Commit Graph

2290 Commits

Author SHA1 Message Date
Florian Bruhin
3a27c45ac9 More cleanup in test_webkitqutescheme 2016-09-14 12:08:35 +02:00
Florian Bruhin
5b527d0f1e Rename test_qutescheme to test_webkitqutescheme 2016-09-14 12:07:26 +02:00
Florian Bruhin
cc1e134f25 Fix test_qutescheme.py 2016-09-14 12:05:15 +02:00
Florian Bruhin
0657eeb8f4 Activate QtWebEngine tests for qute:* 2016-09-14 11:14:04 +02:00
Florian Bruhin
4a14083507 Fix :jseval --world contitional in set_text_field
See #1942
2016-09-13 07:59:21 +02:00
Ryan Roden-Corrent
e23d611b37 Strip quotes from completion pattern.
Given a commandline like:
`:set general editor "gvim -f"|`, the pattern should be 'gvim -f'
rather than '"gvim -f"'.
2016-09-12 22:19:44 -04:00
Ryan Roden-Corrent
e65aba74fd Test pattern for Completion.update_completion.
In the update_completion unit test, verify the `pattern` parameter as
well as the `model`.
2016-09-12 22:19:44 -04:00
Ryan Roden-Corrent
b867b87955 Don't crash Completer on unknown command.
The CommandRunner's fallback parsing behavior treated whitespace
differently than the normal flow. When a user entered an unknown
command, trailing whitespace would be stripped and the cmdline length
would be less than the cursor position.

This is fixed by making the fallback use the ShellLexer just as the
'normal' parsing does.
2016-09-12 22:19:44 -04:00
Ryan Roden-Corrent
52fdad8186 Test on_selection_changed with maxsplit 2016-09-12 22:19:44 -04:00
Ryan Roden-Corrent
d651cc75b0 Fix flake8/pylint errors. 2016-09-12 22:19:44 -04:00
Ryan Roden-Corrent
808a645b40 Fix quick-complete highlighting quirk.
When the commandline reads ':open |', quick-completing the only offered
completion will set the commandline to ':open some_url |'. Since `open`
has `maxsplit=0`, everything after ':open' is (correctly) treated as
one argument. This means completion is opened again with 'some url '
as the pattern (note trailing whitespace), which makes the comletion
menu 'flicker' and stay open even though it was 'supposed' to quick
compelte.

This is fixed by ignoring the next completion request if we just
completed something after maxsplit (because we don't expect any more
completions after the last split).

Resolves #1519.
2016-09-12 22:19:44 -04:00
Ryan Roden-Corrent
127412d91c Simplify update_completion.
Remove the class variables _cursor_part and _empty_item_index. Instead,
split up the commandline around the cursor whenever that information is
needed. Using locals instead of class variables makes the logic easier
to follow and ends up requiring much less code.
2016-09-12 22:19:44 -04:00
Ryan Roden-Corrent
581d7659ba Clean up Completer.on_selection_changed.
Remove the dependency on the class variables _empty_item_index
and _cursor_part to make the code easier to follow. If
_update_completion is refactored in a similar way these variables can
be removed.
2016-09-12 22:19:44 -04:00
Ryan Roden-Corrent
eb384fdda0 Better Completer.on_selection_changed testing.
Preparation for refactoring.
2016-09-12 22:19:44 -04:00
Florian Bruhin
e4192b5158 Yet another flaky test... 2016-09-12 22:49:55 +02:00
Florian Bruhin
1f10c7ff94 Fix broken hint test
The wait we added a few commits earlier was just plain wrong, filtering
the strings isn't going to print that, and shouldn't happen async
anyways...
2016-09-12 22:00:54 +02:00
Florian Bruhin
11abb642ab Remove downloading test with deleting handler
This actually is kind of tricky, and it makes later tests fail because
the tmpdir is suddenly gone...
2016-09-12 20:53:01 +02:00
Florian Bruhin
54c32cc74e Fix long line 2016-09-12 20:41:51 +02:00
Florian Bruhin
07e67740cc Add missing jseval.html 2016-09-12 20:02:03 +02:00
Florian Bruhin
e6680c3c60 Also redirect $HOME for QtWebEngine tests
See https://github.com/The-Compiler/qutebrowser/pull/1637#issuecomment-243043811
2016-09-12 19:55:31 +02:00
Florian Bruhin
08e95f76a6 QtWebEngine: Stabilize/reactivate two hint tests 2016-09-12 18:59:37 +02:00
Florian Bruhin
516bfb5ba3 Un-flaky another test waiting for scrolling 2016-09-12 18:58:31 +02:00
Florian Bruhin
c40325b510 Stabilize "Jumping back after searching" test
The test was flaky because waiting for scrolling didn't actually wait,
as the page logged a scroll position change to 0/0 directly after
loading.

