Auto merge of #24435 - marmeladema:issue-23607/compat, r=jdm

Issue 23607: first pass of changes for compatibility with Python3

As much as i want to migrate entirely to Python3 (see #23607), it will require some time as changes in web-platform-tests are significant and rely on upstream fixes to be merged and synced downstream.
In the meantime, lets improve compatibility with Python3 so that later, migration will be less painful.

Build system is definitely not ready yet for Python3, but its a step in the right direction.

---
<!-- Thank you for contributing to Servo! Please replace each `[ ]` by `[X]` when the step is complete, and replace `___` with appropriate data: -->
- [x ] `./mach build -d` does not report any errors
- [ x] `./mach test-tidy` does not report any errors

<!-- Also, please make sure that "Allow edits from maintainers" checkbox is checked, so that we can help you if you get stuck somewhere along the way.-->

<!-- Pull requests that do not address these steps are welcome, but they will require additional verification as part of the review process. -->

12 files changed, 69 insertions, 56 deletions

diff --git a/python/requirements.txt b/python/requirements.txt
index 7316d105814..438d97e3ddb 100644
--- a/python/requirements.txt
+++ b/python/requirements.txt
@@ -32,4 +32,7 @@ boto3 == 1.4.4
 # https://github.com/servo/servo/pull/18942
 certifi
 
+# For Python3 compatibility
+six
+
 -e python/tidy
diff --git a/python/servo/bootstrap.py b/python/servo/bootstrap.py
index 30c009922d6..e3ccff6f953 100644
--- a/python/servo/bootstrap.py
+++ b/python/servo/bootstrap.py
@@ -11,7 +11,7 @@ import os
 import platform
 import shutil
 import subprocess
-import urllib
+import six.moves.urllib as urllib
 from subprocess import PIPE
 from zipfile import BadZipfile
 
@@ -293,7 +293,7 @@ def windows_msvc(context, force=False):
 
     def prepare_file(zip_path, full_spec):
         if not os.path.isfile(zip_path):
-            zip_url = "{}{}.zip".format(deps_url, urllib.quote(full_spec))
+            zip_url = "{}{}.zip".format(deps_url, urllib.parse.quote(full_spec))
             download_file(full_spec, zip_url, zip_path)
 
         print("Extracting {}...".format(full_spec), end='')
diff --git a/python/servo/bootstrap_commands.py b/python/servo/bootstrap_commands.py
index 0e7b7d68f35..835c7bf21c2 100644
--- a/python/servo/bootstrap_commands.py
+++ b/python/servo/bootstrap_commands.py
@@ -18,7 +18,7 @@ import re
 import subprocess
 import sys
 import traceback
-import urllib2
+import six.moves.urllib as urllib
 import glob
 
 from mach.decorators import (
@@ -220,7 +220,7 @@ class MachCommands(CommandBase):
 
         try:
             content_base64 = download_bytes("Chromium HSTS preload list", chromium_hsts_url)
-        except urllib2.URLError:
+        except urllib.error.URLError:
             print("Unable to download chromium HSTS preload list; are you connected to the internet?")
             sys.exit(1)
 
@@ -244,7 +244,7 @@ class MachCommands(CommandBase):
 
             with open(path.join(preload_path, preload_filename), 'w') as fd:
                 json.dump(entries, fd, indent=4)
-        except ValueError, e:
+        except ValueError as e:
             print("Unable to parse chromium HSTS preload list, has the format changed?")
             sys.exit(1)
 
@@ -258,7 +258,7 @@ class MachCommands(CommandBase):
 
         try:
             content = download_bytes("Public suffix list", list_url)
-        except urllib2.URLError:
+        except urllib.error.URLError:
             print("Unable to download the public suffix list; are you connected to the internet?")
             sys.exit(1)
 
diff --git a/python/servo/build_commands.py b/python/servo/build_commands.py
index 7481663d944..7c4eda9950a 100644
--- a/python/servo/build_commands.py
+++ b/python/servo/build_commands.py
@@ -16,7 +16,7 @@ import platform
 import shutil
 import subprocess
 import sys
-import urllib
+import six.moves.urllib as urllib
 import zipfile
 import stat
 
@@ -498,7 +498,7 @@ class MachCommands(CommandBase):
                 print("Downloading GStreamer dependencies")
                 gst_url = "https://servo-deps.s3.amazonaws.com/gstreamer/%s" % gst_lib_zip
                 print(gst_url)
-                urllib.urlretrieve(gst_url, gst_lib_zip)
+                urllib.request.urlretrieve(gst_url, gst_lib_zip)
                 zip_ref = zipfile.ZipFile(gst_lib_zip, "r")
                 zip_ref.extractall(gst_dir)
                 os.remove(gst_lib_zip)
diff --git a/python/servo/command_base.py b/python/servo/command_base.py
index 7f4e941a4c7..64d6ca242f7 100644
--- a/python/servo/command_base.py
+++ b/python/servo/command_base.py
@@ -7,6 +7,8 @@
 # option. This file may not be copied, modified, or distributed
 # except according to those terms.
 
