
+ EBCDIC_TO_ASCII_MAP = None
+ def _ebcdic_to_ascii(self, s):
+ c = self.__class__
+ if not c.EBCDIC_TO_ASCII_MAP:
+ emap = (0,1,2,3,156,9,134,127,151,141,142,11,12,13,14,15,
+ 16,17,18,19,157,133,8,135,24,25,146,143,28,29,30,31,
+ 128,129,130,131,132,10,23,27,136,137,138,139,140,5,6,7,
+ 144,145,22,147,148,149,150,4,152,153,154,155,20,21,158,26,
+ 32,160,161,162,163,164,165,166,167,168,91,46,60,40,43,33,
+ 38,169,170,171,172,173,174,175,176,177,93,36,42,41,59,94,
+ 45,47,178,179,180,181,182,183,184,185,124,44,37,95,62,63,
+ 186,187,188,189,190,191,192,193,194,96,58,35,64,39,61,34,
+ 195,97,98,99,100,101,102,103,104,105,196,197,198,199,200,
+ 201,202,106,107,108,109,110,111,112,113,114,203,204,205,
+ 206,207,208,209,126,115,116,117,118,119,120,121,122,210,
+ 211,212,213,214,215,216,217,218,219,220,221,222,223,224,
+ 225,226,227,228,229,230,231,123,65,66,67,68,69,70,71,72,
+ 73,232,233,234,235,236,237,125,74,75,76,77,78,79,80,81,
+ 82,238,239,240,241,242,243,92,159,83,84,85,86,87,88,89,
+ 90,244,245,246,247,248,249,48,49,50,51,52,53,54,55,56,57,
+ 250,251,252,253,254,255)
+ import string
+ c.EBCDIC_TO_ASCII_MAP = string.maketrans( \
+ ''.join(map(chr, range(256))), ''.join(map(chr, emap)))
+ return s.translate(c.EBCDIC_TO_ASCII_MAP)
+
+ MS_CHARS = { '\x80' : ('euro', '20AC'),
+ '\x81' : ' ',
+ '\x82' : ('sbquo', '201A'),
+ '\x83' : ('fnof', '192'),
+ '\x84' : ('bdquo', '201E'),
+ '\x85' : ('hellip', '2026'),
+ '\x86' : ('dagger', '2020'),
+ '\x87' : ('Dagger', '2021'),
+ '\x88' : ('circ', '2C6'),
+ '\x89' : ('permil', '2030'),
+ '\x8A' : ('Scaron', '160'),
+ '\x8B' : ('lsaquo', '2039'),
+ '\x8C' : ('OElig', '152'),
+ '\x8D' : '?',
+ '\x8E' : ('#x17D', '17D'),
+ '\x8F' : '?',
+ '\x90' : '?',
+ '\x91' : ('lsquo', '2018'),
+ '\x92' : ('rsquo', '2019'),
+ '\x93' : ('ldquo', '201C'),
+ '\x94' : ('rdquo', '201D'),
+ '\x95' : ('bull', '2022'),
+ '\x96' : ('ndash', '2013'),
+ '\x97' : ('mdash', '2014'),
+ '\x98' : ('tilde', '2DC'),
+ '\x99' : ('trade', '2122'),
+ '\x9a' : ('scaron', '161'),
+ '\x9b' : ('rsaquo', '203A'),
+ '\x9c' : ('oelig', '153'),
+ '\x9d' : '?',
+ '\x9e' : ('#x17E', '17E'),
+ '\x9f' : ('Yuml', ''),}
+
+#######################################################################
+
+
+#By default, act as an HTML pretty-printer.
