Redesign and rearrange Thunderbird Preferences/Options similar to Firefox
Categories
(Thunderbird :: Preferences, enhancement)
Tracking
(Not tracked)
People
(Reporter: Elio, Assigned: Paenglab)
References
(Depends on 1 open bug)
Details
Attachments
(7 files, 26 obsolete files)
41.00 KB,
image/png
|
Details | |
63.54 KB,
patch
|
Paenglab
:
review+
Paenglab
:
ui-review+
|
Details | Diff | Splinter Review |
52.10 KB,
patch
|
Paenglab
:
review+
Paenglab
:
ui-review+
|
Details | Diff | Splinter Review |
74.37 KB,
patch
|
Paenglab
:
review+
Paenglab
:
ui-review+
|
Details | Diff | Splinter Review |
247.57 KB,
patch
|
Paenglab
:
review+
Paenglab
:
ui-review+
|
Details | Diff | Splinter Review |
35.96 KB,
patch
|
Paenglab
:
review+
Paenglab
:
ui-review+
|
Details | Diff | Splinter Review |
37.31 KB,
patch
|
darktrojan
:
review+
|
Details | Diff | Splinter Review |
Comment 1•7 years ago
|
||
Reporter | ||
Comment 2•7 years ago
|
||
Reporter | ||
Comment 3•7 years ago
|
||
Assignee | ||
Comment 4•7 years ago
|
||
Comment 5•7 years ago
|
||
Reporter | ||
Comment 6•7 years ago
|
||
Comment 7•7 years ago
|
||
Reporter | ||
Comment 8•7 years ago
|
||
Reporter | ||
Comment 9•7 years ago
|
||
Comment 10•7 years ago
|
||
Comment 11•7 years ago
|
||
Comment 12•7 years ago
|
||
Comment 13•7 years ago
|
||
Reporter | ||
Comment 14•7 years ago
|
||
Reporter | ||
Comment 15•7 years ago
|
||
Wow, the link again was again the wrong one (not sure what I did in Discourse). Here the thread on Open Source Design: https://discourse.opensourcedesign.net/t/thunderbird-preferences-views-redesign/804/3
So great news is that the Usability Study including testing of the current Preferences view and the upcoming design is completed! We are currently wrapping up things and publish the results soon.
Comment 16•6 years ago
|
||
Is this stalled for a particular reason?
Comment 17•6 years ago
|
||
AFAIK it is waiting for implementation. NEED INFOing Magnus and Alessandro so that it is on their radar.
Magnus, Alessandro - please put this on your todolist so it doesn't get forgotten.
Assignee | ||
Comment 18•6 years ago
|
||
(In reply to Elio Qoshi [:Elio] from comment #15)
So great news is that the Usability Study including testing of the current Preferences view and the upcoming design is completed! We are currently wrapping up things and publish the results soon.
I can't remember seeing results.
I'm also not sure if Alessandro is happy with the Composition / Attachments / filelink presentation.
Reporter | ||
Comment 20•6 years ago
|
||
(In reply to Richard Marti (:Paenglab) from comment #18)
(In reply to Elio Qoshi [:Elio] from comment #15)
So great news is that the Usability Study including testing of the current Preferences view and the upcoming design is completed! We are currently wrapping up things and publish the results soon.
I can't remember seeing results.
I'm also not sure if Alessandro is happy with the Composition / Attachments / filelink presentation.
The results have been published in January and the discussion happened on the UX list. The study can be downloaded here: https://ura.design/projects/thunderbird
We spent a lot of time on the study and involved both community and user feedback. This not being liked comes as a surprise to me when the whole scope of work was tailored in a wider UX meeting with folks who deemed this work important and relatively low hanging. Now that Alessandro joined the core team, it would be great to discuss this and whether some design considerations are not ideal. I understand that there are other priorities now, but at the time this work started (November 2018) it was considered to be also a priority UX-wise.
Comment 21•6 years ago
|
||
Thanks for pulling me into this, I will review the study and the entire thread in order to reopen the discussion later next week.
We're on a bit of a crunch for 68 string freeze right now.
After that, I'll jump right into it.