+from __future__ import print_function
+
 from errno import ENOENT as NO_SUCH_FILE_OR_DIRECTORY
 from glob import glob
 import shutil
@@ -25,7 +27,7 @@ import tarfile
 import zipfile
 from xml.etree.ElementTree import XML
 from servo.util import download_file
-import urllib2
+import six.moves.urllib as urllib
 from bootstrap import check_gstreamer_lib
 
 from mach.decorators import CommandArgument
@@ -105,7 +107,7 @@ def archive_deterministically(dir_to_archive, dest_archive, prepend_path=None):
         # packaging (in case of exceptional situations like running out of disk space).
         # TODO do this in a temporary folder after #11983 is fixed
         temp_file = '{}.temp~'.format(dest_archive)
-        with os.fdopen(os.open(temp_file, os.O_WRONLY | os.O_CREAT, 0644), 'w') as out_file:
+        with os.fdopen(os.open(temp_file, os.O_WRONLY | os.O_CREAT, 0o644), 'w') as out_file:
             if dest_archive.endswith('.zip'):
                 with zipfile.ZipFile(temp_file, 'w', zipfile.ZIP_DEFLATED) as zip_file:
                     for entry in file_list:
@@ -350,15 +352,15 @@ class CommandBase(object):
                 version_line = subprocess.check_output(["rustup" + BIN_SUFFIX, "--version"])
             except OSError as e:
                 if e.errno == NO_SUCH_FILE_OR_DIRECTORY:
-                    print "It looks like rustup is not installed. See instructions at " \
-                          "https://github.com/servo/servo/#setting-up-your-environment"
-                    print
+                    print("It looks like rustup is not installed. See instructions at "
+                          "https://github.com/servo/servo/#setting-up-your-environment")
+                    print()
                     return 1
                 raise
             version = tuple(map(int, re.match("rustup (\d+)\.(\d+)\.(\d+)", version_line).groups()))
             if version < (1, 11, 0):
-                print "rustup is at version %s.%s.%s, Servo requires 1.11.0 or more recent." % version
-                print "Try running 'rustup self update'."
+                print("rustup is at version %s.%s.%s, Servo requires 1.11.0 or more recent." % version)
+                print("Try running 'rustup self update'.")
                 return 1
             toolchain = self.toolchain()
             if platform.system() == "Windows":
@@ -504,15 +506,15 @@ class CommandBase(object):
         nightly_date = nightly_date.strip()
         # Fetch the filename to download from the build list
         repository_index = NIGHTLY_REPOSITORY_URL + "?list-type=2&prefix=nightly"
-        req = urllib2.Request(
+        req = urllib.request.Request(
             "{}/{}/{}".format(repository_index, os_prefix, nightly_date))
         try:
-            response = urllib2.urlopen(req).read()
+            response = urllib.request.urlopen(req).read()
             tree = XML(response)
             namespaces = {'ns': tree.tag[1:tree.tag.index('}')]}
             file_to_download = tree.find('ns:Contents', namespaces).find(
                 'ns:Key', namespaces).text
-        except urllib2.URLError as e:
+        except urllib.error.URLError as e:
             print("Could not fetch the available nightly versions from the repository : {}".format(
                 e.reason))
             sys.exit(1)
diff --git a/python/servo/devenv_commands.py b/python/servo/devenv_commands.py
index ad5a02c270d..266d2cfd6a5 100644
--- a/python/servo/devenv_commands.py
+++ b/python/servo/devenv_commands.py
@@ -14,7 +14,7 @@ from time import time
 import signal
 import sys
 import tempfile
-import urllib2
+import six.moves.urllib as urllib
 import json
 import subprocess
 
