Summary:
Basic "Fact" application with some storage, part of a daemon, and a control binary.
= Goals =
The general idea is that we have various statistics we'd like to compute, like the frequency of image macros, reviewer responsiveness, task close rates, etc. Computing these on page load is expensive and messy. By building an ETL pipeline and running it in a daemon, we can precompute statistics and just pull them out of "stats" tables.
One way to do this is just to completely hard-code everything, e.g. have a daemon that runs every hour which issues a big-ass query and dumps results into a table per-fact or per fact-group. But this has a bunch of drawbacks: adding new stuff to the pipeline is a pain, various fact aggregators can't share much code, updates are slow and expensive, we can never build generic graphs on top of it, etc.
I'm hoping to build an ETL pipeline which is generic enough that we can use it for most things we're interested in without needing schema changes, and so that installs can use it also without needing schema changes, while still being specific enough that it's fast and we can build useful stuff on top of it. I'm not sure if this will actually work, but it would be cool if it does so I'm starting pretty generally and we'll see how far I get. I haven't built this exact sort of thing before so I might be way off.
I'm basing the whole thing on analyzing entire objects, not analyzing changes to objects. So each part of the pipeline is handed an object and told "analyze this", not handed a change. It pretty much deletes all the old data about that thing and then writes new data. I think this is simpler to implement and understand, and it protects us from all sorts of weird issues where we end up with some kind of garbage in the DB and have to wipe the whole thing.
= Facts =
The general idea is that we extract "facts" out of objects, and then the various view interfaces just report those facts. This change has on type of fact, a "raw fact", which is directly derived from an object. These facts are concerete and relate specifically to the object they are derived from. Some examples of such facts might be:
D123 has 9 comments.
D123 uses macro "psyduck" 15 times.
D123 adds 35 lines.
D123 has 5 files.
D123 has 1 object.
D123 has 1 object of type "DREV".
D123 was created at epoch timestamp 89812351235.
D123 was accepted by @alincoln at epoch timestamp 8397981839.
The fact storage looks like this:
<factType, objectPHID, objectA, valueX, valueY, epoch>
Currently, we supprot one optional secondary key (like a user PHID or macro PHID), two optional integer values, and an optional timestamp. We might add more later. Each fact type can use these fields if it wants. Some facts use them, others don't. For instance, this diff adds a "N:*" fact, which is just the count of total objects in the system. These facts just look like:
<"N:*", "PHID-xxxx-yyyy", ...>
...where all other fields are ignored. But some of the more complex facts might look like:
<"DREV:accept", "PHID-DREV-xxxx", "PHID-USER-yyyy", ..., ..., nnnn> # User 'yyyy' accepted at epoch 'nnnn'.
<"FILE:macro", "PHID-DREV-xxxx", "PHID-MACR-yyyy", 17, ..., ...> # Object 'xxxx' uses macro 'yyyy' 17 times.
Facts have no uniqueness constraints. For @vrana's reviewer responsiveness stuff, we can insert multiple rows for each reviewer, e.g.
<"DREV:reviewed", "PHID-DREV-xxxx", "PHID-USER-yyyy", nnnn, ..., mmmm> # User 'yyyy' reviewed revision 'xxxx' after 'nnnn' seconds at 'mmmm'.
The second value (valueY) is mostly because we need it if we sample anything (valueX = observed value, valueY = sample rate) but there might be other uses. We might need to add "objectB" at some point too -- currently we can't represent a fact like "User X used macro Y on revision Z", so it would be impossible to compute macro use rates //for a specific user// based on this schema. I think we can start here though and see how far we get.
= Aggregated Facts =
These aren't implemented yet, but the idea is that we can then take the "raw facts" and compute derived/aggregated/rollup facts based on the raw fact table. For example, the "count" fact can be aggregated to arrive at a count of all objects in the system. This stuff will live in a separate table which does have uniqueness constraints, and come in the next diff.
We might need some kind of time series facts too, not sure about that. I think most of our use cases today are covered by raw facts + aggregated facts.
Test Plan: Ran `bin/fact` commands and verified they seemed to do reasonable things.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran, majak
Maniphest Tasks: T1562
Differential Revision: https://secure.phabricator.com/D3078
Summary: Some people don't like these, so they should be able to turn them off.
