In profiling integration tests, I found a couple places where per-test
overhead could be reduced:
* Avoiding disk IO by synchronizing instead of deleting & copying test
Git repository data. This saves ~100ms per test on my machine
* When flushing queues in `PrintCurrentTest`, invoke `FlushWithContext`
in a parallel.
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
This is a large and complex PR, so let me explain in detail its changes.
First, I had to create new index mappings for Bleve and ElasticSerach as
the current ones do not support search by filename. This requires Gitea
to recreate the code search indexes (I do not know if this is a breaking
change, but I feel it deserves a heads-up).
I've used [this
approach](https://www.elastic.co/guide/en/elasticsearch/reference/7.17/analysis-pathhierarchy-tokenizer.html)
to model the filename index. It allows us to efficiently search for both
the full path and the name of a file. Bleve, however, does not support
this out-of-box, so I had to code a brand new [token
filter](https://blevesearch.com/docs/Token-Filters/) to generate the
search terms.
I also did an overhaul in the `indexer_test.go` file. It now asserts the
order of the expected results (this is important since matches based on
the name of a file are more relevant than those based on its content).
I've added new test scenarios that deal with searching by filename. They
use a new repo included in the Gitea fixture.
The screenshot below depicts how Gitea shows the search results. It
shows results based on content in the same way as the current version
does. In matches based on the filename, the first seven lines of the
file contents are shown (BTW, this is how GitHub does it).
Resolves#32096
---------
Signed-off-by: Bruno Sofiato <bruno.sofiato@gmail.com>
When opening a repository, it will call `ensureValidRepository` and also
`CatFileBatch`. But sometimes these will not be used until repository
closed. So it's a waste of CPU to invoke 3 times git command for every
open repository.
This PR removed all of these from `OpenRepository` but only kept
checking whether the folder exists. When a batch is necessary, the
necessary functions will be invoked.
Noteable additions:
- `redefines-builtin-id` forbid variable names that shadow go builtins
- `empty-lines` remove unnecessary empty lines that `gofumpt` does not
remove for some reason
- `superfluous-else` eliminate more superfluous `else` branches
Rules are also sorted alphabetically and I cleaned up various parts of
`.golangci.yml`.
Fix#29136
Before: The result is a table and all line numbers are all in one row.
After: Use a separate table column for the line numbers.
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Fix for gitea putting everything into one request without batching and
sending it to Elasticsearch for indexing as issued in #28117
This issue occured in large repositories while Gitea tries to
index the code using ElasticSearch.
I've applied necessary changes that takes batch length from below config
(app.ini)
```
[queue.code_indexer]
BATCH_LENGTH=<length_int>
```
and batches all requests to Elasticsearch in chunks as configured in the
above config
The `ToUTF8*` functions were stripping BOM, while BOM is actually valid
in UTF8, so the stripping must be optional depending on use case. This
does:
- Add a options struct to all `ToUTF8*` functions, that by default will
strip BOM to preserve existing behaviour
- Remove `ToUTF8` function, it was dead code
- Rename `ToUTF8WithErr` to `ToUTF8`
- Preserve BOM in Monaco Editor
- Remove a unnecessary newline in the textarea value. Browsers did
ignore it, it seems but it's better not to rely on this behaviour.
Fixes: https://github.com/go-gitea/gitea/issues/28743
Related: https://github.com/go-gitea/gitea/issues/6716 which seems to
have once introduced a mechanism that strips and re-adds the BOM, but
from what I can tell, this mechanism was removed at some point after
that PR.
Refactor Hash interfaces and centralize hash function. This will allow
easier introduction of different hash function later on.
This forms the "no-op" part of the SHA256 enablement patch.
assert.Fail() will continue to execute the code while assert.FailNow()
not. I thought those uses of assert.Fail() should exit immediately.
PS: perhaps it's a good idea to use
[require](https://pkg.go.dev/github.com/stretchr/testify/require)
somewhere because the assert package's default behavior does not exit
when an error occurs, which makes it difficult to find the root error
reason.
This PR removed `unittest.MainTest` the second parameter
`TestOptions.GiteaRoot`. Now it detects the root directory by current
working directory.
