We originally made it per-window in b502280c06 for
issue #228, but that was back when we still needed window IDs for stuff like
message.info.
Nowadays, there's no reason for it to be per-window anymore. The rest of the
download code can deal with one global download manager (because QtWebEngine has
one), and apart from QNAM code which wasn't used here anyways (as tab_id=None)
there was nothing using the window ID anymore.
Also see #3456 which was the original motivation for this change.
What we actually want to test here is that the given type directory is created
and has the correct permission, we don't care much about the basedir itself.
Also, the download dir is not created automatically.
This test failed on Python 3.7 because intermediate directories now aren't
created with the given mode anymore:
https://bugs.python.org/issue19930https://docs.python.org/3.7/whatsnew/3.7.html#changes-in-the-python-api
After reading https://pyyaml.org/wiki/PyYAMLDocumentation again, turns out
Loader.add_constructor and .add_implicit_resolver are actually *class* methods.
In other words, we've been adding dozens of constructors/resolvers to the
default YAML loader object, causing it to slow down massively in other tests
which call configdata.init().
Instead, create our own loader class and only add them once there.
I'm still not sure why this caused the duration to increase with every YAML load
though - that might still be some kind of bug in PyYAML.
Fixes#2777
- fix test_renamed_key()
- fix test_deleted_key()
- combine both test_merge_persist tests using @pytest.mark.parametrize
- fix _handle_migrations(): mark data dirty for renamed and deleted
The test above this one loads hello.txt, but we don't wait for the "load
finished" message, so it can arrive after the previous test already finished and
make this test not wait properly.
However, we also can't easily wait for the load finished message in the
previous test as it only appears with QtWebEngine, not QtWebKit.
As a workaround, we simply load another file in that test, to circumvent this
kind of cross-interaction.
Scope down the new trigger-on-save behavior to only open-editor and
config-edit. Other uses of the editor such as edit-url and edit-command
will behave as before.
These are passing locally but failing in travis. This fixes two possible
timing issues:
- Ensure the signals are set up befor the pidfile is written. The
function that sends the signal waits for the pidfile to exist, so this
ensures we don't miss a signal.
- Wait for the log message indicating that the editor file was read
back, so the test doesn't run through before we get a chance to read
from the editor.
With the previous code, the editor could miss the final signal on a
save-and-exit. This is avoided by always running the file changed
handler on a successful exit, but only firing the signal if the content
actually changed (to avoid double-signalling).
Now that the editor signals on save, the configcommands editing
unittests need to emit the signal in the patch rather than relying on
on_proc_closed to emit the signal.
Now that the editor fires editing_finished on every write, the unit
tests had to be updated.
- Add qtbot to the editor fixture to resolve `QtWarningMsg:
QSocketNotifier: Can only be used with threads started with QThread`
- Use removePaths instead of disconnect to stop the watcher from
signalling. This avoids an error when the editor is forcibly cleaned
up by the tests without the signal ever being connected, but otherwise
has the same behavior as disconnecting the singal.
- wait for a signal on write instead of proc closed
- wait for _watcher.fileChanged in test_unreadable to ensure the write
event is fired before the test exits.
For any command that spawns an editor, tirgger an update on save, not
just on exit.
- :open-editor writes the text field on save
- :edit-url navigates on save
- :edit-url -t opens a new tab on each save
- :edit-command updates the statusbar text on save
- :edit-command --run runs a command on each save
- :config-edit reloads the config on save
Resolves#2307.
Helps mitigate #1596 by allowing users to 'save' partial work, and
notice if there was an error without closing the editor.
Like the spec says, if a value for the @include or @exclude rules starts
and ends with a '/' it should be parsed as a regular expression.
Technically a ECMAScript syntax regular expression, but I am not sure of
the differences and I assume they are far fewer than the similarities.
One that I did see mentioned was that javascript RegExp doesn't support
unicode. Although it apparently does support a 'u' flag now.
Note that code will only be ran for QtWebkit and QWebEngine < 5.8
we rely on the builtin support for metadata it QWebEngine for most
things greasemonkey related. Sadly it seems that they missed the regex
requirement too. I've opened a ticket to track that https://bugreports.qt.io/browse/QTBUG-65484
Always interpret the first word in the command string as the command to
offer completions for, even if that word looks like a flag.
Fixes#3460, where the command string `:-w open` would attempt to offer
completions for `open` but crash because the parsing was thrown off.
By moving the flag-stripping logic to _after_ we determine the command,
`:-w open` interprets `:-w` as the command. Since that is not a valid
command, we won't offer any completions.
