From 747a9bc5b6dfc566c02e4616e5a3002dbe42a95f Mon Sep 17 00:00:00 2001
From: George Edward Bulmer <gebulmer@googlemail.com>
Date: Mon, 11 Dec 2017 21:32:55 +0000
Subject: [PATCH 1/5] Modify usage of re.match to fit re.fullmatch or
 re.search.

re.match features an implicit left anchor, which can be surprising.

re.fullmatch features implicit anchors on both sides, but is aptly named
and unsurprising.

re.search has no such implicit anchors, which ought to be the default
even if a single anchor is needed.
---
 qutebrowser/browser/downloads.py       |  2 +-
 qutebrowser/browser/webengine/spell.py |  2 +-
 qutebrowser/config/configtypes.py      | 10 +++++-----
 qutebrowser/keyinput/basekeyparser.py  |  3 ++-
 qutebrowser/misc/crashdialog.py        |  2 +-
 qutebrowser/misc/keyhintwidget.py      |  2 +-
 qutebrowser/utils/debug.py             |  2 +-
 qutebrowser/utils/docutils.py          |  2 +-
 qutebrowser/utils/urlutils.py          |  4 ++--
 qutebrowser/utils/utils.py             |  2 +-
 10 files changed, 16 insertions(+), 15 deletions(-)

diff --git a/qutebrowser/browser/downloads.py b/qutebrowser/browser/downloads.py
index 5a2daae7e..dc0d83cbc 100644
--- a/qutebrowser/browser/downloads.py
+++ b/qutebrowser/browser/downloads.py
@@ -180,7 +180,7 @@ def transform_path(path):
     path = utils.expand_windows_drive(path)
     # Drive dependent working directories are not supported, e.g.
     # E:filename is invalid
-    if re.match(r'[A-Z]:[^\\]', path, re.IGNORECASE):
+    if re.fullmatch(r'[A-Z]:[^\\]', path, re.IGNORECASE):
         return None
     # Paths like COM1, ...
     # See https://github.com/qutebrowser/qutebrowser/issues/82
diff --git a/qutebrowser/browser/webengine/spell.py b/qutebrowser/browser/webengine/spell.py
index ee2fb7813..9166180d4 100644
--- a/qutebrowser/browser/webengine/spell.py
+++ b/qutebrowser/browser/webengine/spell.py
@@ -30,7 +30,7 @@ from qutebrowser.utils import log
 def version(filename):
     """Extract the version number from the dictionary file name."""
     version_re = re.compile(r".+-(?P<version>[0-9]+-[0-9]+?)\.bdic")
-    match = version_re.match(filename)
+    match = version_re.fullmatch(filename)
     if match is None:
         raise ValueError('the given dictionary file name is malformed: {}'
                          .format(filename))
diff --git a/qutebrowser/config/configtypes.py b/qutebrowser/config/configtypes.py
index 8d0af173b..320798f6b 100644
--- a/qutebrowser/config/configtypes.py
+++ b/qutebrowser/config/configtypes.py
@@ -963,7 +963,7 @@ class Font(BaseType):
     # Gets set when the config is initialized.
     monospace_fonts = None
     font_regex = re.compile(r"""
-        ^(
+        (
             (
                 # style
                 (?P<style>normal|italic|oblique) |
@@ -976,14 +976,14 @@ class Font(BaseType):
                 (?P<size>[0-9]+((\.[0-9]+)?[pP][tT]|[pP][xX]))
             )\           # size/weight/style are space-separated
         )*               # 0-inf size/weight/style tags
-        (?P<family>.+)$  # mandatory font family""", re.VERBOSE)
+        (?P<family>.+)  # mandatory font family""", re.VERBOSE)
 
     def to_py(self, value):
         self._basic_py_validation(value, str)
         if not value:
             return None
 
-        if not self.font_regex.match(value):  # pragma: no cover
+        if not self.font_regex.fullmatch(value):  # pragma: no cover
             # This should never happen, as the regex always matches everything
             # as family.
             raise configexc.ValidationError(value, "must be a valid font")
@@ -1002,7 +1002,7 @@ class FontFamily(Font):
         if not value:
             return None
 
