gered/llm-ls
1
0
Fork 0
Code Issues Pull requests Actions Packages Projects Releases Wiki Activity
65 commits 2 branches 15 tags 358 KiB
Commit graph

65 commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Gered 8cde5ce43f upgrade ollama backend support to use the chat completions api 2024-11-10 18:52:20 -05:00
Gered 29cec6445c add missing string -> LanguageId conversions 
forgot to add "php" before.

also add extra variations for existing Bash and Yaml language types
 2024-11-10 12:52:12 -05:00
Gered 880a729f48 add cmake, dockerfile, php, toml and yaml tree-sitter support 2024-11-10 12:50:50 -05:00
Gered 25b4794d3e update tree-sitter dependencies 
removed tree-sitter-kotlin and tree-sitter-objc as there doesn't seem
to be versions of these compatible with the latest tree-sitter
 2024-11-09 17:47:22 -05:00
Gered 9748c0fd60 pass api_token option to ollama backend 2024-11-03 17:56:39 -05:00
Gered 5abb3c86b4 add rust-toolchain.toml to set rust 1.79.0 to fix build issues 
there are some dependency updates necessary to use 1.80+, but i'm not
really interested in fixes these problems myself, so i'll just sidestep
that problem with this
 2024-11-03 17:14:10 -05:00
Stefan Teofanovic 59febfea52
fix issue 45 - use SystemTime instead of instant for unauthenticated warning (#100)
Some checks failed
test / testbed (map[key:cached name:jaemk/cached node:false parallel:4 python:false]) (push) Has been cancelled
Details
test / testbed (map[key:async-executor name:smol-rs/async-executor node:false parallel:4 python:false]) (push) Has been cancelled
Details
test / testbed (map[key:constrandom name:tkaitchuck/constrandom node:false parallel:8 python:false]) (push) Has been cancelled
Details
test / testbed (map[key:fastapi name:tiangolo/fastapi node:false parallel:8 python:true]) (push) Has been cancelled
Details
test / testbed (map[key:helix name:helix-editor/helix node:false parallel:2 python:false]) (push) Has been cancelled
Details
test / testbed (map[key:huggingface_hub name:huggingface/huggingface_hub node:false parallel:8 python:true]) (push) Has been cancelled
Details
test / testbed (map[key:io-ts name:gcanti/io-ts node:true parallel:8 python:false]) (push) Has been cancelled
Details
test / testbed (map[key:lancedb name:lancedb/lancedb node:false parallel:2 python:false]) (push) Has been cancelled
Details
test / testbed (map[key:picklescan name:mmaitre314/picklescan node:false parallel:8 python:true]) (push) Has been cancelled
Details
test / testbed (map[key:simple name:simple node:false parallel:8 python:false]) (push) Has been cancelled
Details
test / testbed (map[key:starlette name:encode/starlette node:false parallel:8 python:true]) (push) Has been cancelled
Details
test / testbed (map[key:zod name:colinhacks/zod node:true parallel:8 python:false]) (push) Has been cancelled
Details
test / comment_results (push) Has been cancelled
Details 
Co-authored-by: Teofanovic Stefan <stefan.teofanovic@hes-so.ch>
 2024-07-08 10:53:51 +02:00
Luc Georges 479401f3a5
fix(ci): update actions to use nodejs 20 (#96) 2024-05-24 17:33:05 +02:00
Luc Georges 98a12630e7
feat: add backend url route completion (#95) 2024-05-24 16:30:21 +02:00
Luc Georges 0e95bb3589
feat: add llama.cpp backend (#94) 
* feat: add `llama.cpp` backend

* fix(ci): install stable toolchain instead of nightly

* fix(ci): use different model

---------

Co-authored-by: flopes <FredericoPerimLopes@users.noreply.github.com>
 2024-05-24 10:17:48 +02:00
Jeremy Elalouf 078d4c7af2
Feature/multiple encodings handled (#88) 
* feat: support added for multiple encodings
Actually, the only supported encoding was UTF-8.
In cases where the editor sends updates encoded in UTF-16,
the mirror of the user's workspace goes out of sync, leading to a server crash.

Furthermore, UTF-16 is the default and mandatory encoding for the protocol,
and servers must support it.