Assignee | ||
Comment 22•6 years ago
|
||
This implements the "General" pane according to https://www.figma.com/proto/DAIuhXdQ0G9AExbKLCQZ1HSE/Thunderbird-Preferences?node-id=12%3A560&viewport=288%2C524%2C0.426706&scaling=min-zoom&redirected=1
Aceman, Geoff, when I open the prefs I get: TypeError: can't redefine non-configurable property "Downloads"XPCOMUtils.jsm:316:19
. Please can you say me how I can fix this?
Comment 23•6 years ago
|
||
Comment 25•6 years ago
|
||
Sweet, thanks for starting on this.
Indeed, as Geoff pointed out, we shouldn't land this until we have the Search Box implemented and usable, otherwise scrolling through the preferences will be a nightmare.
Assignee | ||
Comment 26•6 years ago
|
||
Comment 27•6 years ago
|
||
(In reply to Alessandro Castellani (:aleca) from comment #25)
Sweet, thanks for starting on this.
Indeed, as Geoff pointed out, we shouldn't land this until we have the Search Box implemented and usable, otherwise scrolling through the preferences will be a nightmare.
I usually try to keep out of UI discussions, but can someone explain to me why following Firefox here is a good idea. A long time ago, FF prefs where neatly divided into categories, now they are all lumped into endless pages which require lots of scrolling, searching, etc.
Or to us the words of TB's lead UX engineer: " ... scrolling through the preferences will be a nightmare." - So why do it?
Assignee | ||
Comment 28•6 years ago
|
||
(In reply to Alessandro Castellani (:aleca) from comment #25)
Sweet, thanks for starting on this.
Indeed, as Geoff pointed out, we shouldn't land this until we have the Search Box implemented and usable, otherwise scrolling through the preferences will be a nightmare.
I'm against it. This will take the AM in pref tab way: Not going further because it's not checked-in. We should land it, it's then in non-official builds and has time to grow until 75, the next official build.
For the search we should do this in three steps: 1. land this bug. 2. convert it to the FX way: a big xhtml <page> with no <prefpane>. 3. now implement the search.
With this way the search should easy to implement as we can more or less copy the FX search implementation.
I'll finish this patches here. But if you don't want to land them, I'm stopping then the maintenance to update this patches and someone other can un-bitrot this patches all the time until it lands. Again, we don't land it on ESR soon, it's then on nightly and later on beta.
Assignee | ||
Comment 29•6 years ago
|
||
Jörg, see the Usability Study at https://ura.design/projects/thunderbird
Comment 30•6 years ago
|
||
Thanks Richard for the clarification.
I see your point and it makes sense, considering it won't land on ESR but it'll stay on nightly until 75.
Upon further inspection of the various sections, I noticed that we have many tabs with only one or two options, so listing them all in one page won't be a problem.
Or to us the words of TB's lead UX engineer: " ... scrolling through the preferences will be a nightmare." - So why do it?
I should have phrase that better, my bad.
I assumed the single panels would turn into an infinite list of options, but that won't be the case.
Many of our tabs are almost empty, and as emerged from the usability study, splitting content into tabs is not the best for accessibility and discoverability.
Comment 31•6 years ago
|
||
Agreed this looks great, and we don't need to have more than the basics for it to land and we do have time to improve it adding a search before 76.
Assignee | ||
Comment 32•6 years ago
|
||
Okay, this part is ready for review.
Geoff, you made a lot on the prefs. So I think you're the right to review this.
The patch moved the parts to General pane. The other panes are still working, only without the moved parts.
In preparation to the move to the FX implementation I already added the data-category="paneGeneral"
which will be used to show/hide the parts depending of the selected category.
Assignee | ||
Comment 33•6 years ago
|
||
This patch should work except the Default startup directory menulist which is empty. When selecting a directory it is shown in the menulist, but after closing/opening of the prefs it is empty again. I see no errors in the console.
Aceman, please can you look what is wrong with my patch? You need to apply 1498332-rearrange-prefs-part-1.patch first.
Comment 34•6 years ago
|
||
Comment 35•6 years ago
|
||
Assignee | ||
Comment 36•6 years ago
|
||
Ui-r comments should be addressed.
Assignee | ||
Comment 37•6 years ago
|
||
Aceman, I updated this patch to apply after part 1. My issue is still the Default startup directory menulist which is empty. When selecting a directory it is shown in the menulist, but after closing/opening of the prefs it is empty again. I see no errors in the console.
Please help me to find the issue.
Comment 38•6 years ago
|
||
Assignee | ||
Comment 39•6 years ago
|
||
Assignee | ||
Comment 40•6 years ago
|
||
Thank you, Alessandro. Fixed the two remaining spacings.