-        match = self.font_regex.match(value)
+        match = self.font_regex.fullmatch(value)
         if not match:  # pragma: no cover
             # This should never happen, as the regex always matches everything
             # as family.
@@ -1039,7 +1039,7 @@ class QtFont(Font):
         font.setStyle(QFont.StyleNormal)
         font.setWeight(QFont.Normal)
 
-        match = self.font_regex.match(value)
+        match = self.font_regex.fullmatch(value)
         if not match:  # pragma: no cover
             # This should never happen, as the regex always matches everything
             # as family.
diff --git a/qutebrowser/keyinput/basekeyparser.py b/qutebrowser/keyinput/basekeyparser.py
index 14efd00ab..658b52d2d 100644
--- a/qutebrowser/keyinput/basekeyparser.py
+++ b/qutebrowser/keyinput/basekeyparser.py
@@ -150,7 +150,8 @@ class BaseKeyParser(QObject):
             A (count, command) tuple.
         """
         if self._supports_count:
-            (countstr, cmd_input) = re.match(r'^(\d*)(.*)', keystring).groups()
+            (countstr, cmd_input) = re.fullmatch(r'(\d*)(.*)',
+                                                 keystring).groups()
             count = int(countstr) if countstr else None
             if count == 0 and not cmd_input:
                 cmd_input = keystring
diff --git a/qutebrowser/misc/crashdialog.py b/qutebrowser/misc/crashdialog.py
index 54827211a..9ef0e573c 100644
--- a/qutebrowser/misc/crashdialog.py
+++ b/qutebrowser/misc/crashdialog.py
@@ -62,7 +62,7 @@ def parse_fatal_stacktrace(text):
         r'(Current )?[Tt]hread [^ ]* \(most recent call first\): *',
         r'  File ".*", line \d+ in (?P<func>.*)',
     ]
-    m = re.match('\n'.join(lines), text)
+    m = re.search('\n'.join(lines), text)
     if m is None:
         # We got some invalid text.
         return ('', '')
diff --git a/qutebrowser/misc/keyhintwidget.py b/qutebrowser/misc/keyhintwidget.py
index 98dc8e275..46d1d3f24 100644
--- a/qutebrowser/misc/keyhintwidget.py
+++ b/qutebrowser/misc/keyhintwidget.py
@@ -87,7 +87,7 @@ class KeyHintView(QLabel):
         Args:
             prefix: The current partial keystring.
         """
-        countstr, prefix = re.match(r'^(\d*)(.*)', prefix).groups()
+        countstr, prefix = re.fullmatch(r'(\d*)(.*)', prefix).groups()
         if not prefix:
             self._show_timer.stop()
             self.hide()
diff --git a/qutebrowser/utils/debug.py b/qutebrowser/utils/debug.py
index 2c86e1142..44699efa1 100644
--- a/qutebrowser/utils/debug.py
+++ b/qutebrowser/utils/debug.py
@@ -184,7 +184,7 @@ def signal_name(sig):
     Return:
         The cleaned up signal name.
     """
-    m = re.match(r'[0-9]+(.*)\(.*\)', sig.signal)
+    m = re.fullmatch(r'[0-9]+(.*)\(.*\)', sig.signal)
     return m.group(1)
 
 
diff --git a/qutebrowser/utils/docutils.py b/qutebrowser/utils/docutils.py
index d645b41ac..e54effe8d 100644
--- a/qutebrowser/utils/docutils.py
+++ b/qutebrowser/utils/docutils.py
@@ -156,7 +156,7 @@ class DocstringParser:
     def _parse_arg_inside(self, line):
         """Parse subsequent argument lines."""
         argname = self._cur_arg_name
-        if re.match(r'^[A-Z][a-z]+:$', line):
+        if re.fullmatch(r'[A-Z][a-z]+:', line):
             if not self.arg_descs[argname][-1].strip():
                 self.arg_descs[argname] = self.arg_descs[argname][:-1]
                 return True
diff --git a/qutebrowser/utils/urlutils.py b/qutebrowser/utils/urlutils.py
index 93c41cec8..50b1632d0 100644
--- a/qutebrowser/utils/urlutils.py
+++ b/qutebrowser/utils/urlutils.py
@@ -306,7 +306,7 @@ def qurl_from_user_input(urlstr):
     """
     # First we try very liberally to separate something like an IPv6 from the
     # rest (e.g. path info or parameters)
-    match = re.match(r'\[?([0-9a-fA-F:.]+)\]?(.*)', urlstr.strip())
+    match = re.fullmatch(r'\[?([0-9a-fA-F:.]+)\]?(.*)', urlstr.strip())
     if match:
         ipstr, rest = match.groups()
     else:
@@ -571,7 +571,7 @@ def incdec_number(url, incdec, count=1, segments=None):
             continue
 
