1965: Reintroduce engine version file r=MarinPostma a=irevoire
Right now if you boot up MeiliSearch and point it to a DB directory created with a previous version of MeiliSearch the existing indexes will be deleted. This [used to be](51d7c84e73) prevented by a startup check which would compare the current engine version vs what was stored in the DB directory's version file, but this functionality seems to have been lost after a few refactorings of the code.
In order to go back to the old behavior we'll need to reintroduce the `VERSION` file that used to be present; I considered reusing the `metadata.json` file used in the dumps feature, but this seemed like the simpler and more approach. As the intent is just to restore functionality, the implementation is quite basic. I imagine that in the future we could build on this and do things like compatibility across major/minor versions and even migrating between formats.
This PR was made thanks to `@mbStavola` and is basically a port of his PR #1860 after a big refacto of the code #1796.
Closes#1840
Co-authored-by: Matt Stavola <m.freitas@offensive-security.com>
Right now if you boot up MeiliSearch and point it to a DB directory created with a previous version of MeiliSearch the existing indexes will be deleted. This used to be prevented by a startup check which would compare the current engine version vs what was stored in the DB directory's version file, but this functionality seems to have been lost after a few refactorings of the code.
In order to go back to the old behavior we'll need to reintroduce the VERSION file that used to be present; I considered reusing the metadata.json file used in the dumps feature, but this seemed like the simpler and more approach. As the intent is just to restore functionality, the implementation is quite basic. I imagine that in the future we could build on this and do things like compatibility across major/minor versions and even migrating between formats.
This PR was made thanks to @mbStavola
Closes#1840
implements:
https://github.com/meilisearch/specifications/blob/develop/text/0085-api-keys.md
- Add tests on API keys management route (meilisearch-http/tests/auth/api_keys.rs)
- Add tests checking authorizations on each meilisearch routes (meilisearch-http/tests/auth/authorization.rs)
- Implement API keys management routes (meilisearch-http/src/routes/api_key.rs)
- Create module to manage API keys and authorizations (meilisearch-auth)
- Reimplement GuardedData to extend authorizations (meilisearch-http/src/extractors/authentication/mod.rs)
- Change X-MEILI-API-KEY by Authorization Bearer (meilisearch-http/src/extractors/authentication/mod.rs)
- Change meilisearch routes to fit to the new authorization feature (meilisearch-http/src/routes/)
- close#1867
1796: Feature branch: Task store r=irevoire a=MarinPostma
# Feature branch: Task Store
## Spec todo
https://github.com/meilisearch/specifications/blob/develop/text/0060-refashion-updates-apis.md
- [x] The update resource is renamed task. The names of existing API routes are also changed to reflect this change.
- [x] Tasks are now also accessible as an independent resource of an index. GET - /tasks; GET - /tasks/:taskUid
- [x] The task uid is not incremented by index anymore. The sequence is generated globally.
- [x] A task_not_found error is introduced.
- [x] The format of the task object is updated.
- [x] updateId becomes uid.
- [x] Attributes of an error appearing in a failed task are now contained in a dedicated error object.
- [x] type is no longer an object. It now becomes a string containing the values of its name field previously defined in the type object.
- [x] The possible values for the type field are reworked to be more clear and consistent with our naming rules.
- [x] A details object is added to contain specific information related to a task payload that was previously displayed in the type nested object. Previous number key is renamed numberOfDocuments.
- [x] An indexUid field is added to give information about the related index on which the task is performed.
- [x] duration format has been updated to express an ISO 8601 duration.
- [x] processed status changes to succeeded.
- [x] startedProcessingAt is updated to startedAt.
- [x] processedAt is updated to finishedAt.
- [x] 202 Accepted requests previously returning an updateId are now returning a summarized task object.
- [x] MEILI_MAX_UDB_SIZE env var is updated MEILI_MAX_TASK_DB_SIZE.
- [x] --max-udb-size cli option is updated to --max-task-db-size.
- [x] task object lists are now returned under a results array.
- [x] Each operation on an index (creation, update, deletion) is now asynchronous and represented by a task.