We work around this by making the generic "And I wait until the scroll
position changed" not wait when it changed to 0/0.
2016-09-12 18:53:56 +02:00
Florian Bruhin
9e20e3a802 Skip :jseval with world tests for PyQt < 5.7.0 2016-09-12 18:30:49 +02:00
Florian Bruhin
c9e3cc04cf Accept a name for --world with :jseval 2016-09-12 18:23:23 +02:00
Florian Bruhin
2281f0c790 Fix :insert-text test for QtWebEngine 2016-09-12 15:59:23 +02:00
Florian Bruhin
b2608d7697 Add --world to :jseval 2016-09-12 15:59:17 +02:00
Florian Bruhin
fa78cc9f69 tests: Allow to mark JS errors as expected 2016-09-12 15:57:02 +02:00
Florian Bruhin
b645a88ade Stabilize :insert-text tests
With QtWebEngine, inserting text into the field is async, so if our
test runs too fast, it would fail. Now we instead log stuff via JS on
changes insteaad, and wait for those log messages in the tests.
2016-09-12 15:25:03 +02:00
Florian Bruhin
d988f919d7 Fix test_quteprocess.py 2016-09-12 12:01:31 +02:00
Florian Bruhin
2388489038 downloads: Be okay with the tmpdir being gone 2016-09-12 11:30:44 +02:00
Florian Bruhin
38a3d118ab Nicer linebreaking in _render_log 2016-09-12 11:24:57 +02:00
Florian Bruhin
469c75662e Fix docstring 2016-09-12 11:24:20 +02:00
Florian Bruhin
8afb02eef7 bdd: Don't log VDEBUG lines without --verbose 2016-09-12 11:23:56 +02:00
Florian Bruhin
fff777404b Use QApplication.postEvent again
We had some funny segfaults reported during scrolling (i.e. with
QApplication.sendEvent), and some code already had to use postpone=True
so there was no segfault...

So now we're back to postEvent again, and eliminated the main reason for
segfaults with it, which was re-using (and -posting) events which had
already been posted.