Therefore, it is imperative to ensure its support.

Now, to avoid any unnecessary conversion, the server negotiates the encoding
with the client during the initialization phase.
This allows the server to choose its preferred encoding in cases where it is available.

See: https://microsoft.github.io/language-server-protocol/specifications/lsp/3.17/specification/#textDocuments

* style: format added code using rustfmt

* feat: cargo clippy warnings fixed

* test: add some unit tests for the new `apply_content_change` function

* feat: review comments fixed
Use of the `document::PositionEncodingKind` enum everywhere it's
possible and TryFrom implemented on it.

Also, LspError are not used anymore.

* feat: review comments fixed, mostly style.
 2024-03-07 13:50:52 +01:00
Luc Georges 386ed53183
test: fix invalid deserialization (#85) 
* test: fix invalid deserialization

* feat: install node or python only when needed

* feat: when action is rerun, run testbed with logs in debug

* feat(ci): update actions to node 20

* feat: log stdout & stderr from setup & build cmd as debug

* fix: bump helix revision

* fix: use helix fork for the time being

* fix(ci): install missing setuptools

* fix(ci): revert back to python 3.10

* fix(ci): disable lance
 2024-02-19 17:25:42 +01:00
Luc Georges 0d035901f0
fix: nameless file crash (#84) 2024-02-19 12:08:32 +01:00
Luc Georges 50b62bd107
fix: AcceptCompletionParams -> RejectCompletionParams (#83) 2024-02-19 11:15:27 +01:00
Luc Georges 0b75e5dd60
refactor: cleanup unused code (#82) 2024-02-19 10:27:01 +01:00
Luc Georges fe1f6aab47
fix: always set return_full_text to false for better UX (#78) 2024-02-13 11:02:50 +01:00
Luc Georges 4437c0c8a6
fix: deserialize url null value w/ default if backend: huggingface (#75) 2024-02-12 20:07:59 +01:00
Luc Georges 86043ce3af
fix: change visiblity of internal functions 2024-02-12 15:58:56 +01:00
Luc Georges 8926969265
docs(README): replace llama.cpp link with python bindings 2024-02-12 11:07:31 +01:00
Luc Georges 4891468c1a
feat: update backend & model parameter (#74) 
* feat: update backend & model parameter

* fix: add `stream: false` in request body for ollama & openai
 2024-02-09 18:42:41 +01:00
Luc Georges 92fc885503
fix: helix editor build crash (#73) 2024-02-08 22:43:56 +01:00
Luc Georges 54b25a8731
feat: add socket connection (#72) 2024-02-07 15:06:58 +01:00
Luc Georges 455b085c96
refactor: error handling (#71) 2024-02-07 12:34:17 +01:00
Luc Georges a9831d5720
refactor: adaptor -> backend (#70) 2024-02-06 21:26:53 +01:00
Markus Hennerbichler 1499fd6cbf
Only warn of rate-limits when using HF endpoint (#58) 
* Only warn of rate-limits when using HF endpoint

Co-authored-by: Luc Georges <McPatate@users.noreply.github.com>
 2024-02-05 21:05:52 +01:00
Markus Hennerbichler c9a44e591c
Fix handling of end-of-file (#60) 
* Fix underflow on empty line

* Update crates/llm-ls/src/main.rs

Co-authored-by: Luc Georges <McPatate@users.noreply.github.com>

* Update crates/llm-ls/src/main.rs

Co-authored-by: Luc Georges <McPatate@users.noreply.github.com>

---------

Co-authored-by: Luc Georges <McPatate@users.noreply.github.com>
 2024-02-05 18:37:24 +01:00
Markus Hennerbichler 16606e5371
Fix off-by-1 error when removing at end (#61) 2024-01-28 23:25:14 +01:00
Markus Hennerbichler c2fbac20ee
Ignore documents with "output" scheme (#59) 2024-01-28 23:21:19 +01:00
Markus Hennerbichler ec2072a621
Fix off-by-1 in prompt creation (#64) 2024-01-28 23:19:15 +01:00
Markus Hennerbichler a5f2e87315
Account for FIM tokens in prompt (#62) 2024-01-28 23:16:01 +01:00
Markus Hennerbichler f40e8cc6ea
Convert generation parameters to snake_case for TGI adapter (#65) 2024-01-28 23:04:38 +01:00
Noah Baldwin 585ea3aae8
feat: add adaptors for various backends (#40) 
* ollama
* tgi
* api-inference
* OpenAI based APIs
 2024-01-02 18:56:01 +01:00
Diego ROJAS 2a433cdf75
Disable ropey unicode_lines feature (#50) 
With the current configuration, Ropey recognises more EOL sequences than the Language Server Protocol. This mismatch can lead to errors when trying to maintain a mirror of the user's documents as the llm-ls might have more lines.