Test Plan:
Toggled the setting on and off; loaded a page in diffusion and differential
that should have symbol cross-references, and saw that they weren't linked
when I had the setting disabled. I also checked that the symbols are still
linked when the setting hasn't been touched.
Reviewers: epriestley, vrana
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3069
Summary: I will need this for tracking line number in Blame previous revision.
Test Plan:
$ hg diff --rev 0:1
$ svn diff -r 64382:64383 README
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3083
Summary:
See title - This simply adds a checkbox to the "Edit User" page in the
admin view, to allow an administrator to re-send the "Welcome to Phabricator"
email.
Test Plan:
Sent myself another welcome email using the checkbox.
Created a new user using the admin panel, to make sure emails still get
sent for new users.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T1524
Differential Revision: https://secure.phabricator.com/D3081
Summary:
We need `$this->old` and `$this->new` in `renderShield()`.
Broken since D2358.
Test Plan: Loaded contents of file with lots of changes.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3080
Summary:
The filename field and the checkbox to select the default image were
overlapping in Firefox on Linux on both the Project Edit page and the
Profile Edit page.
Test Plan: Looked at both of the pages and saw that they rendered better.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3079
Summary: This iterator processes objects that have been updated.
Test Plan:
Ran this test script:
$cursor = null;
$table = new DifferentialRevision();
while (true) {
$iterator = new PhabricatorFactsUpdateIterator($table, $cursor);
foreach ($iterator as $new_cursor => $update) {
echo "{$new_cursor} => D".$update->getID()."\n";
$cursor = $new_cursor;
}
echo "Zzz...\n";
sleep(5);
}
Verified it iterated over every object and then stopped. Made a comment on a differenial revision, verified it iterated over the object after 15 seconds.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1562
Differential Revision: https://secure.phabricator.com/D3077
Summary: PhutilBufferedIterator now implements all the nonspecific logic here.
Test Plan:
Created a test script like this:
$iterator = new LiskMigrationIterator(new DifferentialRevision());
$iterator->setPageSize(3);
foreach ($iterator as $key => $rev) {
echo "{$key}: ".$rev->getID()."\n";
}
Ran it and verified sensible iteration results.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1562
Differential Revision: https://secure.phabricator.com/D3076
Summary: Also move the other tests up so they'll trigger when this stuff is touched.
Test Plan: liberate
Reviewers: nh, btrahan, vrana
Reviewed By: nh
CC: aran
Maniphest Tasks: T1283
Differential Revision: https://secure.phabricator.com/D3074
Summary:
- the current LDAP auth flow expects a DN to look like
cn=ossareh,ou=Users,dc=example,dc=com
- however many LDAP setups have their dn look something like
cn=Mike Ossareh,ou=Users,dc=example,dc=com
Test Plan:
Test if logins work with a LDAP setup which has cn=Full Name
instead of cn=username.
To test you should ensure you set the properties needed to
trigger the search before login as detailed in conf/default.conf.php
Reviewers: epriestley
CC: mbeck, aran, Korvin
Differential Revision: https://secure.phabricator.com/D3072
Summary: Currently, the logout link is this awkward form in the footer since we have to CSRF it. Enable a GET + dialog + confirm workflow instead so the logout link can just be a link instead of this weird mess.
Test Plan: Went to /logout/, logged out.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D3066
Summary: I was poking around to see how this class worked and noticed this variable does nothing.
Test Plan: Careful inspection and reasoning that unused private member variable can be deleted.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3068
Summary:
As we work through @chad's redesign, one thing I want to do is improve the tablet/mobile experience.
Add a "device" behavior which sets a "device-phone", "device-tablet" or "device-desktop" class on the root div. The behavior (device names, width triggers) is mostly based on Bootstrap.
Also adds a preview viewport=meta tag, which makes the iPhone not scale the page like crazy and is a desirable end state, but currently makes the app less usable since things get cut off.
Test Plan:
Added some classes like this:
.device-desktop {
background: blue;
}
.device-tablet {
background: orange;
}
.device-phone {
background: yellow;
}
...and loaded the site on a desktop, iPad and iPhone. Resized the window. Got the right background color in all cases.