Comment 41•6 years ago
|
||
Assignee | ||
Comment 42•6 years ago
|
||
(In reply to Alessandro Castellani (:aleca) from comment #41)
Would be possible to change the icon used in the sidebar? "Composition"
still uses the old pencil
I'll do it in part 4 where I do the CSS cleanup and remove the no more used icons.
Comment 43•6 years ago
|
||
Assignee | ||
Comment 44•6 years ago
|
||
(In reply to Geoff Lankow (:darktrojan) from comment #43)
Comment on attachment 9075738 [details] [diff] [review]
1498332-rearrange-prefs-part-1.patchReview of attachment 9075738 [details] [diff] [review]:
Fine, with some comments. On the whole, given that this is mostly
cut-and-paste, I probably looked too closely at the XUL and not closely
enough at the JS.You'll need to adjust the mochitests before landing this.
I'll plan to look when all patches are ready. I hope I can fix them myself.
::: mail/components/preferences/advanced.js
@@ +39,1 @@The linter doesn't like this empty line.
Fixed. Before I thought I'll let it like it is because this file will be deleted. But better it's clean.
::: mail/components/preferences/compose.inc.xul
@@ +3,5 @@file, You can obtain one at http://mozilla.org/MPL/2.0/.
<prefpane id="paneCompose" label="&paneComposition.title;">
<script src="chrome://messenger/content/preferences/compose.js"/>
<script src="chrome://global/content/contentAreaUtils.js"/>
- <script src="chrome://messenger/content/preferences/downloads.js"/>
What's this doing here?
I had to insert it after contentAreaUtils.js to not get the error from comment 22.
I'll plan to move all <script> to aboutPreferences.xul in a follow-up bug.
::: mail/components/preferences/general.inc.xul
@@ +7,5 @@
- <script src="chrome://mozapps/content/preferences/fontbuilder.js"/>
- <script>
- <![CDATA[
+#ifdef MOZ_WIDGET_GTK- var ICON_URL_APP = "moz-icon://dummy.exe?size=16";
Let's do this with AppConstants.MOZ_WIDGET_GTK and not use preprocessing and
an inline script.
Done.
@@ +34,4 @@
<stringbundle id="bundlePreferences" src="chrome://messenger/locale/preferences/preferences.properties"/>
+#ifdef HAVE_SHELL_SERVICE
- <stringbundle id="bundlePreferences" src="chrome://messenger/locale/preferences/preferences.properties"/>
This stringbundle is already here.
I must be blind. :-(
We should probably stop using <stringbundle> altogether as we change this
code (and use Services.strings instead), but whether you do that or not is
up to you.
Better not in this bug to not mess too much.
@@ +208,5 @@
<hbox>
<radio id="appLocale"
value="false"/>
<!-- label and accesskey will be set dynamically -->
<spacer flex="1"/>
I know this is cut-and-pasted, but I'm going to pick on it anyway. :-)
The <spacer>s are redundant. The outer <hbox> is redundant (as well as
messing up the indent). If the <radiogroup> was align="start", the inner
<hbox>es would be redundant.
Both fixed.
In fact, I see this pattern all over the place. Does it do something I'm not
seeing?
It was to make not the whole line selectable. I can fix them in a follow-up bug.
Indentation.
Fixed all indentations you found.
@@ +488,5 @@
</vbox>
<vbox>
<separator/>
Shouldn't this be outside the vbox?
Moved.
I found an issue with the applicationManager and fixed it (still pointed to gApplicationsPane instead to gGeneralPane).
![]() |
||
Comment 45•6 years ago
|
||
Ah, another change I see no benefit in :)
(In reply to Richard Marti (:Paenglab) from comment #37)
Aceman, I updated this patch to apply after part 1. My issue is still the Default startup directory menulist which is empty. When selecting a directory it is shown in the menulist, but after closing/opening of the prefs it is empty again. I see no errors in the console.
The menulist id=defaultStartupDirList does not contain any items when initAbDefaultStartupDir() is run thus no item is selected and the menulist appears empty. Only later after that the menulist is populated with the list of addressbook as the connectedCallback() of that particular menulist is run.
It seems the "paneload" event (which runs the init functions of the panes) is run sooner than the custom element is built. I don't know when exactly custom elements load and build their contents, but it's not the first problem of their weird timing.