## Todo tech
- [x] Restore Snapshots
- [x] Restore dumps of documents
- [x] Implements the dump of updates
- [x] Error handling
- [x] Fix stats
- [x] Restore the Analytics
- [x] [Add the new analytics](https://github.com/meilisearch/specifications/pull/92/files)
- [x] Fix tests
- [x] ~Deleting tasks when index is deleted (see bellow)~ see #1891 instead
- [x] Improve details for documents addition and deletion tasks
- [ ] Add integration test
- [ ] Test task store filtering
- [x] Rename `UuidStore` to `IndexMetaStore`, and simplify the trait.
- [x] Fix task store initialization: fill pending queue from hard state
- [x] Synchronously return error when creating an index with an invalid index_uid and add test
- [x] Task should be returned in decreasing uid + tests (on index task route)
- [x] Summarized task view
- [x] fix snapshot permissions
## Implementation
### Linked PRs
- #1889
- #1891
- #1892
- #1902
- #1906
- #1911
- #1914
- #1915
- #1916
- #1918
- #1924
- #1925
- #1926
- #1930
- #1936
- #1937
- #1942
- #1944
- #1945
- #1946
- #1947
- #1950
- #1951
- #1957
- #1959
- #1960
- #1961
- #1962
- #1964
### Linked PRs in milli:
- https://github.com/meilisearch/milli/pull/414
- https://github.com/meilisearch/milli/pull/409
- https://github.com/meilisearch/milli/pull/406
- https://github.com/meilisearch/milli/pull/418
### Issues
- close#1687
- close#1786
- close#1940
- close#1948
- close#1949
- close#1932
- close#1956
### Spec patches
- https://github.com/meilisearch/specifications/pull/90
Co-authored-by: Marin Postma <postma.marin@protonmail.com>
1893: Make matches work with numerical value r=MarinPostma a=Thearas
# Pull Request
## What does this PR do?
Implement #1883.
I have test this PR with unit test. It appears to be working properly:
PTAL `@curquiza`
## PR checklist
Please check if your PR fulfills the following requirements:
- [x] Does this PR fix an existing issue?
- [x] Have you read the contributing guidelines?
- [x] Have you made sure that the title is accurate and descriptive of the changes?
Co-authored-by: Thearas <thearas850@gmail.com>
419: fix word pair proximity indexing r=ManyTheFish a=ManyTheFish
# Pull Request
Sort positions before iterating over them during word pair proximity extraction.
fixes [Meilisearch#1913](https://github.com/meilisearch/MeiliSearch/issues/1913)
Co-authored-by: many <maxime@meilisearch.com>
418: change visibility of DocumentDeletionResult r=Kerollmops a=MarinPostma
Change the visibility of `DocumentDeletionResult`, so its fields can be accesses from outside milli.
Co-authored-by: Marin Postma <postma.marin@protonmail.com>
1896: Remove email address from the message at the launch r=irevoire a=curquiza
I suggest removing this email address from the message at the launch since it can encourage people to think this is an email address for support. Is it something we want `@meilisearch/devrel-team` since we mostly redirect them to the forum or the slack?
Co-authored-by: Clémentine Urquizar <clementine@meilisearch.com>
409: remove update_id in UpdateBuilder r=ManyTheFish a=MarinPostma
Removing the `update_id` from `UpdateBuidler`, since it serves no purpose. I had introduced it when working in HA some time ago, but I think there are better ways to do it now, so it can be removed an stop being in our way.