At least during tests this seems to run stable, let's hope it helps for
the people having crashes as well.
2016-09-11 19:37:09 +02:00
Florian Bruhin
ac7ee5ed8b Stabilize :tab-clone tests
Those were flaky with QtWebEngine on OS X:
https://travis-ci.org/The-Compiler/qutebrowser/jobs/159094434
2016-09-11 16:22:20 +02:00
Florian Bruhin
101d30fe1e Make _get_echo_exe_path() private in bdd conftest 2016-09-11 16:15:27 +02:00
Florian Bruhin
1fe2579bb1 Merge branch 'windows-userscripts' of https://github.com/Kingdread/qutebrowser into Kingdread-windows-userscripts 2016-09-11 16:14:25 +02:00
Florian Bruhin
11ee5befcc Fix download tests on Windows 2016-09-11 15:53:46 +02:00
Florian Bruhin
b949e4d73a Start splitting DownloadModel from DownloadManager
This mostly works, apart from remove_item/remove_items not being
available on the model.
2016-09-11 14:58:07 +02:00
Florian Bruhin
4c55963dc3 Fix lint 2016-09-10 17:32:04 +02:00
Florian Bruhin
b2d8905d68 Remove {downloaddir} in downloads.feature
We can just use (tmpdir) instead which we already use everywhere else as
well.
2016-09-10 16:42:27 +02:00
Florian Bruhin
0ba74ad023 Add Qt model test for download model 2016-09-10 16:42:26 +02:00
Florian Bruhin
d13f88f0ac Add lots of new BDD tests for downloads 2016-09-10 16:42:26 +02:00
Florian Bruhin
dc8396986e FIx expected message in test_invocations 2016-09-10 16:42:26 +02:00
Florian Bruhin
a21cd9f56e Set remove-finished-downloads in download tests 2016-09-10 16:42:26 +02:00
Florian Bruhin
17caf80f2a bdd: Add "I wait for the javascript message" step 2016-09-10 16:42:26 +02:00
Florian Bruhin
9f1aacc4f0 Move pdfjs check to the end2end conftest 2016-09-10 16:24:31 +02:00
Florian Bruhin
230cd40b12 Add a test for a misconfigured XDG_DOWNLOAD_DIR
See #866 and #1269.
2016-09-09 18:46:51 +02:00
Florian Bruhin
a4d0449e19 Use a shorter XDG_RUNTIME_DIR for temp_basedir_env
Otherwise with longer test names, we can't listen to the IPC server as
the filename is too long for a local socket.
2016-09-09 18:46:51 +02:00
Florian Bruhin
143a4af91b Improve :download-open tests 2016-09-09 18:46:51 +02:00
Florian Bruhin
2df971a79c Move short_tmpdir fixture to conftest.py 2016-09-09 18:46:51 +02:00
Florian Bruhin
b5b46332c7 Add test for downloading to an inexistent dir 2016-09-09 18:46:51 +02:00
Florian Bruhin
d64241bed9 Add a test for a download without content-size 2016-09-09 18:46:51 +02:00
Florian Bruhin
2177aa574c Improve download BDD tests
This reorganizes the existing tests a bit and adds some new ones.
2016-09-09 18:46:51 +02:00
Florian Bruhin
7611281563 Add basename to "download finished" message 2016-09-09 18:46:51 +02:00
Florian Bruhin
864ec94cde Fix lint 2016-09-09 17:56:58 +02:00
knaggita
9ff3f6810a Remove --qt-* arguments 2016-09-09 17:55:16 +02:00
Florian Bruhin
d0cf452ec8 Remove :prompt-yes and :prompt-no
Those are replaced by :prompt-accept yes and :prompt-accept no
2016-09-09 15:59:00 +02:00
Florian Bruhin
d579697245 Add a value argument to :prompt-accept 2016-09-09 15:48:42 +02:00
Florian Bruhin
71a89bd418 *Really* fix completion with multiple words.
Turns out re.escape also escapes spaces, so we'd need to replace '(\\ )'
groups after escaping. At this point it's easier to just combine spaces
before escaping the pattern.

Fixes #1934.
Supersedes #1935.
2016-09-09 09:03:18 +02:00
Daniel Schadt
b0114768c7 fix spawn tests on Windows
On Windows, no echo.exe exists normally, so calling echo from the tests
is no good idea, since it relies on Cygwin to be installed and in %PATH%
(so that echo.exe is available).

This fixes this by providing a small echo.bat which is callable from the
tests, and then using a platform-specific path to the executable instead
of the hardcoded "echo". This should ensure that the tests pass even on
systems where echo.exe is not installed.

Note that we can't simply use a do-nothing exe (like rundll or hh.exe),
as we're passing parameters, and those executables may behave
differently in the presence of those parameters.
2016-09-08 23:14:10 +02:00
Daniel Schadt
f6c6f766cd add newline at the end of the file 2016-09-08 23:13:00 +02:00
Florian Bruhin
0a3853fcb7 Fix handling of spaces in completion
This is a regression introduced in
faa052ad6f.

