Summary:
After D4825, this information is often available to us in a safe way. Provide it explictly.
This removes or reduces functionality in some cases, but I think we can plug those holes with Conpherence addresses and/or explicit user acknowledgement/config.
Test Plan: Patched a commit with `arc patch` and got the original address out.
Reviewers: btrahan, edward, vrana
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D4828
Summary:
Ugh, just wrote out a huge message, only to lose it with a fat-fingered ctrl-c. Le sigh.
First pass at decoupling the bot from the protocol. Noticeably absent is the command/message coupling. After this design pass I'll give that a go. Could use some advice, thinking that handlers should only create messages (which can be public or private) and not open ended, undefined 'commands'. The problem being that there needs to be some consistant api if we want handlers to be protocol agnostic. Perhaps that's a pipedream, what are your thoughts?
Secondly, a few notes, design review requests on the changes i did make:
# Config. For now i'm passing config through to the adapter. This was mainly to remain backwards compatible on the config. I was thinking it should probably be namespaced into it's own subobject though to distinguish the adapter config from the bot config.
# Adapter selection. This flavor is the one-bot-daemon, config specified protocol version. The upside is that in the future they won't have to run different daemons for this stuff, just have different config, and the door is open for multiple protocol adapters down the road if need be. The downside is that I had to rename the daemon (non-backwards compatible change) and there will need to be some sort of runtime evaluation for instatiation of the adapter. For now I just have a crude switch, but I was thinking of just taking the string they supply as the class name (ala `try { new $clasName(); } catch...`) so as to allow for homegrown adapters, but I wasn't sure how such runtime magic would go over. Also, an alternative would be to make the PhabricatorBot class a non-abstract non-final base class and have the adapters be accompanied by a bot class that just defines their adapter as a property. The upside of which is backwards compatibility (welcome back PhabricatorIRCBot) and perhaps a little bit clearer plugin path for homegrowners.
# Logging. You'll notice I commented out two very important logging lines in the irc adapter. This isn't intended to remain commented out, but I'm not sure what the best way is to get logging at this layer. I'm wary of just composing the daemon back down into the adapter (bi-directional object composition makes my skin crawl), but something needs to happen, obviously. Advice?
That's it. After the feedback on the above, you can either merge down, or wait until i finish the command/message refactor if you don't think the diff will grow too large. Up to you, this all functions as is.
Test Plan: Ran an irc bot, connected, read input, and wrote output including handler integration.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2462
Differential Revision: https://secure.phabricator.com/D4757
Summary:
In the second phase, I want to get rid of the most of `phutil_escape_html()` calls in favor of plain strings or `PhutilSafeHTML`.
This is an example of how it could look.
Test Plan: /api/user.whoami
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2432
Differential Revision: https://secure.phabricator.com/D4823
Summary: do so via event engine. note different order now...
Test Plan: toggled "show beta applications" to off and noted that Conpherence disappeared. Otherwise noted that links showed.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2424
Differential Revision: https://secure.phabricator.com/D4708
Summary: i think the DOM changed in conpherence with the menu upgrades. just noticed that when you select a new conpherence the old one is not de-selected. this fixes it by updating the javascript to ascend one node higher and then use DOM.scry with the right data sigil to get the nodes
Test Plan: read some messages, noted only the one I was reading had the entry highlighted in the left.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2399
Differential Revision: https://secure.phabricator.com/D4819
Summary:
Fixes T2482. After D4792, menus have more formal structure, but previously we were just shoving some `<div>` into the middle of the thing. This no longer works correctly, since we end up with `<div class="nice-formal-div"><div></div>`.
Just put IDs on all the items we're going to show/hide instead so we don't have to render any half-tags.
Test Plan: Home page show/hide works again.
Reviewers: btrahan, chad
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2482
Differential Revision: https://secure.phabricator.com/D4810
Summary: this broke when I moved sorting to the editor. Turns out you can't have custom hooks for the comment transaction, and thus I couldn't update participation inside the editor. This fixes this by removing the empty switch statement for the transaction type inside the parent class editor. This should have no effect other than fixing Conpherence. Note that conpherences will need another message, etc for a given conphernece to fix itself.
Test Plan: Conpherence threads were updated properly!
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2399
Differential Revision: https://secure.phabricator.com/D4817
Summary:
this was originally "just" adding the icons like I had bundled into D4790. It morphed a bit though and does a few things
- adds the icons
- cleans up widget CSS generally a bit so scrolling always works
- phutil_tag -- probably was a bad idea but I wanted to play with it. I think its harder to not break conpherence when you land the branch now maybs. Still up for fixing it immediately post land though.