Reviewers: btrahan, chad
Reviewed By: chad
CC: aran
Differential Revision: https://secure.phabricator.com/D3063
Summary:
This allows getting stats for any arbitrary period, e.g.
- everything
- last week
- week before last week
Test Plan: Ran the script for last week.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3064
Phabricator requires mod_rewrite rule to emulate "routing"
interface between web server and PHP aplication. Since PHP 5.4 where
is built-in web server that can be invoked with
"PHP -S 127.0.0.1:8000", but since it's very simple it don't have
mod_rewrite functionality. But it have routing functionality if .php
file is given via command-line - so this simple fix allows to
use PHP 5.4+ built-in web server to start Phabricator. Useful for
hacking, developing and testing. Use like this:
"php -S 127.0.0.1:8000 ~/Documents/phabricator/webroot/ ~/Documents/phabricator/webroot/index.php"
Summary:
The final goal is to display reviewers response time on homepage.
This is a building block for it.
The algorithm is quite strict - it doesn't count simple comment as response because reviewers would be able to cheat with comments like "I'm overwhelmed right now and will review next week".
We are more liberate in Phabricator where reviewers response with comments without changing the status quite often but I'm not trying to improve response times in Phabricator so this is irrelevant.
Reviewers in Facebook changes status more often (to clean their queue) so I follow this approach.
There is currently no way to track reviewers silently added and removed in Edit Revision but it's not a big deal.
The algorithm doesn't track commandeered revision, there's a TODO for it.
Response times are put in two buckets: `$reviewed` and `$not_reviewed`.
`$reviewed` contains reviewers who took action, `$not_reviewed` contains reviewers who didn't respond on time.
I will probably compute average time from `$reviewed` and raise it for those `$not_reviewed` that are higher than this average.
The idea is to not favor reviewers who were only lucky for being in a group with someone fast.
Test Plan: Passed test.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3062
Summary: I want to link this page from outside.
Test Plan: /phid/?phids=PHID-USER-gsraq7yc66r4stl4c6u3
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3060
Summary:
Currently, we have this cumbersome `PhabricatorRepositoryCommitMessageDetailParser` hook. This is really old and outdated; I want to just use the Differential custom field parser. See T945 for a specific application.
However, it allows installs to override author/committer association. Instead, provide an event hook for doing this.
Test Plan: Added a listener, made every commit resolve to "turtle", parsed some commits, verified the events looked sane and they now correctly were all attributed to "turtle".
Reviewers: btrahan, vrana, nh
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1337
Differential Revision: https://secure.phabricator.com/D3040
Summary: Currently, MySQL/MySQLi connections store passwords in plain text on the object. Allow them to be stored in PhutilOpaqueEnvelopes instead. See D3053.
Test Plan: Loaded site.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D3054
Summary: We pull "retries" and a doc link from PhabricatorEnv directly. Break these dependencies so the classes can move to libphutil.
Test Plan: Browsed site, triggered a schema exception and verified I still got the useful footer text.
Reviewers: btrahan, vrana
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1283
Differential Revision: https://secure.phabricator.com/D3053
Summary: Add an explicit close() method to connections and call it in GlobalLock.
Test Plan:
Wrote a script like this:
$lock = PhabricatorGlobalLock::newLock('test');
echo "LOCK";
$lock->lock();
sleep(10);
echo "UNLOCK";
$lock->unlock();
sleep(9999);
Using `SHOW FULL PROCESSLIST`, verified the connection closed after 10 seconds with both the "MySQL" and "MySQLi" implementations.
Reviewers: btrahan, vrana
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1470
Differential Revision: https://secure.phabricator.com/D3035
Summary:
The URL /jump/?jump=%s now does the same thing as the POST version,
except that if it couldn't jump to anything, it loads /jump/ with the
query filled in so you can just press enter (we don't save searches
without a CSRF token).
(hsb: Sorry for stealing your task! It hadn't been updated in two months
so I figured you were likely not actively working on it.)
Test Plan:
Loaded given URL with different queries (including various
flavors of nothing). Search worked as expected.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T1036
Differential Revision: https://secure.phabricator.com/D3047
Summary: Our auditors requested displaying this field and I can image that it can be useful.