See: https://docs.rs/ropey/1.6.0/ropey/index.html#a-note-about-line-breaks
See: https://microsoft.github.io/language-server-protocol/specifications/lsp/3.17/specification/#textDocuments
 2023-12-15 16:11:05 +01:00
Johan von Forstner b1d0eb4ffe
add support for Kotlin language (#52) 2023-12-15 15:01:40 +01:00
Luc Georges 6c4e0e4176
feat: parallelise at hole level (#44) 
* feat: parallelise at hole level

* fix(ci): move strategy to testbed job

* feat: output json results file

* fix(ci): install jq

* fix(ci): add missing `runs-on`

* fix(ci): add dependency to testbed job

* fix(ci): invalid artifact key name

* fix(ci): add missing i in fastapi key

* feat(ci): make CI run different # of threads per repo

* fix(ci): results.json not in markdown

* feat: round output values

* fix: avoid creating zombie processes

* fix: check on word instead of line

* feat: recreate holes for long CI
 2023-11-17 18:05:45 +01:00
Luc Georges 3ad64a32df
feat!: get completions camel case (#48) 
* feat!: make API camelCase

* fix(testbed): update llm-ls API

* feat: sync docs with incremental change
 2023-11-15 13:36:27 +01:00
Luc Georges 59185abfd9
feat!: make API camelCase (#46) 
API had inconsistencies in case
 2023-11-09 12:00:39 +01:00
Luc Georges f58085b812
fix(ci): outdated lance revision (#43) 2023-11-07 13:24:22 +01:00
Luc Georges c7affd0da9
feat: testbed (#39) 2023-11-06 21:26:37 +01:00
Luc Georges 4aacd7087b
feat: improve logging (#34) 
* feat: imporve logging

* feat: bump to `0.4.0`

* docs: add features section in README
 2023-10-18 11:12:37 +02:00
Luc Georges fdd55dd5c1
fix: new dependencies break CI (#31) 
* fix: install `arm-linux-gnueabihf-g++` as suggested by err

* fix: restrict visibility of modules

* fix: add `g++-aarch64-linux-gnu`
 2023-10-11 21:12:18 +02:00
Luc Georges cdbf76fd43
feat: improve suggestions based on AST (#30) 
* feat: improve suggestions based on AST

* feat: bump version to `0.3.0`
 2023-10-11 19:33:57 +02:00
Luc Georges b6d6c6cccd
feat: send warning to client when unauthenticated (#27) 2023-10-10 14:43:12 +02:00
Luc Georges f90bb13fbd
docs: add llm-intellij link (#26) 2023-10-04 16:51:00 +02:00
Luc Georges bad31bd017
docs: mark llm-vscode as compatible (#23) 2023-09-25 15:41:47 +02:00
Luc Georges fbaf98203f
fix: don't use tokenizer on config error (#22) 2023-09-25 15:10:29 +02:00
Luc Georges 787f2a1a26
feat: improve tokenizer config (#21) 
* feat: improve tokenizer config

* fix: add untagged decorator to `TokenizerConfig`

* feat: bump version to `0.2.0`
 2023-09-21 17:57:19 +02:00
Luc Georges eeb443feb3
feat: add user agent (#20) 
* feat: add user agent

* feat: add mock_server to repo

* feat: bump to `0.1.1`
 2023-09-21 14:32:21 +02:00
Luc Georges 7f9c7855d5
feat: add tokens to clear (#19) 2023-09-20 19:24:38 +02:00
Luc Georges 5378a67ce8
fix: advertise correct version to client (#16) 2023-09-15 18:24:14 +02:00