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Fix#26723
Add `ChangeDefaultBranch` to the `notifier` interface and implement it
in `indexerNotifier`. So when changing the default branch,
`indexerNotifier` sends a message to the `indexer queue` to update the
index.
---------
Co-authored-by: techknowlogick <matti@mdranta.net>
Fix#24662.
Replace #24822 and #25708 (although it has been merged)
## Background
In the past, Gitea supported issue searching with a keyword and
conditions in a less efficient way. It worked by searching for issues
with the keyword and obtaining limited IDs (as it is heavy to get all)
on the indexer (bleve/elasticsearch/meilisearch), and then querying with
conditions on the database to find a subset of the found IDs. This is
why the results could be incomplete.
To solve this issue, we need to store all fields that could be used as
conditions in the indexer and support both keyword and additional
conditions when searching with the indexer.
## Major changes
- Redefine `IndexerData` to include all fields that could be used as
filter conditions.
- Refactor `Search(ctx context.Context, kw string, repoIDs []int64,
limit, start int, state string)` to `Search(ctx context.Context, options
*SearchOptions)`, so it supports more conditions now.
- Change the data type stored in `issueIndexerQueue`. Use
`IndexerMetadata` instead of `IndexerData` in case the data has been
updated while it is in the queue. This also reduces the storage size of
the queue.
- Enhance searching with Bleve/Elasticsearch/Meilisearch, make them
fully support `SearchOptions`. Also, update the data versions.
- Keep most logic of database indexer, but remove
`issues.SearchIssueIDsByKeyword` in `models` to avoid confusion where is
the entry point to search issues.
- Start a Meilisearch instance to test it in unit tests.
- Add unit tests with almost full coverage to test
Bleve/Elasticsearch/Meilisearch indexer.
---------
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Refactor `modules/indexer` to make it more maintainable. And it can be
easier to support more features. I'm trying to solve some of issue
searching, this is a precursor to making functional changes.
Current supported engines and the index versions:
| engines | issues | code |
| - | - | - |
| db | Just a wrapper for database queries, doesn't need version | - |
| bleve | The version of index is **2** | The version of index is **6**
|
| elasticsearch | The old index has no version, will be treated as
version **0** in this PR | The version of index is **1** |
| meilisearch | The old index has no version, will be treated as version
**0** in this PR | - |
## Changes
### Split
Splited it into mutiple packages
```text
indexer
├── internal
│ ├── bleve
│ ├── db
│ ├── elasticsearch
│ └── meilisearch
├── code
│ ├── bleve
│ ├── elasticsearch
│ └── internal
└── issues
├── bleve
├── db
├── elasticsearch
├── internal
└── meilisearch
```
- `indexer/interanal`: Internal shared package for indexer.
- `indexer/interanal/[engine]`: Internal shared package for each engine
(bleve/db/elasticsearch/meilisearch).
- `indexer/code`: Implementations for code indexer.
- `indexer/code/internal`: Internal shared package for code indexer.
- `indexer/code/[engine]`: Implementation via each engine for code
indexer.
- `indexer/issues`: Implementations for issues indexer.
### Deduplication
- Combine `Init/Ping/Close` for code indexer and issues indexer.
- ~Combine `issues.indexerHolder` and `code.wrappedIndexer` to
`internal.IndexHolder`.~ Remove it, use dummy indexer instead when the
indexer is not ready.
- Duplicate two copies of creating ES clients.
- Duplicate two copies of `indexerID()`.
### Enhancement
- [x] Support index version for elasticsearch issues indexer, the old
index without version will be treated as version 0.
- [x] Fix spell of `elastic_search/ElasticSearch`, it should be
`Elasticsearch`.
- [x] Improve versioning of ES index. We don't need `Aliases`:
- Gitea does't need aliases for "Zero Downtime" because it never delete
old indexes.
- The old code of issues indexer uses the orignal name to create issue
index, so it's tricky to convert it to an alias.
- [x] Support index version for meilisearch issues indexer, the old
index without version will be treated as version 0.
- [x] Do "ping" only when `Ping` has been called, don't ping
periodically and cache the status.
- [x] Support the context parameter whenever possible.
- [x] Fix outdated example config.