Test Plan: /audit/
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3044
Summary:
I want to move queryfx() and family to libphutil, for @chad and others (see T1283). We need to break a few dependencies to do this.
Since AphrontWriteGuard is independently useful, I broke the dependency between it and AphrontRequest rather than between Connection and WriteGuard. I'll move its implementation to libphutil in a future diff.
Test Plan: Loaded site, submitted CSRF form successfully, monkeyed with CSRF token, submitted CSRF form, got error.
Reviewers: btrahan, vrana
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1283
Differential Revision: https://secure.phabricator.com/D3042
Summary: I changed this from `getName` to `getFullName` to make attached revisions, etc., render with "Dnnn", but accidentally made all users render as "username (Full Name)". Be a little more surgical in application of full names.
Test Plan: Created a task and attached a CC, a task and a revision. Verified the task and revision rendered with "Tnn", "Dnn" but the CC rendered as "username".
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D3041
Summary: We currently omit email from Git author/committer lookups, which gives us some bad results when identify commit authors. Include email. Also simplify this block a little bit.
Test Plan: Ran "reparse.php --message" on several commits, verified that the author/committer seemed reasonable with var_dump()s.
Reviewers: btrahan, vrana, nh
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1337
Differential Revision: https://secure.phabricator.com/D3039
Summary: This returns a list of PHIDs, not objects which need to be pulled.
Test Plan:
Repro'd fatal locally, verified it was fixed.
Repro steps are:
- Create an arc project associated with a repository, with indexed language(s) and subprojects.
- View a file in that repository.
Reviewers: nh, btrahan
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D3038
Summary:
See D3033, T1529. We currently transform "scp-style" `user@host:path` URIs into normal `ssh://user@host/path` URIs. This is undesirable for two reasons:
- The paths aren't always equivalent. They are for GitHub, which is why I missed this originally, but in the general case the ":path" is resolved relatively and the "/path" is resolved absolutely. So this transformation can break things.
- It confuses users, who do not think of "git@host:path" URIs as SSH URIs even though the SSH protocol is implied.
So stop using them, and just use the "git@host:path" URIs instead. This is a bit messy since we have some validation built up on top of URIs. Hopefully we can get rid of more of this in the future as we simplify repository management.
Test Plan: Unit tests cover this stuff pretty well. Made a new git repository with a "git@host:path" style URI and did pull/discover on it, verified the right URI was used.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1529
Differential Revision: https://secure.phabricator.com/D3036
Summary: MetaMTA + daemons used to be pretty hard but @nh landed some patches a while ago that make it way eaiser. Back off the "ooh scary config" text in the documentation, since this option will just work for ~every install now.
Test Plan: Read it.
Reviewers: btrahan, vrana, nh
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1525
Differential Revision: https://secure.phabricator.com/D3037
Summary: created a PhabricatorInlineCommentPreviewController so controllers in Diffusion and Differential respectively just have to handle the URI mapping and data loading like good little controllers.
Test Plan:
left inline comments on commits, deleted inline commits, submitted inline comments -- all worked well
did the same on some diffs
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T1176
Differential Revision: https://secure.phabricator.com/D3034
Summary:
- The most common workflow complaint I've seen recently is something like "how do I use Differential with a branch full of random code that me and several other developers all commit to"? There are some okay answers ("commandeer") but I think the best answer is "don't do that". Add a document explaining how development works at Facebook (and many other companies) without the use of feature branching, why it's better, and how you can lay the technical groundwork you need to to stop doing this.
- Add a general "smaller commits are better" and "your commit messsage should provide context" document.
- Minor updates to other stuff as my understanding of Mercurial has been refined.
Test Plan: Generated and read documentation.
Reviewers: btrahan, vrana, schrockn
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D3025
Summary: See detailed discussion in T1543.
Test Plan:
- Enabled multiplexing.
- Set user A to "enable Re".
- Created a task owned by user B with user A cc'd.
- Verified A got no "Re:" before this patch.
- Applied patch.
- Verified A got "Re:" after this patch.
Reviewers: nh, btrahan, vrana
Reviewed By: nh
CC: aran
Maniphest Tasks: T1543
Differential Revision: https://secure.phabricator.com/D3031