Assignee | ||
Comment 46•6 years ago
|
||
Thanks Aceman. Geoff, do you know how this could be solved?
Assignee | ||
Comment 47•6 years ago
|
||
I think also with the menulist problem in part 2 we can go further with the ui-r.
This patch implements the "Privacy & Security" pane.
Assignee | ||
Comment 48•6 years ago
|
||
This is the patch for the "Chat" pane. Now with the fixed chat crash this works too without additional back-out patch.
Assignee | ||
Comment 49•6 years ago
|
||
This is the "Calendar" part.
The mockup forgot the Categories. I added them at the bottom of the pane, okay?
![]() |
||
Comment 50•6 years ago
|
||
Also notice the long page can't be scrolled by Home/End/Page Up/Page Down, unless you click some option. It is not enough to click any empty space inside the page, as would work in any other web page.
But this bug is in Firefox too.
Comment 51•6 years ago
|
||
The menulist id=defaultStartupDirList does not contain any items when initAbDefaultStartupDir() is run thus no item is selected and the menulist appears empty. Only later after that the menulist is populated with the list of addressbook as the connectedCallback() of that particular menulist is run.
It looks like putting setTimeout(…) around the call to that function from gComposePane.init does the trick.
![]() |
||
Comment 52•6 years ago
|
||
Maybe we could run whole gComposePane.init from a timeout.
Anyway those are all hack on hacks. What is the proper way to handle custom elements and run after they are already initialized?
We have more problems like this in the preferences pane (e.g. opening the Prefs tab always throws a problem with this.mSelectedInternal from some uninitialized menulist).
Assignee | ||
Comment 53•6 years ago
|
||
I tried the setTimeout(…) this weekend it didn't work for me. How should the command look and which value did you use?
Comment 54•6 years ago
|
||
Comment 55•6 years ago
•
|
||
Comment 56•6 years ago
|
||
Updated•6 years ago
|
Comment 57•6 years ago
|
||
Comment 58•6 years ago
|
||
Comment 59•6 years ago
|
||
(In reply to :aceman from comment #52)
Maybe we could run whole gComposePane.init from a timeout.
Anyway those are all hack on hacks. What is the proper way to handle custom elements and run after they are already initialized?
We have more problems like this in the preferences pane (e.g. opening the Prefs tab always throws a problem with this.mSelectedInternal from some uninitialized menulist).
I don't know about proper ways to do this, but I think it's happening because all the initialisation code is running as soon as the parser gets to it. Does that prevent the CE bindings from applying beforehand? I guess so.
We've already got a function at the top of preferences.js which if put in a no-delay timeout should run after the CE stuff.
Assignee | ||
Comment 60•6 years ago
|
||
I added now the setTimeout() hack with 1 ms delay which works for me, is more needed for slower systems? I hope, this is okay for now and a better solution can be found in a new bug.
Patch is also updated after landing of bug 1564294.
Assignee | ||
Comment 61•6 years ago
|
||
Assignee | ||
Comment 62•6 years ago
|
||
Added a <separator/> before the "When message directed at you arrive:" text.
Assignee | ||
Comment 63•6 years ago
|
||
Alessandro, does this look better?
Updated•6 years ago
|
Updated•6 years ago
|
Assignee | ||
Updated•6 years ago
|
Assignee | ||
Comment 64•6 years ago
|
||
Updated after bug 1562560.
Assignee | ||
Comment 65•6 years ago
|
||
Assignee | ||
Comment 66•6 years ago
|
||
Assignee | ||
Comment 67•6 years ago
|
||
Assignee | ||
Comment 68•6 years ago
|
||
Comment 69•6 years ago
|
||
(In reply to Richard Marti (:Paenglab) from comment #60)
I added now the setTimeout() hack with 1 ms delay which works for me, is more needed for slower systems? I hope, this is okay for now and a better solution can be found in a new bug.
It's not a timing issue, it's an order of operations issue. setTimeout
even with 0 delay is basically saying "do this when you've finished doing everything else".
I think what we should do is put this line inside a timeout. That would delay all of the things listening for a paneload
event, and also seems to solve another problem I've been noticing (that sometimes nothing appears on the tab if it's open when Thunderbird starts).
Comment 70•6 years ago
|
||
Comment 71•6 years ago
|
||
I've noticed that when scrolled down in a pane, if you change pane the new pane is also scrolled down. That's not right.