- [x] Give up the requeue logic of issues indexer: When indexing fails,
call Ping to check if it was caused by the engine being unavailable, and
only requeue the task if the engine is unavailable.
- It is fragile and tricky, could cause data losing (It did happen when
I was doing some tests for this PR). And it works for ES only.
- Just always requeue the failed task, if it caused by bad data, it's a
bug of Gitea which should be fixed.
---------
Co-authored-by: Giteabot <teabot@gitea.io>
Before there was a "graceful function": RunWithShutdownFns, it's mainly
for some modules which doesn't support context.
The old queue system doesn't work well with context, so the old queues
need it.
After the queue refactoring, the new queue works with context well, so,
use Golang context as much as possible, the `RunWithShutdownFns` could
be removed (replaced by RunWithCancel for context cancel mechanism), the
related code could be simplified.
This PR also fixes some legacy queue-init problems, eg:
* typo : archiver: "unable to create codes indexer queue" => "unable to
create repo-archive queue"
* no nil check for failed queues, which causes unfriendly panic
After this PR, many goroutines could have better display name:
## ⚠️ Breaking
The `log.<mode>.<logger>` style config has been dropped. If you used it,
please check the new config manual & app.example.ini to make your
instance output logs as expected.
Although many legacy options still work, it's encouraged to upgrade to
the new options.
The SMTP logger is deleted because SMTP is not suitable to collect logs.
If you have manually configured Gitea log options, please confirm the
logger system works as expected after upgrading.
## Description
Close#12082 and maybe more log-related issues, resolve some related
FIXMEs in old code (which seems unfixable before)
Just like rewriting queue #24505 : make code maintainable, clear legacy
bugs, and add the ability to support more writers (eg: JSON, structured
log)
There is a new document (with examples): `logging-config.en-us.md`
This PR is safer than the queue rewriting, because it's just for
logging, it won't break other logic.
## The old problems
The logging system is quite old and difficult to maintain:
* Unclear concepts: Logger, NamedLogger, MultiChannelledLogger,
SubLogger, EventLogger, WriterLogger etc
* Some code is diffuclt to konw whether it is right:
`log.DelNamedLogger("console")` vs `log.DelNamedLogger(log.DEFAULT)` vs
`log.DelLogger("console")`
* The old system heavily depends on ini config system, it's difficult to
create new logger for different purpose, and it's very fragile.
* The "color" trick is difficult to use and read, many colors are
unnecessary, and in the future structured log could help
* It's difficult to add other log formats, eg: JSON format
* The log outputer doesn't have full control of its goroutine, it's
difficult to make outputer have advanced behaviors
* The logs could be lost in some cases: eg: no Fatal error when using
CLI.
* Config options are passed by JSON, which is quite fragile.
* INI package makes the KEY in `[log]` section visible in `[log.sub1]`
and `[log.sub1.subA]`, this behavior is quite fragile and would cause
more unclear problems, and there is no strong requirement to support
`log.<mode>.<logger>` syntax.
## The new design
See `logger.go` for documents.
## Screenshot
<details>
</details>
## TODO
* [x] add some new tests
* [x] fix some tests
* [x] test some sub-commands (manually ....)
---------
Co-authored-by: Jason Song <i@wolfogre.com>
Co-authored-by: delvh <dev.lh@web.de>
Co-authored-by: Giteabot <teabot@gitea.io>
# ⚠️ Breaking
Many deprecated queue config options are removed (actually, they should
have been removed in 1.18/1.19).
If you see the fatal message when starting Gitea: "Please update your
app.ini to remove deprecated config options", please follow the error
messages to remove these options from your app.ini.
Example:
```
2023/05/06 19:39:22 [E] Removed queue option: `[indexer].ISSUE_INDEXER_QUEUE_TYPE`. Use new options in `[queue.issue_indexer]`
2023/05/06 19:39:22 [E] Removed queue option: `[indexer].UPDATE_BUFFER_LEN`. Use new options in `[queue.issue_indexer]`
2023/05/06 19:39:22 [F] Please update your app.ini to remove deprecated config options
```
Many options in `[queue]` are are dropped, including:
`WRAP_IF_NECESSARY`, `MAX_ATTEMPTS`, `TIMEOUT`, `WORKERS`,
`BLOCK_TIMEOUT`, `BOOST_TIMEOUT`, `BOOST_WORKERS`, they can be removed
from app.ini.
