Summary:
- Fixes T2257. We wrote a 0-length file (erroneously?) and currently throw when retrieving it. This also happens if you intentionally upload an empty file. I'm not sure what happened with the image: we check for errors during the write, so its existence implies S3 told us the write was successful and then lost the data. Since this is a one-off, I'm not too worried about it. The indistinguishable case of an actually empty file is fixed, at least.
- Writes to a directory like "phabricator/ab/cd/efgh" instead of "phabricator/abcdefgh". When I had to go look for the file on S3 it took a few minutes of scrolling since the web interface isn't very fast. Make it so a file can be located by navigating through pieces of the hash.
Test Plan: Viewed an empty file, no fatal. Viewed the file from T2257 locally, no fatal (no data either, but it's gone). Uploaded a file, saw a nice path.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2257
Differential Revision: https://secure.phabricator.com/D4303
Summary: We have enough z-index rules that they're fairly hard to visualize with "git grep". Consolidate them. Then fix T2253 (missing z-index on left menu background).
Test Plan: Made a Differential window really narrow, then scrolled it horizontally.
Reviewers: btrahan, chad, ender
Reviewed By: chad
CC: aran
Maniphest Tasks: T2253
Differential Revision: https://secure.phabricator.com/D4302
Summary:
This is basicaly a light version of D4286. The major problem with D4286 is that it's a huge leap and completely replaces the setup process in one step.
Instead, I want to do this:
- Add the post-setup warnings (yellow bar with "6 unresolved warnings...").
- Copy all setup checks into post-setup warnings (so every check has an old-style check and a new-style check).
- Run that for a little bit and make sure it's stable.
- Implement fatal post-setup checks (the red screen, vs the yellow bar).
- Run that for a little bit.
- Nuke setup mode and delete all the old checks.
This should give us a bunch of very gradual steps toward the brave new world of simpler setup.
Test Plan:
- Faked APC setup failures, saw warnings raise.
- Verified that this runs after restart (get + set).
- Verified that this costs us only one cache hit after first-run (get only).
Reviewers: btrahan, codeblock, vrana, chad
Reviewed By: codeblock
CC: aran
Maniphest Tasks: T2228
Differential Revision: https://secure.phabricator.com/D4295
Summary:
See discussion in T2221. Before we can move configuration to the database, we have a bootstrapping problem: we need database credentials to live //somewhere// if we can't guess them (and we can only really guess localhost / root / no password).
Some options for this are:
- Have them live in ENV variables.
- These are often somewhat unfamiliar to users.
- Scripts would become a huge pain -- you'd have to dump a bunch of stuff into ENV.
- Some environments have limited ability to set ENV vars.
- SSH is also a pain.
- Have them live in a normal config file.
- This probably isn't really too awful, but:
- Since we deploy/upgrade with git, we can't currently let them edit a file which already exists, or their working copy will become dirty.
- So they have to copy or create a file, then edit it.
- The biggest issue I have with this is that it will be difficult to give specific, easily-followed directions from Setup. The instructions need to be like "Copy template.conf.php to real.conf.php, then edit these keys: x, y, z". This isn't as easy to follow as "run script Y".
- Have them live in an abnormal config file with script access (this diff).
- I think this is a little better than a normal config file, because we can tell users 'run phabricator/bin/config set mysql.user phabricator' and such, which is easier to follow than editing a config file.
I think this is only a marginal improvement over a normal config file and am open to arguments against this approach, but I think it will be a little easier for users to deal with than a normal config file. In most cases they should only need to store three values in this file -- db user/host/pass -- since once we have those we can bootstrap everything else. Normal config files also aren't going away for more advanced users, we're just offering a simple alternative for most users.
This also adds an ENVIRONMENT file as an alternative to PHABRICATOR_ENV. This is just a simple way to specify the environment if you don't have convenient access to env vars.
Test Plan: Ran `config set x y`, verified writes. Wrote to ENVIRONMENT, ran `PHABRICATOR_ENV= ./bin/repository`.