@@ -201,7 +201,7 @@ class MachCommands(CommandBase):
              category='devenv')
     def rustup(self):
         url = get_static_rust_lang_org_dist() + "/channel-rust-nightly-date.txt"
-        nightly_date = urllib2.urlopen(url, **get_urlopen_kwargs()).read()
+        nightly_date = urllib.request.urlopen(url, **get_urlopen_kwargs()).read()
         toolchain = "nightly-" + nightly_date
         filename = path.join(self.context.topdir, "rust-toolchain")
         with open(filename, "w") as f:
diff --git a/python/servo/lints/wpt_lint.py b/python/servo/lints/wpt_lint.py
index a1c7e528c77..e2e44c729ae 100644
--- a/python/servo/lints/wpt_lint.py
+++ b/python/servo/lints/wpt_lint.py
@@ -7,6 +7,8 @@
 # option. This file may not be copied, modified, or distributed
 # except according to those terms.
 
+from __future__ import print_function
+
 import os
 import sys
 
@@ -20,7 +22,7 @@ class Lint(LintRunner):
     def _get_wpt_files(self, suite):
         working_dir = os.path.join(WPT_PATH, suite, '')
         file_iter = self.get_files(working_dir, exclude_dirs=[])
-        print '\nRunning the WPT lint on %s...' % working_dir
+        print('\nRunning the WPT lint on %s...' % working_dir)
         for f in file_iter:
             if filter_file(f):
                 yield f[len(working_dir):]
diff --git a/python/servo/mutation/mutator.py b/python/servo/mutation/mutator.py
index f92b388e487..8018ddc6116 100644
--- a/python/servo/mutation/mutator.py
+++ b/python/servo/mutation/mutator.py
@@ -7,6 +7,8 @@
 # option. This file may not be copied, modified, or distributed
 # except according to those terms.
 
+from __future__ import print_function
+
 import fileinput
 import re
 import random
@@ -28,7 +30,7 @@ def init_variables(if_blocks):
 def deleteStatements(file_name, line_numbers):
     for line in fileinput.input(file_name, inplace=True):
         if fileinput.lineno() not in line_numbers:
-            print line.rstrip()
+            print(line.rstrip())
 
 
 class Strategy:
@@ -48,7 +50,7 @@ class Strategy:
             for line in fileinput.input(file_name, inplace=True):
                 if fileinput.lineno() == mutation_line_number:
                     line = re.sub(self._replace_strategy['regex'], self._replace_strategy['replaceString'], line)
-                print line.rstrip()
+                print(line.rstrip())
             return mutation_line_number
 
 
diff --git a/python/servo/package_commands.py b/python/servo/package_commands.py
index c487a8c290d..40bbd61b92c 100644
--- a/python/servo/package_commands.py
+++ b/python/servo/package_commands.py
@@ -19,7 +19,7 @@ import shutil
 import subprocess
 import sys
 import tempfile
-import urllib
+import six.moves.urllib as urllib
 
 from mach.decorators import (
     CommandArgument,
@@ -594,7 +594,7 @@ class PackageCommands(CommandBase):
                 "/secrets/v1/secret/project/servo/" +
                 name
             )
-            return json.load(urllib.urlopen(url))["secret"]
+            return json.load(urllib.request.urlopen(url))["secret"]
 
         def get_s3_secret():
             aws_access_key = None
diff --git a/python/servo/testing_commands.py b/python/servo/testing_commands.py
index 2863c314426..1f7651f814b 100644
--- a/python/servo/testing_commands.py
+++ b/python/servo/testing_commands.py
@@ -18,10 +18,11 @@ import copy
 from collections import OrderedDict
 import time
 import json
-import urllib2
+import six.moves.urllib as urllib
 import base64
 import shutil
 import subprocess
+from six import iteritems
 
 from mach.registrar import Registrar
 from mach.decorators import (
@@ -59,7 +60,7 @@ TEST_SUITES = OrderedDict([
               "include_arg": "test_name"}),
 ])
 
-TEST_SUITES_BY_PREFIX = {path: k for k, v in TEST_SUITES.iteritems() if "paths" in v for path in v["paths"]}
+TEST_SUITES_BY_PREFIX = {path: k for k, v in iteritems(TEST_SUITES) if "paths" in v for path in v["paths"]}
 
 
 def create_parser_wpt():
@@ -158,7 +159,7 @@ class MachCommands(CommandBase):
                 return 1
 
         test_start = time.time()
-        for suite, tests in selected_suites.iteritems():
+        for suite, tests in iteritems(selected_suites):
             props = suites[suite]
             kwargs = props.get("kwargs", {})
             if tests:
@@ -174,7 +175,7 @@ class MachCommands(CommandBase):
     def suite_for_path(self, path_arg):
         if os.path.exists(path.abspath(path_arg)):
             abs_path = path.abspath(path_arg)
-            for prefix, suite in TEST_SUITES_BY_PREFIX.iteritems():
+            for prefix, suite in iteritems(TEST_SUITES_BY_PREFIX):
                 if abs_path.startswith(prefix):
                     return suite
         return None
@@ -510,9 +511,9 @@ class MachCommands(CommandBase):
                 elif tracker_api.endswith('/'):
                     tracker_api = tracker_api[0:-1]
 
-                query = urllib2.quote(failure['test'], safe='')
-                request = urllib2.Request("%s/query.py?name=%s" % (tracker_api, query))
-                search = urllib2.urlopen(request)
+                query = urllib.parse.quote(failure['test'], safe='')
+                request = urllib.request.Request("%s/query.py?name=%s" % (tracker_api, query))
+                search = urllib.request.urlopen(request)
                 data = json.load(search)
                 if len(data) == 0:
                     actual_failures += [failure]
@@ -521,11 +522,11 @@ class MachCommands(CommandBase):
             else:
                 qstr = "repo:servo/servo+label:I-intermittent+type:issue+state:open+%s" % failure['test']
                 # we want `/` to get quoted, but not `+` (github's API doesn't like that), so we set `safe` to `+`
-                query = urllib2.quote(qstr, safe='+')
-                request = urllib2.Request("https://api.github.com/search/issues?q=%s" % query)
+                query = urllib.parse.quote(qstr, safe='+')
+                request = urllib.request.Request("https://api.github.com/search/issues?q=%s" % query)
                 if encoded_auth:
                     request.add_header("Authorization", "Basic %s" % encoded_auth)
-                search = urllib2.urlopen(request)
+                search = urllib.request.urlopen(request)
                 data = json.load(search)
                 if data['total_count'] == 0:
                     actual_failures += [failure]
diff --git a/python/servo/util.py b/python/servo/util.py
index 818240d39e6..38127093ae8 100644
--- a/python/servo/util.py
+++ b/python/servo/util.py
@@ -16,11 +16,11 @@ import platform
 import shutil
 from socket import error as socket_error
 import stat
-import StringIO
+from io import BytesIO
 import sys
 import time
 import zipfile
-import urllib2
+import six.moves.urllib as urllib
 
 
 try:
@@ -101,10 +101,10 @@ def download(desc, src, writer, start_byte=0):
     dumb = (os.environ.get("TERM") == "dumb") or (not sys.stdout.isatty())
 
     try:
-        req = urllib2.Request(src)
+        req = urllib.request.Request(src)
         if start_byte:
-            req = urllib2.Request(src, headers={'Range': 'bytes={}-'.format(start_byte)})
-        resp = urllib2.urlopen(req, **get_urlopen_kwargs())
+            req = urllib.request.Request(src, headers={'Range': 'bytes={}-'.format(start_byte)})
+        resp = urllib.request.urlopen(req, **get_urlopen_kwargs())
 
         fsize = None
         if resp.info().getheader('Content-Length'):
@@ -136,16 +136,16 @@ def download(desc, src, writer, start_byte=0):
 
         if not dumb:
             print()
-    except urllib2.HTTPError, e:
+    except urllib.error.HTTPError as e:
         print("Download failed ({}): {} - {}".format(e.code, e.reason, src))
         if e.code == 403:
             print("No Rust compiler binary available for this platform. "
                   "Please see https://github.com/servo/servo/#prerequisites")
         sys.exit(1)
-    except urllib2.URLError, e:
+    except urllib.error.URLError as e:
         print("Error downloading {}: {}. The failing URL was: {}".format(desc, e.reason, src))
         sys.exit(1)
-    except socket_error, e:
+    except socket_error as e:
         print("Looks like there's a connectivity issue, check your Internet connection. {}".format(e))
         sys.exit(1)
     except KeyboardInterrupt:
@@ -154,7 +154,7 @@ def download(desc, src, writer, start_byte=0):
 
 
 def download_bytes(desc, src):
-    content_writer = StringIO.StringIO()
+    content_writer = BytesIO()
     download(desc, src, content_writer)
     return content_writer.getvalue()
 
diff --git a/python/tidy/servo_tidy/tidy.py b/python/tidy/servo_tidy/tidy.py
index 0069f344fa7..2c9985bca34 100644
--- a/python/tidy/servo_tidy/tidy.py
+++ b/python/tidy/servo_tidy/tidy.py
@@ -7,6 +7,8 @@
 # option. This file may not be copied, modified, or distributed
 # except according to those terms.
 