Update the description to mention the number of columns and change the
default to ["white", "white", "white"] to make it more obvious that
multiple colors can be specified. This also satisfies the config test
that expects the default value for ListOrValue types to be a list.
One other test had to be tweaked to use a config option that is still
just a QtColor rather than a ListOrValue.
While it is possible to provide just two colors, it is "undefined
behavior". It will use the first color as the third color, but that is
an artifact of the implementation and therefore not documented (though
also not an error, as it is harmless).
colors.completion.fg is now a list instead of a QColor. As this test
specifically wanted to test a QColor, I just changed it to a different
config option.
command-accept --rapid will run the command without clearing the prompt,
allowing "rapid fire" commands. For example, one could open completion
for `open -t` and open several tabs in a row.
The default binding is ctrl+enter.
Resolves#588.
I mistakenly checked the length of wheres instead of words. This fixes
that check, renames 'wheres' to 'where_clause' to be clear
that it is a string and not an array, and adds a test.
There are various small changes here:
- If the process is already finished, we don't try to terminate it.
- On Windows, we use QProcess::kill instead of QProcess::terminate, as terminate
will only work with processes which have a GUI loop.
- We assert that quitting the suprocess actually worked.
Fixes#3384
Perviously, 'foo bar' would match 'foo/bar' but not 'bar/foo'. Now it
will match both, using a query with a WHERE clause like:
WHERE ((url || title) like '%foo%' AND (url || title) like '%bar%')
This does not seem to change the performance benchmark. However, it does
create a new query for every character added rather than re-running the
same query with different parameters. We could re-use queries if we
maintained a list like self._queries=[1_arg_query, 2_arg_query, ...].
However, it isn't clear that such a complexity would be necessary.
Resolves#1651.
This test is supposed to ensure that user scripts don't run on iframes
when the @noframes directive is set in the greasemonkey metadata. It is
failing sometimes on travis but passing on local test runs. Personally I
haven't actually ran the whole test suite through, just the javascript
tests. It maybe be some stale state that only shows up when you run the
whole suite. It may be some timing issue that only shows up on travis
because ???. Hopefully this stops the red x from showing up on the PR.
For some reason, if we don't wait for about:blank to be fully loaded with
Qt 5.10, we get the next LoadStatus.finished notification with about:blank as
URL.
This is most likely caused by the changes in
https://codereview.qt-project.org/#/c/202924/
See #3003
This reverts commit e72e8b8556.
Now that the SQL category works in isolation, it is possible to hide
quickmarks/bookmarks when those categories are empty.
Fixes#960
While QSortFilterProxyModel emits layoutChanged when changing the
pattern, QSqlQueryModel emits modelReset. As only layoutChanged was
connected, a HistoryCategory would only work in a model that also had at
least one ListCategory.
The simplest solution is to have the parent model emit the signal
directly. This also emits a single signal on a pattern change rather
that one for each child model.
Resolves#3016.
If we don't do this, earlier tests can affect later ones when e.g. using "...
should not be logged", as we don't really wait until a test has been fully
finished.
Test document-end and noframes. Because coverage.py told me to.
Hopefully this doesn't slow the test run down too much, particularly the
"should not be logged" bit.
I'm just reusing and existing test html page that used an iframe because
I'm lazy.
Just runs a greasemonkey script on a test page and uses console.log to
ensure it is running.
Tests @include, and basic happy path greasemonkey.py operation (loading
and parsing script, scrip_for on webkit), only testing document-start
injecting point but that is the troublsome one at this point.
Tested on py35 debian unstable (oldwebkit and qtwebengine5.9) debian
stable qtwebengine5.7.
Note the extra :reload call for qt5.7 because document-start scripts
don't seem to run on the first page load with the current insertion
point. I need to look into this more to look at ways of fixing this.
I can't reproduce this, but someone on KDE reported always getting a crash (as
msg.splitlines()[0] gives an IndexError) when trying to select a file with
Qt 5.9.3.
If an editor is open on a form in a tab and that tab is closed, rewire
the callback to print a warning. Previously, the callback would access a
deleted C++ object and cause a crash.
Resolves#2758.
Updated requirements and adjusted the configuration in `.flake8`; other
files have been modified where the lack of per-file auto-ignore caused
problems, where putty's `# flake8: disable=` syntax could be replaced
with a simpler `noqa`, or where pylint directives already suppressed the
same error.
Show an error message if the user edits the command such that it is
missing a start character (:, /, or ?). Previously, this would cause the
browser to crash.
Resolves#3326.