Reviewers: btrahan, vrana, codeblock
Reviewed By: codeblock
CC: aran
Maniphest Tasks: T2221
Differential Revision: https://secure.phabricator.com/D4294
Summary:
* When we restored to the default value, we did, in fact delete the row from the
database, but then a few lines later down, we saved it again. This patch causes
the controller to return early on delete, like it was supposed to do to begin
with.
* When checking the user's input value for `null` (since PHP's JSON encoder will
return `null` on failure), check the value that the user gave, not the value
that we default to (which is often `null` anyway). Oops.
Test Plan:
* Saved an empty text field and saw the delete work properly and NOT get
re-added.
* Put `null` in the text field, and saved successfully.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4300
Summary:
The placeholder property remains null despite it is set to a custom
value.
Test Plan:
Use a custom placeholder in `AphrontFormTokenizerControl`
```
id(new AphrontFormTokenizerControl())
->setPlaceholder('My custom placeholder...')
->setDatasource('/typeahead/common/projects/');
```
The placeholder should be set to "My custom placeholder..."
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4297
Summary:
As mentioned by @epriestley in an inline on D4290, we should show what happens
if the user leaves the box blank.
Test Plan: Went to edit a setting and saw the default below the text box.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin, asherkin
Differential Revision: https://secure.phabricator.com/D4293
Summary:
This is somewhat clowny, particularly in how it handles JSON encode/decode, but
I've commented why I did things the way I did. The goal is to store minified JSON
but show pretty-printed JSON where possible, to the user editing it.
Test Plan:
* Went to /config/ and saw a list of keys from the `default` config.
* Clicked on one of them, submitted the default value successfully.
* Changed the value to invalid JSON and got a decent error.
* Changed the value to valid JSON and checked the DB to confirm it saved.
* Confirmed the DB values were minified.
* Confirmed the user-facing values were pretty-printed where they could be.
* Confirmed that PHIDs were getting assigned properly and that isDeleted
properly defaulted to false/0.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2246
Differential Revision: https://secure.phabricator.com/D4290
Summary:
This replaces D4042 and gives you the option to search for existing Phrication
documents only.
Test Plan:
# Reindexed ALL THE THINGS, saw existing documents show up in search in all cases.
# Deleted a document and saw it only show up when "Open and Closed Documents" are
shown.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4289
Summary: Currently, we have a configuration stack for unit tests, but they're built in to `PhabricatorEnv`. Pull them out and formalize them, so we can add more configuration sources (e.g., database).
Test Plan: Ran unit tests, web requests, scripts. This code had fairly good existing test coverage.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2223, T2221
Differential Revision: https://secure.phabricator.com/D4284
Summary:
In the past, we did some additional magic on `$response_string` (adding profiling headers? Or DarkConsole?), so we could not share the pathway with HTTPSink. We no longer do this; share the pathways.
Also remove error handler initialization (duplicated in PhabricatorEnv), and move $sink initialization earlier. My general goal here is to allow PhabricatorSetup to emit a normal Response object and share as much code as possible with normal pages.
Test Plan: Loaded page.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2228
Differential Revision: https://secure.phabricator.com/D4285
Summary:
We have a bunch of code duplication now between __init_script__.php and webroot/index.php. Consoldiate these methods and move them into PhabricatorEnv.
Merge PhabricatorRequestOverseer into PhabricatorStartup.
Test Plan: Loaded page, ran script. Wiped PHABRICATOR_ENV; loaded page, ran script; got errors.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2223
Differential Revision: https://secure.phabricator.com/D4283
Summary:
We have a lot of mess to get through before we can load libphutil and enter Phabricator code properly. Move it to a dedicated class.
I'm probably going to merge PhabricatorRequestOverseer into this, although the check that lives there now is kind of weird. It also does not really need to be a pre-load check and could be handled better.
I stopped shoving stuff in here once I got to ENV stuff, I'm going to tackle that next.
Test Plan: Ran phabricator normally; introduced fatals and misconfigurations. Grepped for changed symbols.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran, asherkin
Maniphest Tasks: T2223
Differential Revision: https://secure.phabricator.com/D4282
Summary:
See T2062. This cache allows us to essentially implement this sort of block:
if (this_code_has_not_run_since_the_last_server_restart()) {
...