Fixes #1934.
2016-09-08 21:59:18 +02:00
Florian Bruhin
93428c4e2b Skip instead of xfail JS prompt test on WebEngine
We probably get stuck with a visual JS prompt, so for some reason the
following header tests fail.
2016-09-07 20:25:56 +02:00
Florian Bruhin
26025214ed Mark session loading tests with @qtwebengine_flaky 2016-09-07 19:03:41 +02:00
Florian Bruhin
a119e9d5b0 Mark :insert-text with undo test as xfail on OS X 2016-09-07 19:01:24 +02:00
Florian Bruhin
d299fd260c Fix adblock tests 2016-09-07 18:53:11 +02:00
Florian Bruhin
f3d540b375 Fix adblock tests 2016-09-07 18:47:39 +02:00
Florian Bruhin
44d1056e54 QtWebEngine: Implement custom HTTP headers 2016-09-07 18:29:38 +02:00
Florian Bruhin
3ce9e795a1 Add a simple end2end test for :adblock-update 2016-09-07 16:27:27 +02:00
Florian Bruhin
8cef35517f Add a manual test file for uploading files 2016-09-07 13:40:05 +02:00
Florian Bruhin
fca37abf55 QtWebEngine: Implement webelem.classes 2016-09-07 12:31:48 +02:00
Florian Bruhin
522049132b Add a test for hinting with ACE editor 2016-09-07 12:28:12 +02:00
Florian Bruhin
3e1583bb1c Make WebKitElement._is_visible private
It makes a lot of sense for this to be in webkitelem.py, but it should
not be public API as it's only used internally and can't be implemented
here with QtWebEngine.
2016-09-07 11:58:39 +02:00
Florian Bruhin
60c86a08c4 Get rid of WebElement.debug_text 2016-09-07 11:47:39 +02:00
Florian Bruhin
2da827462e Add a test to make sure check_coverage files exist 2016-09-07 11:37:21 +02:00
Florian Bruhin
e700d11447 Add check_vanished() in webkitelem.insert_text 2016-09-07 11:29:26 +02:00
Florian Bruhin
8f9cfcf232 Get rid of webelem.run_js_async 2016-09-07 11:27:37 +02:00
Florian Bruhin
78d64f4791 Get rid of webkitelem.focus_elem 2016-09-07 11:24:28 +02:00
Florian Bruhin
c267776491 Add missing :leave-mode for insert-mode test
Otherwise we hint while still in insert mode - while this seems to work
fine, let's do things the cleaner way.
2016-09-07 11:21:27 +02:00
Florian Bruhin
948fa033c7 Implement :insert-text for QtWebEngine 2016-09-07 11:20:32 +02:00
Florian Bruhin
ee5a972069 Move auto-insert-mode handling to browsertab
This means auto-insert-mode now works correctly with QtWebEngine.
2016-09-07 10:24:27 +02:00
Florian Bruhin
919196714b QtWebEngine: Implement WebEngineScroller.at_bottom 2016-09-07 09:32:51 +02:00
Florian Bruhin
cf6cffeb08 tests: Ignore "Xlib: sequence lost" error 2016-09-06 22:10:34 +02:00
Florian Bruhin
08302c5a5f Apply @qtwebengine_flaky (too) liberally 2016-09-06 22:10:28 +02:00
Florian Bruhin
1106104700 Add a @qtwebengine_flaky marker 2016-09-06 21:43:02 +02:00
Florian Bruhin
37b1c26e1c Don't check active win for WebEngine session test 2016-09-06 21:22:50 +02:00
Florian Bruhin
e3e7366863 Make pylint shut up 2016-09-06 21:21:37 +02:00
Daniel Schadt
22ac19b151 style fixes 2016-09-06 20:33:48 +02:00
Florian Bruhin
c58a0e926f Skip scrolling tests on Qt < 5.7
This fixes tests on OS X on Travis, since Homebrew is still on Qt 5.6.
2016-09-06 20:29:58 +02:00
Florian Bruhin
22d9c49169 Stabilize scrolling in :repeat-command test 2016-09-06 19:34:38 +02:00
Florian Bruhin
010a52efb8 Stabilize :tab-detach test in invoke.feature
When we detach a tab, we need to wait until the page is actually loaded
in the newly detached tab before saving the session.
2016-09-06 19:33:09 +02:00
Florian Bruhin
c59a1535c5 Don't check active window in session test
This is flaky as we don't know which window is actually focused.
2016-09-06 19:29:12 +02:00
Daniel Schadt
0ab1902c98 add an userscript test for Windows 2016-09-06 18:41:24 +02:00
nanjekyejoannah
d3e19ec8fc Add general -> yank-ignored-url-parameters 2016-09-06 18:19:40 +02:00
Florian Bruhin
e77bf62ace Remove unnecessary @qtwebengine_createWindow tags
Since we did now remove the error message, some more tests actually run
fine without the createWindow patch.
2016-09-06 18:04:54 +02:00
Florian Bruhin
7356e32b94 QtWebEngine: Don't search the Holy Grail 2016-09-06 17:38:13 +02:00
Florian Bruhin
fcaadb9352 Stabilize :save-session --force test 2016-09-06 17:04:13 +02:00
Florian Bruhin
c6d6483929 bdd: Show what exactly failed with InvalidLine 2016-09-06 17:00:25 +02:00