Test Plan: played with conphernece a bit. Used FF and Chrome to verify CSS was looking okay-ish.
Reviewers: epriestley, chad
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2399
Differential Revision: https://secure.phabricator.com/D4814
Summary: T2326 tells the tale. this is the fix.
Test Plan: verified that queries that shouldn't be sortable weren't. also had a phlog in there a bit to sanity check things faster
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2326
Differential Revision: https://secure.phabricator.com/D4816
Summary:
I attempted to test every interface. I probably missed some stuff, but I at least have some level of confidence that the `phutil_tag` branch is fairly stable.
Fixed these issues:
[1] Fixed a Herald issue with object links in transcripts. Some actions return
links; this was previously hard-coded.
[2] DarkConsole refactoring created an issue where the "`" event handler registered too many times.
[3] Fixed a bug where `strlen($value)` was being checked, but fields may now return array(). Possibly we should implement phutil_is_empty_html() or similar.
[4] Fixed a undefined variable issue for image edit transactions.
[5] Fixed an issue with rendering participant transactions. This required phutil_safe_html() because `pht()` can't handle `array()` for `%s`.
[6] Fixed an issue where feed was entirely overescaped by reverting an overly ambitious render_tag -> tag.
[7] Fixed an issue with strict tables and inserting `''` instead of `0` into an integer column.
[8] Fixed an issue where • was shown escaped.
[9] Fixed an issue where "no data" was overescaped.
[10] Fixed an issue with strict tables and inserting `''` instead of `0` into an integer column.
[11] Fixed an issue with strict tables and inserting `''`.
[12] Fixed an issue with missing space after ":" for mini panels.
Encountered (but did not fix) these issues:
[X1] "e" works incorrectly on comments you are not allowed to edit. Did not fix.
[X2] Coverage currently interacts incorrectly with "--everything" for Phutil tests.
Test Plan:
- Viewed Differential.
- Created a diff via copy/paste.
- Viewed standalone diff.
- Jumped to diff via changeset table.
- Created a revision.
- Updated revision.
- Added a comment.
- Edited revision dependencies.
- Edited revision tasks.
- Viewed MetaMTA transcripts.
- Viewed Herald transcripts [1].
- Downloaded raw diff.
- Flagged / unflagged revision.
- Added/edited/deleted inline comment.
- Collapsed/expanded file.
- Did show raw left.
- Did show raw right.
- Checked previews for available actions.
- Clicked remarkup buttons
- Used filetree view.
- Used keyboard: F, j, k, J, K, n, p, t, h, "?" [2] [X1].
- Created a meme.
- Uploaded a file via drag and drop.
- Viewed a revision with no reviewers.
- Viewed a revision with >100 files.
- Viewed various other revisions [3].
- Viewed an image diff.
- Added image diff inline comments.
- Viewed Maniphest.
- Ran various queries.
- Created task.
- Created similar task.
- Added comments to tasks.
- Ran custom query.
- Saved custom query.
- Edited custom queries.
- Drag-reordered tasks.
- Batch edited tasks.
- Exported tasks to excel.
- Looked at reports (issue in T2311 notwithstanding).
- Viewed Diffusion.
- Browsed Git, SVN, HG repositories.
- Looked at history, browse, change, commit views.
- Viewed audit.
- Performed various audit searches.
- Viewed Paste.
- Performed paste searches.
- Created, edited, forked paste.
- Viewed Phriction.
- Edited a page.
- Viewed edit history.
- Used search typeahead to search for user / application.
- Used search to search for text.
- Viewed Phame.
- Viewed Blog, Post.
- Viewed live post.
- Published/unpublished post.
- Previewed post.
- Viewed Pholio.
- Edited/commented mock.
- Viewed ponder.
- Viewed question.
- Added answer/comment.
- Viewed Diviner.
- Viewed Conpherence [4] [5].
- Made Conpherence updates.
- Viewed calendar.
- Created status.
- Viewed status.
- Viewed Feed [6].
- Viewed Projects.
- Viewed project detail.
- Edited project.
- Viewed Owners.
- Viewed package detail.
- Edited package [7].
- Viewed flags.
- Edited flag.
- Deleted flag.
- Viewed Herald.
- Viewed rules.