         # Get the last number in a string
-        match = re.match(r'(.*\D|^)(0*)(\d+)(.*)', getter())
+        match = re.fullmatch(r'(.*\D|^)(0*)(\d+)(.*)', getter())
         if not match:
             continue
 
diff --git a/qutebrowser/utils/utils.py b/qutebrowser/utils/utils.py
index 8debc6688..efd2b2cc8 100644
--- a/qutebrowser/utils/utils.py
+++ b/qutebrowser/utils/utils.py
@@ -869,7 +869,7 @@ def expand_windows_drive(path):
     # E:\. The correct way to specifify drive E: is "E:\", but most users
     # probably don't use the "multiple working directories" feature and expect
     # "E:" and "E:\" to be equal.
-    if re.match(r'[A-Z]:$', path, re.IGNORECASE):
+    if re.fullmatch(r'[A-Z]:', path, re.IGNORECASE):
         return path + "\\"
     else:
         return path

From f74832328f23fa71e73067e086887bb14cf0ae65 Mon Sep 17 00:00:00 2001
From: George Edward Bulmer <gebulmer@googlemail.com>
Date: Mon, 11 Dec 2017 23:06:52 +0000
Subject: [PATCH 2/5] Modify re.match usage in scripts directory.

---
 scripts/asciidoc2html.py   | 6 +++---
 scripts/dev/run_vulture.py | 2 +-
 scripts/dictcli.py         | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/scripts/asciidoc2html.py b/scripts/asciidoc2html.py
index ce29becc0..bd51b27ef 100755
--- a/scripts/asciidoc2html.py
+++ b/scripts/asciidoc2html.py
@@ -154,17 +154,17 @@ class AsciiDoc:
                     hidden = False
                 elif line == "The Compiler <mail@qutebrowser.org>\n":
                     continue
-                elif re.match(r'^:\w+:.*', line):
+                elif re.fullmatch(r':\w+:.*', line):
                     # asciidoc field
                     continue
 
                 if not found_title:
-                    if re.match(r'^=+$', line):
+                    if re.fullmatch(r'=+', line):
                         line = line.replace('=', '-')
                         found_title = True
                         title = last_line.rstrip('\n') + " | qutebrowser\n"
                         title += "=" * (len(title) - 1)
-                    elif re.match(r'^= .+', line):
+                    elif re.fullmatch(r'= .+', line):
                         line = '==' + line[1:]
                         found_title = True
                         title = last_line.rstrip('\n') + " | qutebrowser\n"
diff --git a/scripts/dev/run_vulture.py b/scripts/dev/run_vulture.py
index 9d21ad428..11cfce792 100755
--- a/scripts/dev/run_vulture.py
+++ b/scripts/dev/run_vulture.py
@@ -132,7 +132,7 @@ def filter_func(item):
         True if the missing function should be filtered/ignored, False
         otherwise.
     """
-    return bool(re.match(r'[a-z]+[A-Z][a-zA-Z]+', item.name))
+    return bool(re.fullmatch(r'[a-z]+[A-Z][a-zA-Z]+', item.name))
 
 
 def report(items):
diff --git a/scripts/dictcli.py b/scripts/dictcli.py
index 2742de8ad..98754d628 100755
--- a/scripts/dictcli.py
+++ b/scripts/dictcli.py
@@ -142,7 +142,7 @@ def parse_entry(entry):
     dict_re = re.compile(r"""
         (?P<filename>(?P<code>[a-z]{2}(-[A-Z]{2})?).*)\.bdic
     """, re.VERBOSE)
-    match = dict_re.match(entry['name'])
+    match = dict_re.fullmatch(entry['name'])
     if match is not None:
         return match.group('code'), match.group('filename')
     else:

From b07a4c8c2864d1ab54fdb4e58f36144e92eb0f8e Mon Sep 17 00:00:00 2001
From: George Edward Bulmer <gebulmer@googlemail.com>
Date: Tue, 12 Dec 2017 13:25:35 +0000
Subject: [PATCH 3/5] Add explicit left anchor to directory check regex.