Also the chat pane jumps to the bottom the first time it is shown.
![]() |
||
Comment 72•6 years ago
|
||
(In reply to Geoff Lankow (:darktrojan) from comment #69)
I think what we should do is put this line inside a timeout. That would delay all of the things listening for a
paneload
event, and also seems to solve another problem I've been noticing (that sometimes nothing appears on the tab if it's open when Thunderbird starts).
That's what I was aiming at in comment 52.
Comment 73•6 years ago
|
||
Comment 74•6 years ago
|
||
Comment 75•6 years ago
|
||
Assignee | ||
Comment 76•6 years ago
|
||
Added the
setTimeout(function() {pane.dispatchEvent(new CustomEvent("paneload"))}, 1);
and the
document.getElementById("preferencesContainer").scrollTo(0, 0);
to preferences.js.
Assignee | ||
Comment 77•6 years ago
|
||
(In reply to Geoff Lankow (:darktrojan) from comment #70)
Comment on attachment 9077455 [details] [diff] [review]
1498332-rearrange-prefs-part-2.patchReview of attachment 9077455 [details] [diff] [review]:
::: mail/components/preferences/applications.inc.xul
@@ -1,4 @@-# This Source Code Form is subject to the terms of the Mozilla Public
-# License, v. 2.0. If a copy of the MPL was not distributed with this
-# file, You can obtain one at http://mozilla.org/MPL/2.0/.
- <prefpane id="paneApplications" label="&paneAttachments.title;">
You'll need to change a reference to this in downloads.js, that was added
with bug 1562560.
fixed
::: mail/components/preferences/compose.inc.xul
@@ +70,3 @@
<hbox align="center" pack="start">
<label value ="&languagePopup.label;" control="languageMenuList" accesskey="&languagePopup.accessKey;"/>
There's a stray space here. It's not your fault, but please fix it.
fixed
@@ +276,5 @@
href="https://addons.thunderbird.net/thunderbird/tag/filelink"
value="&findCloudFileProviders.label;"/>
</vbox>
<separator class="thin" orient="vertical"/>
<vbox flex="1" style="height: 400px;">
This height needs to be on the parent box. If you add too many providers,
the whole thing expands.
Moved to the parent <hbox>
Additionally, the random image from the WeTransfer provider now overflows
and you can't see the explanation text below it (not that it helps, often).
Made it 480px height.
::: mail/components/preferences/compose.js
@@ +74,3 @@}
- setTimeout(function() {gComposePane.initAbDefaultStartupDir()}, 1);
See comment 69.
Is now in patch-1.
Assignee | ||
Comment 78•6 years ago
|
||
(In reply to Geoff Lankow (:darktrojan) from comment #73)
Comment on attachment 9077456 [details] [diff] [review]
1498332-rearrange-prefs-part-3.patchReview of attachment 9077456 [details] [diff] [review]:
Nothing too exciting here.
::: mail/components/preferences/privacy.js
@@ +297,5 @@
- */
- openTextLink(evt) {
- // Opening links behind a modal dialog is poor form. Work around flawed
- // text-link handling by opening in browser if we'd instead get a content
- // tab behind the modal options dialog.
This can probably go. It's not a modal dialog any more!
Removed the function as theis="text-link"
does the opening of the link.
@@ +328,5 @@
- initSubmitCrashes() {
- var checkbox = document.getElementById("submitCrashesBox");
- try {
var cr = Cc["@mozilla.org/toolkit/crash-reporter;1"].
getService(Ci.nsICrashReporter);
Let's line the dot up with the bracket while we're here.
Done
@@ +346,5 @@
- updateSubmitCrashes() {
- var checkbox = document.getElementById("submitCrashesBox");
- try {
var cr = Cc["@mozilla.org/toolkit/crash-reporter;1"].
getService(Ci.nsICrashReporter);
And this one.
Done
Assignee | ||
Comment 79•6 years ago
|
||
(In reply to Geoff Lankow (:darktrojan) from comment #74)
Comment on attachment 9077457 [details] [diff] [review]
1498332-rearrange-prefs-part-4.patchReview of attachment 9077457 [details] [diff] [review]:
::: mail/components/preferences/chat.inc.xul
@@ +17,5 @@
- <groupbox data-category="paneChat">
<label><html:h2 data-l10n-id="chat-status-title"/></label>
<!-- Startup -->
<hbox align="center">
<label id ="chatStartupAction" value="&startupAction.label;"
Another stray space. I don't know why these stand out so much.