}
This will let us do setup checks automatically (i.e., without a specialized setup mode) without imposing hundreds of milliseconds of `git submodule status` and similar checks on every page load, even if an install does not have APC.
Broadly, the major goals here are:
- Reduce user errors and support costs related to misconfiguration (e.g., failure to update submodules).
- Simplify setup and configuration (remove 'phabricator.setup', remove/reduce PHABRICATOR_ENV).
- Move as much configuration to the web as possible (required for SaaS).
Test Plan:
Added this block to webroot/index.php:
$cache = PhabricatorCaches::getSetupCache();
$result = $cache->getKeys(array('x'));
if (empty($result['x'])) {
phlog('Cache miss + set.');
$cache->setKeys(array('x' => 'y'));
} else {
phlog('Cache hit.');
}
Verified it used APC correctly.
Disabled APC and verified it degraded to a reasonable disk-based behavior.
If we miss both of these we end up with no actual caching, but that's the best we can do. This code will also run too early in setup for it to be appropriate to raise exceptions out of this pathway -- later on, we can raise a warning that APC is not installed.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2227, T2062
Differential Revision: https://secure.phabricator.com/D4281
Test Plan: Clicked on next month, didn't see year 20121.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4277
Summary:
This fixes two separate issues:
# `getTextStatus()` is used for machine readable data in handles and user.info method. Broken since D3810.
# Status may contain date. Broken since beginning but masked by the fact that CSS ignores unknown class names.
Test Plan:
Displayed revision with reviewer away.
Called `user.addstatus`.
Edited status in calendar.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
CC: nh, aran, Korvin
Differential Revision: https://secure.phabricator.com/D4275
Summary:
The search indexing API has several problems right now:
- Always runs in-process.
- It would be nice to push this into the task queue for performance. However, the API currently passses an object all the way through (and some indexers depend on preloaded object attributes), so it can't be dumped into the task queue at any stage since we can't serialize it.
- Being able to use the task queue will also make rebuilding indexes faster.
- Instead, make the API phid-oriented.
- No uniform indexing API.
- Each "Editor" currently calls SomeCustomIndexer::indexThing(). This won't work with AbstractTransactions. The API is also just weird.
- Instead, provide a uniform API.
- No uniform CLI.
- We have `scripts/search/reindex_everything.php`, but it doesn't actually index everything. Each new document type needs to be separately added to it, leading to stuff like D3839. Third-party applications can't provide indexers.
- Instead, let indexers expose documents for indexing.
- Not application-oriented.
- All the indexers live in search/ right now, which isn't the right organization in an application-orietned view of the world.
- Instead, move indexers to applications and load them with SymbolLoader.
Test Plan:
- `bin/search index`
- Indexed one revision, one task.
- Indexed `--type TASK`, `--type DREV`, etc., for all types.
- Indexed `--all`.
- Added the word "saboteur" to a revision, task, wiki page, and question and then searched for it.
- Creating users is a pain; searched for a user after indexing.
- Creating commits is a pain; searched for a commit after indexing.
- Mocks aren't currently loadable in the result view, so their indexing is moot.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: 20after4, aran
Maniphest Tasks: T1991, T2104
Differential Revision: https://secure.phabricator.com/D4261
Summary:
See discussion in D4204. Facebook currently has a 314MB remarkup cache with a 55MB index, which is slow to access. Under the theory that this is an index size/quality problem (the current index is on a potentially-384-byte field, with many keys sharing prefixes), provide a more general index with fancy new features:
- It implements PhutilKeyValueCache, so it can be a component in cache stacks and supports TTL.
- It has a 12-byte hash-based key.
- It automatically compresses large blocks of data (most of what we store is highly-compressible HTML).
Test Plan:
- Basics:
- Loaded /paste/, saw caches generate and save.
- Reloaded /paste/, saw the page hit cache.
- GC:
- Ran GC daemon, saw nothing.
- Set maximum lifetime to 1 second, ran GC daemon, saw it collect the entire cache.
