Summary: So they're maybe a little easier to deal with? I'm going to take this formally to "plz @chad plz help" land.
Test Plan: {F20329}
Reviewers: btrahan, vrana, chad
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D3609
Summary:
Fact engines loading dependent objects are super slow because they load them one by one.
This diff put each page in a Lisk set allowing engines to use `loadRelatives()`.
It also introduces `clearSet()` method which is somewhat neccessary in PHP < 5.3 or with disabled cyclic [[ http://php.net/gc | GC ]].
Test Plan:
$iterator = new PhabricatorFactUpdateIterator(new DifferentialRevision());
foreach ($iterator as $revision) {
$diffs = $revision->loadRelatives(new DifferentialDiff(), 'revisionID');
echo memory_get_usage() . "\n";
}
Experienced not-steadily-increasing memory usage and much faster loading.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3247
Summary:
- Adds a new "Applications" application.
- Builds an application list via application config instead of via hard-coding, so we can move toward better concepts of installing/uninstalling applications, etc.
- Applications indicate that they need attention with notice counts and brief status messages rathern than 50 giant tables of all sorts of app data.
I want to try replacing the home screen with this screen, pretty much. Not sure if this is totally crazy or not. What does everyone else think?
Test Plan: Will add screenshots.
Reviewers: btrahan, chad, vrana, alanh
Reviewed By: vrana
CC: aran, davidreuss, champo
Maniphest Tasks: T1569
Differential Revision: https://secure.phabricator.com/D3129
Summary:
Firefox has a limit of ~500,000 elements which can be passed in literal array.
This amount of data is meaningless anyway because even Retina displays don't have such resolution.
Limit the amount of data to mitigate browser limitations and also reduce the page size.
Ensure that first and last element is passed.
I considered also reducing the granularity to days but I want new repositories to have nice precise graph.
Test Plan: Displayed the chart in Firefox.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3134
Summary: For any count fact, allow a chart to be drawn. INCREDIBLY POWERFUL DATA ANALYSIS PLATFORM.
Test Plan: Drew a chart of object counts. Drew the Maniphest burn chart.
Reviewers: vrana, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1562
Differential Revision: https://secure.phabricator.com/D3099
Summary:
- Add PhabricatorApplication. This is a general class that I have grand designs for, but used here to allow applications to provide objects for analysis by the facts appliction.
- Add FactCursors, to keep track of where iterators are.
- Make the daemon do something sort of useful.
- Add `bin/fact cursors` for showing and managing objects and cursors.
- Add some options to `bin/fact analyze`.
Test Plan:
- `bin/fact cursors`, `bin/fact cursors --reset DifferentialRevision`, `bin/fact cursors --reset X`
- `bin/fact analyze`, `bin/fact analyze --all`, `bin/fact analyze --iterator DifferentialRevision --skip-aggregates`
- `bin/phd debug fact`
Reviewers: vrana, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1562
Differential Revision: https://secure.phabricator.com/D3098
Summary: Not totally sure about this but I think it's okay?
Test Plan: Loaded /fact/, got a more readable page.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1562
Differential Revision: https://secure.phabricator.com/D3090
Summary:
Some facts are aggregations of other facts. For example, we may compute how many times each macro is used in each object as a "raw fact":
Dnnn uses macro "psyduck" 6 times.
But we want to present this data in aggregate form, e.g. "order macros by popularity". We can do this at runtime and it probably won't be too awful a query, but we can also aggregate it cheaply:
Macro "psyduck" is used 3920 times across all objects.
...and then do a query like "select macros ordered by usage".
"Aggregate" facts support facts like this. The aggregate facts I've implemented are:
- Count of all objects.
- Count of objects of type X.
- Last time facts were updated.
These clearly fit the "aggregate" facts template well. I'm not 100% sure macros do. We can use this table to answer a question like "What are the most popular macros, ordered by use?" We can also use it to answer a question like "What are the most popular macros in the last 6 months?", if we build a specific fact for that. But we can't use it to answer a question like "What are the most popular macros between times X and Y?". Maybe that's important; maybe not.
This seems like a good fit for at least some types of facts.
I'll de-magic the keys a bit in the next diff.
Test Plan: Ran the engines and got some aggregated facts about other facts.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1562
Differential Revision: https://secure.phabricator.com/D3089
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