---
 qutebrowser/browser/downloads.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/qutebrowser/browser/downloads.py b/qutebrowser/browser/downloads.py
index dc0d83cbc..354e74c9b 100644
--- a/qutebrowser/browser/downloads.py
+++ b/qutebrowser/browser/downloads.py
@@ -180,7 +180,7 @@ def transform_path(path):
     path = utils.expand_windows_drive(path)
     # Drive dependent working directories are not supported, e.g.
     # E:filename is invalid
-    if re.fullmatch(r'[A-Z]:[^\\]', path, re.IGNORECASE):
+    if re.search(r'^[A-Z]:[^\\]', path, re.IGNORECASE):
         return None
     # Paths like COM1, ...
     # See https://github.com/qutebrowser/qutebrowser/issues/82

From 9ca6baca4f624e99e76eae39b1bfe306525d1750 Mon Sep 17 00:00:00 2001
From: George Edward Bulmer <gebulmer@googlemail.com>
Date: Tue, 12 Dec 2017 15:07:37 +0000
Subject: [PATCH 4/5] Modify instances of re.match to fullmatch or search.

This applies the changes to the tests directory only.
---
 tests/end2end/conftest.py             | 4 ++--
 tests/end2end/fixtures/quteprocess.py | 2 +-
 tests/end2end/fixtures/testprocess.py | 2 +-
 tests/unit/misc/test_checkpyver.py    | 4 ++--
 tests/unit/utils/test_debug.py        | 4 ++--
 tests/unit/utils/test_utils.py        | 8 ++++----
 6 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/tests/end2end/conftest.py b/tests/end2end/conftest.py
index b2f7b1484..7f80d6acb 100644
--- a/tests/end2end/conftest.py
+++ b/tests/end2end/conftest.py
@@ -74,7 +74,7 @@ def _get_version_tag(tag):
         (?P<version>\d+\.\d+(\.\d+)?)
     """, re.VERBOSE)
 
-    match = version_re.match(tag)
+    match = version_re.fullmatch(tag)
     if not match:
         return None
 
@@ -125,7 +125,7 @@ def _get_dictionary_tag(tag):
         (?P<event>must_have_dict|cannot_have_dict)=(?P<dict>[a-z]{2}-[A-Z]{2})
     """, re.VERBOSE)
 
-    match = dict_re.match(tag)
+    match = dict_re.fullmatch(tag)
     if not match:
         return None
 
diff --git a/tests/end2end/fixtures/quteprocess.py b/tests/end2end/fixtures/quteprocess.py
index a0b48ab4b..7accddc03 100644
--- a/tests/end2end/fixtures/quteprocess.py
+++ b/tests/end2end/fixtures/quteprocess.py
@@ -48,7 +48,7 @@ def is_ignored_qt_message(message):
     """Check if the message is listed in qt_log_ignore."""
     regexes = pytest.config.getini('qt_log_ignore')
     for regex in regexes:
-        if re.match(regex, message):
+        if re.search(regex, message):
             return True
     return False
 
diff --git a/tests/end2end/fixtures/testprocess.py b/tests/end2end/fixtures/testprocess.py
index 570a3125c..eb7123388 100644
--- a/tests/end2end/fixtures/testprocess.py
+++ b/tests/end2end/fixtures/testprocess.py
@@ -335,7 +335,7 @@ class Process(QObject):
         if expected is None:
             return True
         elif isinstance(expected, regex_type):
-            return expected.match(value)
+            return expected.fullmatch(value)
         elif isinstance(value, (bytes, str)):
             return utils.pattern_match(pattern=expected, value=value)
         else:
diff --git a/tests/unit/misc/test_checkpyver.py b/tests/unit/misc/test_checkpyver.py
index 6487eb263..b61fadc59 100644
--- a/tests/unit/misc/test_checkpyver.py
+++ b/tests/unit/misc/test_checkpyver.py
@@ -44,7 +44,7 @@ def test_python2():
         pytest.skip("python2 not found")
     assert not proc.stdout
     stderr = proc.stderr.decode('utf-8')
-    assert re.match(TEXT, stderr), stderr
+    assert re.fullmatch(TEXT, stderr), stderr
     assert proc.returncode == 1
 