+from __future__ import print_function
+
 import contextlib
 import fnmatch
 import imp
@@ -14,7 +16,7 @@ import itertools
 import json
 import os
 import re
-import StringIO
+from io import StringIO
 import subprocess
 import sys
 
@@ -23,6 +25,7 @@ import toml
 import voluptuous
 import yaml
 from licenseck import OLD_MPL, MPL, APACHE, COPYRIGHT, licenses_toml, licenses_dep_toml
+from six import iteritems
 topdir = os.path.abspath(os.path.dirname(sys.argv[0]))
 wpt = os.path.join(topdir, "tests", "wpt")
 
@@ -342,7 +345,7 @@ def check_flake8(file_name, contents):
         "E501",  # 80 character line length; the standard tidy process will enforce line length
     }
 
-    output = StringIO.StringIO()
+    output = StringIO()
     with stdout_redirect(output):
         check_code(contents, ignore=ignore)
     for error in output.getvalue().splitlines():
@@ -378,7 +381,7 @@ def check_lock(file_name, contents):
         if name not in packages_by_name:
             yield (1, "duplicates are allowed for `{}` but it is not a dependency".format(name))
 
-    for (name, packages) in packages_by_name.iteritems():
+    for (name, packages) in iteritems(packages_by_name):
         has_duplicates = len(packages) > 1
         duplicates_allowed = name in exceptions
 
@@ -422,7 +425,7 @@ def check_lock(file_name, contents):
                     visited_whitelisted_packages[dependency_name][package_name] = True
 
     # Check if all the exceptions to blocked packages actually depend on the blocked package
-    for dependency_name, package_names in blocked_packages.iteritems():
+    for dependency_name, package_names in iteritems(blocked_packages):
         for package_name in package_names:
             if not visited_whitelisted_packages[dependency_name].get(package_name):
                 fmt = "Package {} is not required to be an exception of blocked package {}."
@@ -514,7 +517,7 @@ def check_manifest_dirs(config_file, print_text=True):
         lines = conf_file.splitlines(True)
 
     if print_text:
-        print '\rChecking the wpt manifest file...'
+        print('\rChecking the wpt manifest file...')
 
     p = parser.parse(lines)
     paths = rec_parse(wpt_path("web-platform-tests"), p)
@@ -908,7 +911,7 @@ def check_config_file(config_file, print_text=True):
         lines = conf_file.splitlines(True)
 
     if print_text:
-        print '\rChecking the config file...'
+        print('\rChecking the config file...')
 
     config_content = toml.loads(conf_file)
     exclude = config_content.get("ignore", {})
@@ -995,7 +998,7 @@ def parse_config(config_file):
 
 def check_directory_files(directories, print_text=True):
     if print_text:
-        print '\rChecking directories for correct file extensions...'
+        print('\rChecking directories for correct file extensions...')
     for directory, file_extensions in directories.items():
         files = sorted(os.listdir(directory))
         for filename in files:
@@ -1015,7 +1018,7 @@ def collect_errors_for_files(files_to_check, checking_functions, line_checking_f
     if not has_element:
         raise StopIteration
     if print_text:
-        print '\rChecking files for tidiness...'
+        print('\rChecking files for tidiness...')
 
     for filename in files_to_check:
         if not os.path.exists(filename):
@@ -1037,7 +1040,7 @@ def collect_errors_for_files(files_to_check, checking_functions, line_checking_f
 
 def get_dep_toml_files(only_changed_files=False):
     if not only_changed_files:
-        print '\nRunning the dependency licensing lint...'
+        print('\nRunning the dependency licensing lint...')
         for root, directories, filenames in os.walk(".cargo"):
             for filename in filenames:
                 if filename == "Cargo.toml":
@@ -1136,11 +1139,11 @@ def scan(only_changed_files=False, progress=True, stylo=False):
     error = None
     for error in errors:
         colorama.init()
-        print "\r\033[94m{}\033[0m:\033[93m{}\033[0m: \033[91m{}\033[0m".format(*error)
+        print("\r\033[94m{}\033[0m:\033[93m{}\033[0m: \033[91m{}\033[0m".format(*error))
 
-    print
+    print()
     if error is None:
         colorama.init()
-        print "\033[92mtidy reported no errors.\033[0m"
+        print("\033[92mtidy reported no errors.\033[0m")
 
     return int(error is not None)