Fixed
::: mail/locales/en-US/messenger/preferences/preferences.ftl
@@ +50,5 @@privacy-certificates-title = Certificates
+chat-pane-header = Chat
+
+chat-status-title =StatusSpace.
Fixed
Assignee | ||
Comment 80•6 years ago
|
||
(In reply to Geoff Lankow (:darktrojan) from comment #75)
Comment on attachment 9077458 [details] [diff] [review]
1498332-rearrange-prefs-part-5.patchReview of attachment 9077458 [details] [diff] [review]:
This pane looks a lot better than it did. Well done.
::: calendar/lightning/content/messenger-overlay-preferences.xul
@@ +33,5 @@flex="1" insertbefore="paneAdvanced" label="&lightning.preferencesLabel;">
- <hbox id="chatPaneCategory"
These are all id="chatPaneCategory". And now that I've noticed it, there's
two in chat.inc.xul with the same id.
Fixed here and in Chat.
::: calendar/locales/en-US/chrome/calendar/preferences/general.dtd
@@ +19,5 @@<!ENTITY pref.defaults.label "Default Values for New Items">
<!ENTITY pref.events.label "Events">
<!ENTITY pref.tasks.label "Tasks">+<!ENTITY pref.default_event_task_length.label "Default Event and Tasks Length">
Task should be singular.
Fixed
::: calendar/locales/en-US/chrome/calendar/preferences/views.dtd
@@ +6,5 @@- If this ==> … <== doesn't look like an ellipsis (three dots in a row), - your editor isn't using UTF-8 encoding and may munge up the document!
-->
+<!ENTITY pref.calendar.view.format.caption "Formats">
I don't think Formats is really the right label, and I wonder if we even
need one in this spot.
I removed it.
Assignee | ||
Comment 81•6 years ago
|
||
I'm now looking for the tests.
Assignee | ||
Comment 82•6 years ago
|
||
Shoot, introduced 3 linting errors in one line with the setTimeout(function() { pane.dispatchEvent(new CustomEvent("paneload")); }, 1);
Assignee | ||
Comment 83•6 years ago
|
||
I tried what I could with my limited knowledge.
See https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=d6cbe5f4eb206263c5d1c1bd8d421556d22f9f20
Aceman, Geoff, please could you look what is wrong with my tests?
I've set for the no more existing tabs in the pane the undefined like https://searchfox.org/comm-central/source/mail/components/preferences/test/browser/browser_privacy.js#6 already had. I think this works now but something else is wrong in the tests.
For removing the aTabID I filed bug 1565789.
Comment 84•6 years ago
|
||
Once bug 1565789 is done, you should use it in the tests, as it looks like a lot of the failures are from clicking on things that are off-screen.
Assignee | ||
Comment 85•6 years ago
|
||
This is my latest test patch with the scrollTo elements for bug 1565789.
What do I wrong?
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=fc3dec7218b86e76dd15cdfb0ece357e7aad64b5
Or please, could you look to fix it?
Comment 86•6 years ago
|
||
Looks like the only test failing is testAlarmDefaultValue.js. And the linter tells us what the problem is:
TEST-UNEXPECTED-ERROR | /builds/worker/checkouts/gecko/comm/calendar/test/mozmill/testAlarmDefaultValue.js:35:8 | 'content_tab_e' is not defined. (no-undef)
Assignee | ||
Comment 87•6 years ago
|
||
(In reply to Jorg K (GMT+2) from comment #86)
Looks like the only test failing is testAlarmDefaultValue.js. And the linter tells us what the problem is:
TEST-UNEXPECTED-ERROR | /builds/worker/checkouts/gecko/comm/calendar/test/mozmill/testAlarmDefaultValue.js:35:8 | 'content_tab_e' is not defined. (no-undef)
Changed
- ({ content_tab_e, content_tab_eid } = collector.getModule("content-tab-helpers"));
- content_tab_eid = collector.getModule("content-tab-helpers");
But what about the bct errors and the Z2 and Z8 errors in debug?
Comment 88•6 years ago
|
||
Z8 is the same issue, Z2 seems intermittent, and bct is an issue, indeed.