 
@@ -64,7 +64,7 @@ def test_patched_no_errwindow(capfd, monkeypatch):
     checkpyver.check_python_version()
     stdout, stderr = capfd.readouterr()
     assert not stdout
-    assert re.match(TEXT, stderr), stderr
+    assert re.fullmatch(TEXT, stderr), stderr
 
 
 def test_patched_errwindow(capfd, mocker, monkeypatch):
diff --git a/tests/unit/utils/test_debug.py b/tests/unit/utils/test_debug.py
index e32f564c8..b7b5cf3ca 100644
--- a/tests/unit/utils/test_debug.py
+++ b/tests/unit/utils/test_debug.py
@@ -90,8 +90,8 @@ class TestLogTime:
 
             assert len(caplog.records) == 1
 
-            pattern = re.compile(r'^Foobar took ([\d.]*) seconds\.$')
-            match = pattern.match(caplog.records[0].msg)
+            pattern = re.compile(r'Foobar took ([\d.]*) seconds\.')
+            match = pattern.fullmatch(caplog.records[0].msg)
             assert match
 
             duration = float(match.group(1))
diff --git a/tests/unit/utils/test_utils.py b/tests/unit/utils/test_utils.py
index f6eef7f4f..66f5e852d 100644
--- a/tests/unit/utils/test_utils.py
+++ b/tests/unit/utils/test_utils.py
@@ -863,7 +863,7 @@ class TestOpenFile:
         cmdline = '{} -c pass'.format(executable)
         utils.open_file('/foo/bar', cmdline)
         result = caplog.records[0].message
-        assert re.match(
+        assert re.fullmatch(
             r'Opening /foo/bar with \[.*python.*/foo/bar.*\]', result)
 
     @pytest.mark.not_frozen
@@ -872,7 +872,7 @@ class TestOpenFile:
         cmdline = '{} -c pass {{}} raboof'.format(executable)
         utils.open_file('/foo/bar', cmdline)
         result = caplog.records[0].message
-        assert re.match(
+        assert re.fullmatch(
             r"Opening /foo/bar with \[.*python.*/foo/bar.*'raboof'\]", result)
 
     @pytest.mark.not_frozen
@@ -882,7 +882,7 @@ class TestOpenFile:
         config_stub.val.downloads.open_dispatcher = cmdline
         utils.open_file('/foo/bar')
         result = caplog.records[1].message
-        assert re.match(
+        assert re.fullmatch(
             r"Opening /foo/bar with \[.*python.*/foo/bar.*\]", result)
 
     def test_system_default_application(self, caplog, config_stub, mocker):
@@ -890,7 +890,7 @@ class TestOpenFile:
                          new_callable=mocker.Mock)
         utils.open_file('/foo/bar')
         result = caplog.records[0].message
-        assert re.match(
+        assert re.fullmatch(
             r"Opening /foo/bar with the system application", result)
         m.assert_called_with(QUrl('file:///foo/bar'))
 

From 4844a68bfca3ef66f4b44c9611ecac9e9a7b6062 Mon Sep 17 00:00:00 2001
From: George Edward Bulmer <gebulmer@googlemail.com>
Date: Tue, 12 Dec 2017 21:42:10 +0000
Subject: [PATCH 5/5] Change a rogue re.fullmatch into a re.search.

Sometimes tests were timing out. Perhaps with fullmatch these tests were
rarely failing to parse the logs for the information.
---
 tests/end2end/fixtures/testprocess.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/end2end/fixtures/testprocess.py b/tests/end2end/fixtures/testprocess.py
index eb7123388..7836ab79a 100644
--- a/tests/end2end/fixtures/testprocess.py
+++ b/tests/end2end/fixtures/testprocess.py
@@ -335,7 +335,7 @@ class Process(QObject):
         if expected is None:
             return True
         elif isinstance(expected, regex_type):
-            return expected.fullmatch(value)
+            return expected.search(value)
         elif isinstance(value, (bytes, str)):
             return utils.pattern_match(pattern=expected, value=value)
         else: