Commit Graph

1849 Commits

Author SHA1 Message Date
Tamo
e1053989c0
add a fuzzer on milli 2022-01-12 17:57:54 +01:00
bors[bot]
559e019de1
Merge #424
424: Store the geopoint in three dimensions r=Kerollmops a=irevoire

Related to this issue: https://github.com/meilisearch/MeiliSearch/issues/1872

Fix the whole computation of distance for any “geo” operations (sort or filter). Now when you sort points they are returned to you in the right order.
And when you filter on a specific radius you only get points included in the radius.

This PR changes the way we store the geo points in the RTree.
Instead of considering the latitude and longitude as orthogonal coordinates, we convert them to real orthogonal coordinates projected on a sphere with a radius of 1.
This is the conversion formulae.
![image](https://user-images.githubusercontent.com/7032172/145990456-eefe840a-384f-4486-848b-81d0036814ec.png)
Which, in rust, translate to this function:
```rust
pub fn lat_lng_to_xyz(coord: &[f64; 2]) -> [f64; 3] {
    let [lat, lng] = coord.map(|f| f.to_radians());
    let x = lat.cos() * lng.cos();
    let y = lat.cos() * lng.sin();
    let z = lat.sin();

    [x, y, z]
}
```

Storing the points on a sphere is easier / faster to compute than storing the point on an approximation of the real earth shape.
But when we need to compute the distance between two points we still need to use the haversine distance which works with latitude and longitude.
So, to do the fewest search-time computation possible I'm now associating every point with its `DocId` and its lat/lng.

Co-authored-by: Tamo <tamo@meilisearch.com>
2022-01-10 15:23:43 +00:00
bors[bot]
660eac50b2
Merge #427
427: Handle escaped characters in filters r=Kerollmops a=irevoire



Co-authored-by: Tamo <tamo@meilisearch.com>
2022-01-10 15:01:23 +00:00
Tamo
92804f6f45
apply clippy suggestions 2022-01-10 15:59:04 +01:00
Tamo
0fcde35a20
Update filter-parser/src/value.rs
Co-authored-by: Clément Renault <clement@meilisearch.com>
2022-01-10 15:53:44 +01:00
Tamo
3c7ea1d298
Apply code suggestions
Co-authored-by: Clément Renault <clement@meilisearch.com>
2022-01-10 15:19:21 +01:00
bors[bot]
74594be234
Merge #429
429: Benchmark CIs: not use a default label to call the GH runner r=irevoire a=curquiza

Since we now have multiple self-hosted github runners, we need to differentiate them calling them in the CI. The `self-hosted` label is the default one, so we need to use the unique and appropriate one for the benchmark machine

<img width="925" alt="Capture d’écran 2022-01-04 à 15 42 18" src="https://user-images.githubusercontent.com/20380692/148079840-49cd7878-5912-46ff-8ab8-bf646777f782.png">


Co-authored-by: Clémentine Urquizar <clementine@meilisearch.com>
2022-01-04 15:41:08 +00:00
Clémentine Urquizar
3d99686f7a
Change self-hosted label by benchmarks 2022-01-04 16:01:01 +01:00
bors[bot]
c039562723
Merge #428
428: Reintroduce the gitignore for the fuzzer r=Kerollmops a=irevoire

Reintroduce the gitignore in the fuzz directory

Co-authored-by: Tamo <tamo@meilisearch.com>
2022-01-04 12:09:06 +00:00
Tamo
9bdcd42b9b
reintroduce the gitignore for the fuzzer 2022-01-04 13:07:32 +01:00
bors[bot]
4cae691b86
Merge #425
425: Push the result of the benchmarks to influxdb r=irevoire a=irevoire

Now execute a benchmark for every PR merged into main and then upload the results to influxdb.

Co-authored-by: Tamo <tamo@meilisearch.com>
2022-01-04 11:04:16 +00:00
Tamo
6a1216bd51
Integrate telegraf into our CI 2022-01-04 11:59:05 +01:00
Tamo
02a21fd309
Handle the escapes of quote in the filters 2022-01-04 04:04:10 +01:00
Tamo
98a365aaae
store the geopoint in three dimensions 2021-12-14 12:21:24 +01:00
Tamo
d671d6f0f1
remove an unused file 2021-12-13 19:27:34 +01:00
bors[bot]
11a056d116
Merge #422
422: Prefer returning `None` instead of using an `FilterCondition::Empty` state r=Kerollmops a=Kerollmops

This PR is related to the issue comment https://github.com/meilisearch/MeiliSearch/issues/1338#issuecomment-989322889 which exhibits the fact that when a filter is known to be empty no results are returned which is wrong, the filter should not apply as no restriction is done on the documents set.

The filter system on the milli side has introduced an Empty state which was used in this kind of situation but I found out that it is not needed and that when we parse a filter and that it is empty we can simply return `None` as the `Filter::from_array` constructor does. So I removed it and added tests!