# The problem
The old queue package has some legacy problems:
* complexity: I doubt few people could tell how it works.
* maintainability: Too many channels and mutex/cond are mixed together,
too many different structs/interfaces depends each other.
* stability: due to the complexity & maintainability, sometimes there
are strange bugs and difficult to debug, and some code doesn't have test
(indeed some code is difficult to test because a lot of things are mixed
together).
* general applicability: although it is called "queue", its behavior is
not a well-known queue.
* scalability: it doesn't seem easy to make it work with a cluster
without breaking its behaviors.
It came from some very old code to "avoid breaking", however, its
technical debt is too heavy now. It's a good time to introduce a better
"queue" package.
# The new queue package
It keeps using old config and concept as much as possible.
* It only contains two major kinds of concepts:
* The "base queue": channel, levelqueue, redis
* They have the same abstraction, the same interface, and they are
tested by the same testing code.
* The "WokerPoolQueue", it uses the "base queue" to provide "worker
pool" function, calls the "handler" to process the data in the base
queue.
* The new code doesn't do "PushBack"
* Think about a queue with many workers, the "PushBack" can't guarantee
the order for re-queued unhandled items, so in new code it just does
"normal push"
* The new code doesn't do "pause/resume"
* The "pause/resume" was designed to handle some handler's failure: eg:
document indexer (elasticsearch) is down
* If a queue is paused for long time, either the producers blocks or the
new items are dropped.
* The new code doesn't do such "pause/resume" trick, it's not a common
queue's behavior and it doesn't help much.
* If there are unhandled items, the "push" function just blocks for a
few seconds and then re-queue them and retry.
* The new code doesn't do "worker booster"
* Gitea's queue's handlers are light functions, the cost is only the
go-routine, so it doesn't make sense to "boost" them.
* The new code only use "max worker number" to limit the concurrent
workers.
* The new "Push" never blocks forever
* Instead of creating more and more blocking goroutines, return an error
is more friendly to the server and to the end user.
There are more details in code comments: eg: the "Flush" problem, the
strange "code.index" hanging problem, the "immediate" queue problem.
Almost ready for review.
TODO:
* [x] add some necessary comments during review
* [x] add some more tests if necessary
* [x] update documents and config options
* [x] test max worker / active worker
* [x] re-run the CI tasks to see whether any test is flaky
* [x] improve the `handleOldLengthConfiguration` to provide more
friendly messages
* [x] fine tune default config values (eg: length?)
## Code coverage:
Fixes#22714
### Changes:
1. Add a token filter which named "camelCase" between custom unicode
token filter and "to_lower" token filter when add custom analyzer.
### Notice:
If users want this feature to work, they should delete folder under
{giteaPath}/data/indexers and restart application. Then application will
create a new IndexMapping.
### Screenshots:
### Others:
I originally attempted to give users the ability to configure the
"token_filters" in the "app.ini" file. But I found that if users does
not strictly follow a right order to register "token_filters", they
won't get the expected results. I think it is difficult to ask users to
do this. So I finally give up this idea.
---------
Co-authored-by: Jason Song <i@wolfogre.com>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Change all license headers to comply with REUSE specification.
Fix#16132
Co-authored-by: flynnnnnnnnnn <flynnnnnnnnnn@github>
Co-authored-by: John Olheiser <john.olheiser@gmail.com>
A testing cleanup.
This pull request replaces `os.MkdirTemp` with `t.TempDir`. We can use the `T.TempDir` function from the `testing` package to create temporary directory. The directory created by `T.TempDir` is automatically removed when the test and all its subtests complete.
This saves us at least 2 lines (error check, and cleanup) on every instance, or in some cases adds cleanup that we forgot.
Reference: https://pkg.go.dev/testing#T.TempDir
```go
func TestFoo(t *testing.T) {
// before
tmpDir, err := os.MkdirTemp("", "")
require.NoError(t, err)
defer os.RemoveAll(tmpDir)
// now
tmpDir := t.TempDir()
}
```
Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>