Comment 89•6 years ago
|
||
You need to rerun this, it was totally busted due to bug 1566338.
Comment 90•6 years ago
|
||
I'll take a look at these tests tomorrow. There's a number of things to fix up.
Assignee | ||
Comment 91•6 years ago
|
||
ESLint is now satisfied but the calendar error still exists.
Assignee | ||
Comment 92•6 years ago
|
||
Fixed the not centred "Fonts for:" title and removed a no more needed rule.
Comment 93•6 years ago
|
||
I've fixed the remaining test failures. There's some refactoring that could be done, but as it's test code and it works, I don't really care about that.
Assignee | ||
Comment 94•6 years ago
|
||
Many thanks for fixing the tests, Geoff.
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=df0cf1bcabc975003d0a29708cf7ef7e2c85594b looks good -> c-n
This needs to land together with bug 1565789.
Comment 95•6 years ago
|
||
Sorry, I'm confused here. Does bug 1565789 need go first or after part 5 as Geoff had it in his try?
Bug 1498332 - Fix the tests after the prefs rearrange. r=darktrojan
Bug 1565789 - Update the preferences callers after the prefs redesign in Bug 1498332. r=darktrojan
Bug 1498332 - Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
Bug 1498332 - Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
Bug 1498332 - Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
Bug 1498332 - Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
Bug 1498332 - Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
What about the stufftofix patch, was this merged into the test-fix patch here? I can't see, for example, these changes:
https://hg.mozilla.org/try-comm-central/rev/09bd26f4103b4a598a86af950dc39c84de30722b#l1.12
in the patch.
Comment 96•6 years ago
|
||
Does bug 1565789 need go first or after part 5 as Geoff had it in his try?
If it needs to be shuffled like this, or can it go first, but then the commit message isn't right:
"Update the preferences callers after the prefs redesign in bug 1498332."
The test patch doesn't apply if the patch from bug 1565789 does't come first in some way.
Assignee | ||
Comment 97•6 years ago
|
||
The order of the two bugs doesn't matter. They have to go together to not get test errors.
Comment 98•6 years ago
|
||
Not quite. 1565789 need to be applied or else Geoff's test changes patch doesn't apply. But I'm happy to put it first and change the commit message. Anyway, I'm still unsure what to do with this:
https://hg.mozilla.org/try-comm-central/rev/09bd26f4103b4a598a86af950dc39c84de30722b
since this was part of the try run but doesn't appear to be part of the patch.
Comment 99•6 years ago
|
||
I folded my changes into the last patch and uploaded the combined patch.
Comment 100•6 years ago
|
||
Pushed by mozilla@jorgk.com:
https://hg.mozilla.org/comm-central/rev/3440c68aebee
Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
https://hg.mozilla.org/comm-central/rev/4aac8099a1ed
Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
https://hg.mozilla.org/comm-central/rev/719b71ad6690
Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
https://hg.mozilla.org/comm-central/rev/0795c763e706
Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
https://hg.mozilla.org/comm-central/rev/5f73dd30816c
Redesign and rearrange Thunderbird Preferences/Options similarly to Firefox. r=darktrojan, ui-r=aleca
https://hg.mozilla.org/comm-central/rev/62bf2165d66e
Fix the tests after the prefs rearrange. r=darktrojan
Updated•6 years ago
|
Comment 101•6 years ago
|
||
So now we integrate the Account Manager into this?
Comment 102•6 years ago
|
||
(In reply to Jorg K (GMT+2) from comment #101)
So now we integrate the Account Manager into this?
We should first implement the Search Box, and then move to the Account Manager.
I'm also getting this error since this patch landed whenever I close the preferences:
JavaScript error: resource://gre/modules/SessionStoreFunctions.jsm, line 122: NS_ERROR_FILE_NOT_FOUND:
Is it related or known?
Assignee | ||
Comment 103•6 years ago
|
||
(In reply to Alessandro Castellani (:aleca) from comment #102)
I'm also getting this error since this patch landed whenever I close the preferences:
JavaScript error: resource://gre/modules/SessionStoreFunctions.jsm, line 122: NS_ERROR_FILE_NOT_FOUND:
Is it related or known?
This file is only referenced in m-c. This is more related to bug 1563171 which landed two days ago.
Comment 104•6 years ago
|
||
Removed the AM from the summary since it was misleading.
Comment 105•4 years ago
|
||
Description
•