- Deflate:
- Selected row formats from the database, saw a mixture of 'raw' and 'deflate' storage.
- Used profiler to verify that 'deflate' is fast (12 calls @ 220us on my paste list).
- Ran unit tests
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D4259
Summary: This data structure is a `dict<int, list<Comment>>` now, where the `int` is the line number.
Test Plan:
- Created a diff changing an image.
- Added inline comments on the left and right sides of the diff.
- Saw some exceptions and general sadness.
- Applied patch.
- Reloaded page.
- Everything worked great.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D4264
Summary: D4270 missed a spot I think.
Test Plan: Called it.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4271
Summary: D4249 missed a spot I think.
Test Plan: no more fatal
Reviewers: vrana, epriestley
Reviewed By: vrana
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4270
Summary: basically when we're on a page we have a URIPath so set that and pass it over. I wasn't super duper happy with this but it seems to be the best way to pass this data.
Test Plan: verified the og:url was correct on my dev instance of phacility blog for both the main blog and individual post view
Reviewers: epriestley
Reviewed By: epriestley
CC: chad, aran, Korvin
Differential Revision: https://secure.phabricator.com/D4269
Summary: this makes it more sensical when you hit "share" from a bookmarklet or cut and paste a link into FB, basically by having post-specific data when sharing a post.
Test Plan: looked at generated HTML on my test blog
Reviewers: epriestley
Reviewed By: epriestley
CC: chad, aran, Korvin
Differential Revision: https://secure.phabricator.com/D4266
Summary: Sets any event text color to white, reguardless if its an anchor or not.
Test Plan: Have events from multiple people in the calendar, view events.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4263
Summary:
When previewing, save drafts. When loading objects, restore drafts if they are available.
Depends on: D665
Test Plan:
- Viewed a Mock.
- Typed text into the comment box.
- Reloaded the page.
- Text still there.
- Hit submit, got my comment.
- Reloaded the page.
- Draft correctly deleted.
- Repeated for Macros.
Reviewers: btrahan, chad, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2104
Differential Revision: https://secure.phabricator.com/D4252
Summary:
Implements previews for Macros and Pholio.
(Design is nonfinal -- kind of split the difference between `diff_full_view.png`, laziness, and space concerns. Next couple diffs will add more stuff here.)
Test Plan: {F28055}
Reviewers: btrahan, chad
Reviewed By: btrahan
CC: aran, vrana
Maniphest Tasks: T2104
Differential Revision: https://secure.phabricator.com/D4246
Summary: Does this seem reasonable? It's a bit more compact than digest() (6 bits / byte instead of 4 bits / byte) and 72 bits of entropy @ 12 bytes instead of 128 bits of entropy @ 32 bytes. I feel like it's important to preserve the printability, though, and this seemed like a fairly good balance of concerns.
Test Plan: unit tests
Reviewers: vrana
Reviewed By: vrana
CC: aran, yemao932
Differential Revision: https://secure.phabricator.com/D4253
Summary: Fixes T2210. Recently, we require unique keys on menu items, but it's currently possible in Maniphest to save the same custom query under multiple names. Avoid exploding in this case (we'll hide the duplicates). This isn't a great fix, but makes Maniphest usable again.
Test Plan: Saved the same query twice, laoded page, got exception, applied patch, loaded page, saw duplicate query stripped.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2210
Differential Revision: https://secure.phabricator.com/D4247
Summary: We use "D<id>" for revisions.
Test Plan: Looked at revision.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4258
Summary: 'cuz new fluid layouts require the westerlyness. Looks like D4126 started the N and W implementation but didn't finish it...? note I had to do the shifting of the 5 pixels in javascript; using the CSS didn't work for me in chrome.
Test Plan: uiexample, and hoping it goes well when deployed in prod for differential case
Reviewers: epriestley
Reviewed By: epriestley
CC: chad, aran, Korvin
Maniphest Tasks: T2211
Differential Revision: https://secure.phabricator.com/D4257
Summary: This is used in every other view.
Test Plan: Browsed around.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4248
Summary:
I messed up D4244 (and didn't test it properly) - this change should load
the raw content so we can use it.