Co-authored-by: Marin Postma <postma.marin@protonmail.com>
1904: Update mini-dashboard version to v0.1.5 r=curquiza a=curquiza
Update the mini-dashboard with its latest version (v0.1.5)
Check with `@mdubus,` replaces https://github.com/meilisearch/MeiliSearch/pull/1903Fixes#1898
Co-authored-by: Clémentine Urquizar <clementine@meilisearch.com>
1897: Add ARM image for Docker to CI r=irevoire a=curquiza
Fixes#1315
- [x] Publish MeiliSearch's docker image for `arm64`
- [x] Add `workflow_dispatch` event in case we need to re-trigger it after a failure without creating a new release
- [x] Use our own server to run the github runner since this CI is really slow (1h instead of 4h)
- [x] Open an issue for a refactor by merging both files in one file (https://github.com/meilisearch/MeiliSearch/issues/1901)
Co-authored-by: Clémentine Urquizar <clementine@meilisearch.com>
414: improve update result types r=ManyTheFish a=MarinPostma
Inprove the returned meta when performing document additions and deletions:
- On document addition return the number of indexed documents and the total number of documents in the index after the indexion
- On document deletion return the number of deleted documents, and the remaining number of documents in the index after the deletion is performed
I also fixed a potential bug when performing a document deletion and the primary key couldn't be found: before we assumed that the db was empty and returned that no documents were deleted, but since we checked before that the db wasn't empty, entering this branch is actually a bug, and now returns a 'MissingPrimaryKey' error.
Co-authored-by: Marin Postma <postma.marin@protonmail.com>
returned metaimprove document addition returned metaimprove document
addition returned metaimprove document addition returned metaimprove
document addition returned metaimprove document addition returned
metaimprove document addition returned meta
400: Rewrite the filter parser and add a lot of tests r=irevoire a=irevoire
This PR is a complete rewrite of #358, which was reverted in #403.
You can already try this PR in Meilisearch here https://github.com/meilisearch/MeiliSearch/pull/1880.
Since writing a parser is quite complicated, I moved all the logic to another workspace called `filter_parser`.
In this workspace, we don't know anything about milli, the filterable fields / field ID or anything.
As you can see in its `cargo.toml`, it has only three dependencies entirely focused on the parsing part:
```
nom = "7.0.0"
nom_locate = "4.0.0"
```
But introducing this new workspace made some changes necessary on the “AST”. Now the parser only returns `Tokens` (a simple `&str` with a bit of context). Everything is interpreted when we execute the filter later in milli.
This crate provides a new error type for all filter related errors.
---------
## Errors
Currently, we have multiple kinds of errors. Sometimes we are generating errors looking like that: (for `name = truc`)
```
Attribute `name` is not filterable. Available filterable attributes are: ``.
```
While sometimes pest was generating errors looking like that:
```
Invalid syntax for the filter parameter: ` --> 1:7
|
1 | name =
| ^---
|
= expected word`.
```
Which most people were seeing like that: (for `name =`)
```
Invalid syntax for the filter parameter: ` --> 1:7\n |\n1 | name =\n | ^---\n |\n = expected word`.
```
-----------
With this PR, the error format is unified between all errors.
All errors follow this more straightforward format:
```
The error message.
[from char]:[to char] filter
```
This should be way easier to read when embedded in the JSON for a human. And it should also allow us to parse the errors easily and provide highlighting or something with a frontend playground.
Here is an example of the two previous errors with the new format:
For `name = truc`:
```
Attribute `name` is not filterable. Available filterable attributes are: ``.
1:4 name = truc
```
Or in one line:
```
Attribute `name` is not filterable. Available filterable attributes are: ``.\n1:4 name = truc
```
And for `name =`:
```
Was expecting a value but instead got nothing.
7:7 name =
```
Or in one line:
```
Was expecting a value but instead got nothing.\n7:7 name =
```
Also, since we now have control over the parser, we can generate more explicit error messages so a lot of new errors have been created. I tried to be as helpful as possible for the user; here is a little overview of the new error message you can get when misusing a filter:
```
Expression `"truc` is missing the following closing delimiter: `"`.
8:13 name = "truc
```
The `_geoRadius` filter is an operation and can't be used as a value.
8:30 name = _geoRadius(12, 13, 14)
```
etc
## Tests
A lot of tests have been written in the `filter_parser` crate. I think there is a unit test for every part of the syntax.
But since we can never be sure we covered all the cases, I also fuzzed the new parser A LOT (for ±8 hours on 20 threads). And the code to fuzz the parser is included in the workspace, so if one day we need to change something to the syntax, we'll be able to re-use it by simply running:
```
cargo fuzz run --release parse
```
## Milli
I renamed the type and module `filter_condition.rs` / `FilterCondition` to `filter.rs` / `Filter`.
Co-authored-by: Tamo <tamo@meilisearch.com>