On the MeiliSearch side, we just need to match on a `None` and completely ignore the filter in such a case.

Co-authored-by: Clément Renault <clement@meilisearch.com>
2021-12-09 15:03:04 +00:00
Clément Renault
94011bb9a8
Fix the benchmarks to work with optional filters 2021-12-09 12:14:16 +01:00
Clément Renault
1c6c89f345
Fix the binaries that use the new optional filters 2021-12-09 11:57:53 +01:00
Clément Renault
25faef67d0
Remove the database setup in the filter_depth test 2021-12-09 11:57:53 +01:00
Clément Renault
65519bc04b
Test that empty filters return a None 2021-12-09 11:57:53 +01:00
Clément Renault
ef59762d8e
Prefer returning None instead of the Empty Filter state 2021-12-09 11:57:52 +01:00
bors[bot]
80dcfd5c3e
Merge #421
421: Introduce the depth method on FilterCondition r=Kerollmops a=Kerollmops

This PR introduces the depth method on the FilterCondition type to be able to react to it. It is meant to be used to reject filters that go too deep and can make the engine stack overflow.

Co-authored-by: Clément Renault <clement@meilisearch.com>
2021-12-09 10:28:52 +00:00
Clément Renault
ee856a7a46
Limit the max filter depth to 2000 2021-12-07 17:36:45 +01:00
Clément Renault
32bd9f091f
Detect the filters that are too deep and return an error 2021-12-07 17:20:11 +01:00
Clément Renault
90f49eab6d
Check the filter max depth limit and reject the invalid ones 2021-12-07 16:32:48 +01:00
Clément Renault
49c2db9485
Change the depth function to return the token depth 2021-12-07 16:06:10 +01:00
Clément Renault
57502fcf6a
Introduce the depth method on FilterCondition 2021-12-06 17:35:20 +01:00
bors[bot]
c83b77304a
Merge #420
420: Update milli 0.21.0 r=ManyTheFish a=ManyTheFish

Update all modules to 0.21.0

Co-authored-by: many <maxime@meilisearch.com>
2021-11-30 17:22:12 +00:00
many
1b3923b5ce
Update all packages to 0.21.0 2021-11-29 12:17:59 +01:00
bors[bot]
26629a3f9e
Merge #419
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>
2021-11-23 10:21:05 +00:00
many
8970246bc4
Sort positions before iterating over them during word pair proximity extraction 2021-11-22 18:16:54 +01:00
bors[bot]
cc32519a2d
Merge #418
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>
2021-11-22 14:45:55 +00:00
Marin Postma
6e977dd8e8 change visibility of DocumentDeletionResult 2021-11-22 15:44:44 +01:00
bors[bot]
68f1db123a
Merge #416
416: Update tokenizer v0.2.6 r=Kerollmops a=ManyTheFish



Co-authored-by: many <maxime@meilisearch.com>
2021-11-18 16:01:11 +00:00
many
35f9499638
Export tokenizer from milli 2021-11-18 16:57:12 +01:00
many
64ef5869d7
Update tokenizer v0.2.6 2021-11-18 16:56:05 +01:00
bors[bot]
2c14efa8a2
Merge #409
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>
2021-11-16 14:59:09 +00:00
Marin Postma
6eb47ab792 remove update_id in UpdateBuilder 2021-11-16 13:07:04 +01:00
bors[bot]
21b78f3926
Merge #414
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>
2021-11-15 09:06:10 +00:00
Marin Postma
09b4281cff improve document addition 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
metaimprove document addition returned meta
2021-11-10 14:08:36 +01:00
Marin Postma
721fc294be improve document deletion returned meta
returns both the remaining number of documents and the number of deleted
documents.
2021-11-10 14:08:18 +01:00
bors[bot]
8dff08d772
Merge #400
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>
2021-11-09 16:09:34 +00:00
Irevoire
7c3017734a
re-ignore the ! symbol when generating a good error message 2021-11-09 17:08:04 +01:00
Tamo
bff48681d2
Re-order the operator
Co-authored-by: Clément Renault <clement@meilisearch.com>
2021-11-09 17:05:36 +01:00
Irevoire
519d6b2bf3
remove the ! syntax for the not 2021-11-09 16:47:54 +01:00
Irevoire
73df873f44
fix typos 2021-11-09 16:41:10 +01:00
Irevoire
99197387af
fix the test with the new escaped format 2021-11-09 16:41:10 +01:00
Tamo
f28600031d
Rename the filter_parser crate into filter-parser
Co-authored-by: Clément Renault <clement@meilisearch.com>
2021-11-09 16:41:10 +01:00
Irevoire
0ea0146e04
implement deref &str on the tokens 2021-11-09 11:34:10 +01:00
Irevoire
a211a9cdcd
update the error format so it can be easily parsed by someone else 2021-11-09 11:19:30 +01:00