- Created rule.
- Edited rule.
- Viewed edit log.
- Viewed transcripts.
- Inspected a transcript.
- Viewed People.
- Viewed list.
- Administrated user.
- Checked username/delete stuff.
- Looked at create/import LDAP/activity logs.
- Looked at a user profile.
- Looked at user about page.
- Looked at Repositories.
- Edited repository.
- Edited arcanist project.
- Looked at daemons.
- Looked at all daemons [8].
- Viewed combined log.
- Looked at configuration.
- Edited configuration.
- Looked at setup issues [9].
- Looked at current settings.
- Looked at application list.
- Installed / uninstalled applications [10].
- Looked at mailing lists.
- Created a mailing list.
- Edited a mailing list.
- Looked at sent mail.
- Looked at received mail.
- Looked at send/receive tests.
- Looked at settings.
- Clicked through all the panels.
- Looked at slowvote.
- Created a slowvote [11].
- Voted in a slowvote.
- Looked at Macro.
- Created a macro.
- Edited a macro.
- Commented on a macro.
- Looked at Countdown.
- Created a Countdown.
- Looked at it.
- Looked at Drydock.
- Poked around a bit.
- Looked at Fact.
- Poked around a bit.
- Looked at files.
- Looked at a file.
- Uploaded a file.
- Looked at Conduit.
- Made a Conduit call.
- Looked at UIExamples.
- Looked at PHPAST.
- Looked at PHIDs.
- Looked at notification menu.
- Looked at notification detail.
- Logged out.
- Logged in.
- Looked at homepage [12].
- Ran `arc unit --everything --no-coverage` [X2].
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Maniphest Tasks: T2432
Differential Revision: https://secure.phabricator.com/D4807
Summary:
See D4812.
- This preference disables the file tree completely.
- It defaults off, so users who want it will have to go turn it on.
- Maybe slightly cleaner would be doing this if the tree was hidden and then ajaxing it in if you press "F", but that's complicated and I don't want to bother.
- Generally, I think this element is useful to something like 5% of users and not useful to 95%.
Test Plan: Enabled and disabled file tree. Looked at commits and revisions; verified they reflected the setting correctly.
Reviewers: vrana
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D4813
Summary: This is the most requested feature in FB by far.
Test Plan:
Toggled, verified that data are saved.
Reloaded, toggled, toggled, toggled, verified that data are saved.
Reloaded.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4812
Summary: Add a field where you can put the gravatar email address to pull an image for the profile picture from
Test Plan: Tried uploading a file, replacing with default, and various combinations and they all still work.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2105
Differential Revision: https://secure.phabricator.com/D4809
Summary: Remove css class that was setting static width
Test Plan: Loaded user profile edit page, stretched browser around, saw that the text was happy
Reviewers: epriestley, btrahan
Reviewed By: btrahan
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4806
Summary:
We are doing a better job in 1) of D3069#3 and 2) is just dumb.
Let's see if someone notices this change.
Test Plan: /settings/panel/display/
Reviewers: nh, epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4789
Summary:
Convert a final few `render_tag()` calls to `tag()` calls. This leaves us with 36 calls:
- 9 are in Conpherence, and will be converted after merging. Using Views makes the most sense here, to get access to renderHTMLView() and such.
- 2 are the definition and its library map entry.
- 3 are in the documentation.
- I believe the remaining 22 are too difficult to convert pre-merge. About half are in code which is slated for destruction; the other half are in the base implementations of very common classes (like PhabricatorStandardPageView) and can only be converted by converting the entire codebase.
My plan at this point is:
- Test the branch thoroughly.
- Merge to master.
- Over time, resolve the remaining issues: lint means things shouldn't get any worse.
Test Plan: See inlines.
Reviewers: vrana
Reviewed By: vrana
CC: aran, edward
Maniphest Tasks: T2432
Differential Revision: https://secure.phabricator.com/D4802
Test Plan: Ran the migration on a single revision, verified DB, called `loadUnsubscribedPHIDs()`.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4786
Summary: Suggest the MySQL mode STRICT_ALL_TABLES during setup if it is not set. Small improvement to the phabricator.developer-mode comments.
Test Plan: Set the global sql_mode to include or exclude STRICT_ALL_TABLES and check for desired behavior.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4803
Summary:
Make `PhabricatorMenuView` more flexible, so callers can add items to the beginning/end/middle.