Test Plan: use the conduit console and actually test the paste conduit methods
Reviewers: vrana, epriestley, btrahan
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4245
Summary:
The caching introduced by rP7e37eb48273eef87e6e6811703fb5d85a3e07a81
broke the use of PhabricatorPaste::getContent() for forking/editing
pastes and downloading pastes with arc paste. I think this fixes all
the appropriate callsites.
Test Plan: fork a paste in the web ui
Reviewers: vrana, epriestley, btrahan
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4244
Summary:
See f5c2a2ab4b (commitcomment-2333247)
Copy of working implementation from PHPMailerLite.
Also expose the SSL/TLS options.
Test Plan: Switched to this mailer, configured Gmail SMTP, sent email. Verified email arrived intact.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran, mbeck
Differential Revision: https://secure.phabricator.com/D4239
Summary: We allow user to specify the host to listen to but we listen on 127.0.0.1 instead.
Test Plan:
Hardcoded the path and verified that I can connect to the host from other machine.
Verified that `localhost` still doesn't allow remote connections.
Reviewers: ddfisher, epriestley, nh
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4238
Summary:
- Original command is in SSH_ORIGINAL_COMMAND, not normal argv.
- Use PhutilShellLexer to parse it.
- Fix a protocol encoding issue with ConduitSSHWorkflow. I think I'm going to make this protocol accept multiple commands anyway because SSH pipes are crazy expensive to build (even locally, they're ~300ms).
Test Plan: With other changes, successfully executed "arc list --conduit-uri=ssh://localhost:2222".
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T550
Differential Revision: https://secure.phabricator.com/D4232
Summary:
- Build "sshd-auth" (for authentication) and "sshd-exec" (for command execution) binaries. These are callable by "sshd-vcs", located [[https://github.com/epriestley/sshd-vcs | in my account on GitHub]]. They are based on precursors [[https://github.com/epriestley/sshd-vcs-glue | here on GitHub]] which I deployed for TenXer about a year ago, so I have some confidence they at least basically work.
- The problem this solves is that normally every user would need an account on a machine to connect to it, and/or their public keys would all need to be listed in `~/.authorized_keys`. This is a big pain in most installs. Software like Gitosis/Gitolite solve this problem by giving you an easy way to add public keys to `~/.authorized_keys`, but this is pretty gross.
- Roughly, instead of looking in `~/.authorized_keys` when a user connects, the patched sshd instead runs `echo <public key> | sshd-auth`. The `sshd-auth` script looks up the public key and authorizes the matching user, if they exist. It also forces sshd to run `sshd-exec` instead of a normal shell.
- `sshd-exec` receives the authenticated user and any command which was passed to ssh (like `git receive-pack`) and can route them appropriately.
- Overall, this permits a single account to be set up on a server which all Phabricator users can connect to without any extra work, and which can safely execute commands and apply appropriate permissions, and disable users when they are disabled in Phabricator and all that stuff.
- Build out "sshd-exec" to do more thorough checks and setup, and delegate command execution to Workflows (they now exist, and did not when I originally built this stuff).
- Convert @btrahan's conduit API script into a workflow and slightly simplify it (ConduitCall did not exist at the time it was written).
The next steps here on the Repository side are to implement Workflows for Git, SVN and HG wire protocols. These will mostly just proxy the protocols, but also need to enforce permissions. So the approach will basically be:
- Implement workflows for stuff like `git receive-pack`.
- These workflows will implement enough of the underlying protocol to determine what resource the user is trying to access, and whether they want to read or write it.
- They'll then do a permissons check, and kick the user out if they don't have permission to do whatever they are trying to do.
- If the user does have permission, we just proxy the rest of the transaction.
Next steps on the Conduit side are more simple:
- Make ConduitClient understand "ssh://" URLs.
Test Plan: Ran `sshd-exec --phabricator-ssh-user epriestley conduit differential.query`, etc. This will get a more comprehensive test once I set up sshd-vcs.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T603, T550
Differential Revision: https://secure.phabricator.com/D4229