+if __name__ == '__main__':
+ import sys
+ soup = BeautifulSoup(sys.stdin)
+ print soup.prettify()
diff --git a/Tools/Scripts/webkitpy/thirdparty/__init__.py b/Tools/Scripts/webkitpy/thirdparty/__init__.py
index b26e9c0..44f0f8d 100644
--- a/Tools/Scripts/webkitpy/thirdparty/__init__.py
+++ b/Tools/Scripts/webkitpy/thirdparty/__init__.py
@@ -42,6 +42,7 @@
from distutils import spawn
from webkitpy.common.system.autoinstall import AutoInstaller
from webkitpy.common.system.filesystem import FileSystem
+from webkitpy.common.system.executive import Executive
_THIRDPARTY_DIR = os.path.dirname(__file__)
_AUTOINSTALLED_DIR = os.path.join(_THIRDPARTY_DIR, "autoinstalled")
@@ -77,8 +78,9 @@
class AutoinstallImportHook(object):
- def __init__(self, filesystem=None):
+ def __init__(self, filesystem=None, executive=None):
self._fs = filesystem or FileSystem()
+ self._executive = executive or Executive()
def _ensure_autoinstalled_dir_is_in_sys_path(self):
# Some packages require that the are being put somewhere under a directory in sys.path.
@@ -124,6 +126,8 @@
self._install_pytest()
elif '.requests' in fullname:
self._install_requests()
+ elif '.bs4' in fullname:
+ self._install_beautifulsoup()
def _install_six(self):
self._install("https://files.pythonhosted.org/packages/16/d8/bc6316cf98419719bd59c91742194c111b6f2e85abac88e496adefaf7afe/six-1.11.0.tar.gz",
@@ -196,6 +200,20 @@
self._install("https://files.pythonhosted.org/packages/01/62/ddcf76d1d19885e8579acb1b1df26a852b03472c0e46d2b959a714c90608/requests-2.22.0.tar.gz",
"requests-2.22.0/requests")
+ def _install_beautifulsoup(self):
+ if sys.version_info < (3, 0):
+ return
+
+ self._install_requests()
+ self._ensure_autoinstalled_dir_is_in_sys_path()
+ self._install("https://files.pythonhosted.org/packages/7f/4e/95a13527e18b6f1a15c93f1c634b86d5fa634c5619dce695f4e0cd68182f/soupsieve-1.9.4.tar.gz",
+ "soupsieve-1.9.4/soupsieve")
+ did_download_bs4 = self._install("https://files.pythonhosted.org/packages/86/cd/495c68f0536dcd25f016e006731ba7be72e072280305ec52590012c1e6f2/beautifulsoup4-4.8.1.tar.gz",
+ "beautifulsoup4-4.8.1/bs4")
+ if did_download_bs4:
+ self._executive.run_command(['2to3', '-w', self._fs.join(_AUTOINSTALLED_DIR, 'bs4')])
+
+
def _install_pylint(self):
self._ensure_autoinstalled_dir_is_in_sys_path()
if (not self._fs.exists(self._fs.join(_AUTOINSTALLED_DIR, "pylint")) or
@@ -291,7 +309,7 @@
def _install(self, url, url_subpath=None, target_name=None):
installer = AutoInstaller(target_dir=_AUTOINSTALLED_DIR)
- installer.install(url=url, url_subpath=url_subpath, target_name=target_name)
+ return installer.install(url=url, url_subpath=url_subpath, target_name=target_name)
def get_latest_pypi_url(self, package_name, url_subpath_format='{name}-{version}/{lname}'):
json_url = "https://pypi.org/pypi/%s/json" % package_name
diff --git a/Tools/Scripts/webkitpy/thirdparty/mock.py b/Tools/Scripts/webkitpy/thirdparty/mock.py
index 015c19e..d578574 100644
--- a/Tools/Scripts/webkitpy/thirdparty/mock.py
+++ b/Tools/Scripts/webkitpy/thirdparty/mock.py
@@ -118,7 +118,7 @@
self.call_count = 0
self.call_args_list = []
self.method_calls = []
- for child in self._children.itervalues():
+ for child in self._children.values():
child.reset_mock()
if isinstance(self._return_value, Mock):
self._return_value.reset_mock()