In particular, this allows event handlers to receive a $menu and call `addMenuItemToLabel('activity', ...)` or similar, for D4708.
Test Plan: Unit tests. Browsed site. Home page, Conpherence, and other pages with menus look correct.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D4792
Summary:
By default, order applications in application order. See discussion in D4708.
Principally, this is intended to make sure that application event handlers are registered in order, and thus fire in order.
Test Plan:
Looked at /applications/, homepage tiles, verified they both still work.
I didn't actually test the event handler bit since it's fairly complicated to test blind; D4708 should provide a test case.
Reviewers: btrahan, Afaque_Hussain
Reviewed By: Afaque_Hussain
CC: aran
Differential Revision: https://secure.phabricator.com/D4791
Test Plan:
Looked at file with lint errors in Diffusion.
I've also tried inline comments in Differential but it failed.
I'll try it again after you land all your diffs.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2432
Differential Revision: https://secure.phabricator.com/D4755
Summary: Regression to original behavior.
Test Plan: Clicked on it twice, didn't see confirmation dialog.
Reviewers: epriestley, codeblock
Reviewed By: codeblock
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4788
Summary: Applied fixes for issues mentioned in D4737#1&2
Test Plan: Verified that scripts seem to work as intended.
Reviewers: epriestley
CC: aran, Korvin
Maniphest Tasks: T2446
Differential Revision: https://secure.phabricator.com/D4782
Summary:
Try to lock the screen to whatever the user is looking at as we load changesets.
Notably, this improves the use case of taking a known action on a diff. Currently, you have to wait for everything to load or the comments keep getting scrolled down. After this change, the comments stay in the same place on screen.
Test Plan:
Raised the autoload changeset limit from 100 to 1000, looked at a 220 changeset diff.
- Reloaded it while scrolled at the top; normal behavior (no scrolling).
- Reloaded it, scrolled to the bottom. Comment area now stable.
- Reloaded it, kind of scrolled around the middle? Behavior seemed stable/reasonable. This one is kind of heursitic so it's hard to say I'm getting it totally right or not, but it's less important than the "bottom" case.
Reviewers: vrana, btrahan, chad, dctrwatson
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D4774
Summary: Refactor options related to verbose error reporting and forcing disk reads into a single developer option.
Test Plan: Run Phabricator with the developer-mode option set and check that errors print stack traces, static assets are always reloaded, etc.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4780
Summary:
Added isNormalMouseEvent() that returns true if left mouse button triggered event click, mousedown or mouseup.
Modified isNormalMouseClick() to use new function.
Test Plan: Verified that new function works for click, mousedown and mouseup events.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2475
Differential Revision: https://secure.phabricator.com/D4778
Summary: Fixed T2398
Test Plan: Ran a local test. It looked a tad better.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2398
Differential Revision: https://secure.phabricator.com/D4779
Summary: Write documentation about how to use phutil_tag(), etc., safely.
Test Plan:
Read carefully.
{F31223}
Reviewers: vrana, btrahan, chad
Reviewed By: vrana
CC: aran
Maniphest Tasks: T2432
Differential Revision: https://secure.phabricator.com/D4769
Summary: Fixed T2395
Test Plan: When impact.ttf was added to resources/font, it was being used. When renamed, tuffy was.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2395
Differential Revision: https://secure.phabricator.com/D4700
Summary: Also convert to `phutil_tag()`.
Test Plan: Displayed revision with hidden comments.
Reviewers: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4772
Summary:
I missed these in review, but here are a couple of tweaks:
- Call `setWorkflow(true)` on the actions. This makes the dialogs pop up on the same page with Javascript if it's available.
- When the user installs/uninstalls an application, send them back to the application's detail page, not the application list.
Test Plan:
- Uninstalled an application (saw dialog, got sent back to detail page).
- Installed an application (saw dialog, got sent back to detail page).
- Canceled an application uninstall.
Reviewers: Afaque_Hussain
Reviewed By: Afaque_Hussain
CC: aran
Differential Revision: https://secure.phabricator.com/D4762
Summary: If you render multiple tags, they squish together.
Test Plan:
Now they look a little better:
{F31191}
Reviewers: Afaque_Hussain
Reviewed By: Afaque_Hussain
CC: aran
Differential Revision: https://secure.phabricator.com/D4761
Summary: It's a little bit confusing that you couldn't use %d but kind of expected taken that the number will be formatted so it's not a number anymore.
Test Plan: /paste/
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4764
