mirror of
https://github.com/hwchase17/langchain.git
synced 2025-09-10 07:21:03 +00:00
Framework for supporting more languages in LanguageParser (#13318)
## Description I am submitting this for a school project as part of a team of 5. Other team members are @LeilaChr, @maazh10, @Megabear137, @jelalalamy. This PR also has contributions from community members @Harrolee and @Mario928. Initial context is in the issue we opened (#11229). This pull request adds: - Generic framework for expanding the languages that `LanguageParser` can handle, using the [tree-sitter](https://github.com/tree-sitter/py-tree-sitter#py-tree-sitter) parsing library and existing language-specific parsers written for it - Support for the following additional languages in `LanguageParser`: - C - C++ - C# - Go - Java (contributed by @Mario928 https://github.com/ThatsJustCheesy/langchain/pull/2) - Kotlin - Lua - Perl - Ruby - Rust - Scala - TypeScript (contributed by @Harrolee https://github.com/ThatsJustCheesy/langchain/pull/1) Here is the [design document](https://docs.google.com/document/d/17dB14cKCWAaiTeSeBtxHpoVPGKrsPye8W0o_WClz2kk) if curious, but no need to read it. ## Issues - Closes #11229 - Closes #10996 - Closes #8405 ## Dependencies `tree_sitter` and `tree_sitter_languages` on PyPI. We have tried to add these as optional dependencies. ## Documentation We have updated the list of supported languages, and also added a section to `source_code.ipynb` detailing how to add support for additional languages using our framework. ## Maintainer - @hwchase17 (previously reviewed https://github.com/langchain-ai/langchain/pull/6486) Thanks!! ## Git commits We will gladly squash any/all of our commits (esp merge commits) if necessary. Let us know if this is desirable, or if you will be squash-merging anyway. <!-- Thank you for contributing to LangChain! Replace this entire comment with: - **Description:** a description of the change, - **Issue:** the issue # it fixes (if applicable), - **Dependencies:** any dependencies required for this change, - **Tag maintainer:** for a quicker response, tag the relevant maintainer (see below), - **Twitter handle:** we announce bigger features on Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out! Please make sure your PR is passing linting and testing before submitting. Run `make format`, `make lint` and `make test` to check this locally. See contribution guidelines for more information on how to write/run tests, lint, etc: https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md If you're adding a new integration, please include: 1. a test for the integration, preferably unit tests that do not rely on network access, 2. an example notebook showing its use. It lives in `docs/extras` directory. If no one reviews your PR within a few days, please @-mention one of @baskaryan, @eyurtsev, @hwchase17. --> --------- Co-authored-by: Maaz Hashmi <mhashmi373@gmail.com> Co-authored-by: LeilaChr <87657694+LeilaChr@users.noreply.github.com> Co-authored-by: Jeremy La <jeremylai511@gmail.com> Co-authored-by: Megabear137 <zubair.alnoor27@gmail.com> Co-authored-by: Lee Harrold <lhharrold@sep.com> Co-authored-by: Mario928 <88029051+Mario928@users.noreply.github.com> Co-authored-by: Bagatur <baskaryan@gmail.com> Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
This commit is contained in:
Ian Gregory
GitHub
@@ -9,7 +9,35 @@
|
||||
"\n",
|
||||
"This notebook covers how to load source code files using a special approach with language parsing: each top-level function and class in the code is loaded into separate documents. Any remaining code top-level code outside the already loaded functions and classes will be loaded into a separate document.\n",
|
||||
"\n",
|
||||
"This approach can potentially improve the accuracy of QA models over source code. Currently, the supported languages for code parsing are Python and JavaScript. The language used for parsing can be configured, along with the minimum number of lines required to activate the splitting based on syntax."
|
||||
"This approach can potentially improve the accuracy of QA models over source code.\n",
|
||||
"\n",
|
||||
"The supported languages for code parsing are:\n",
|
||||
"\n",
|
||||
"- C (*)\n",
|
||||
"- C++ (*)\n",
|
||||
"- C# (*)\n",
|
||||
"- COBOL\n",
|
||||
"- Go (*)\n",
|
||||
"- Java (*)\n",
|
||||
"- JavaScript (requires package `esprima`)\n",
|
||||
"- Kotlin (*)\n",
|
||||
"- Lua (*)\n",
|
||||
"- Perl (*)\n",
|
||||
"- Python\n",
|
||||
"- Ruby (*)\n",
|
||||
"- Rust (*)\n",
|
||||
"- Scala (*)\n",
|
||||
"- TypeScript (*)\n",
|
||||
"\n",
|
||||
"Items marked with (*) require the packages `tree_sitter` and `tree_sitter_languages`.\n",
|
||||
"It is straightforward to add support for additional languages using `tree_sitter`,\n",
|
||||
"although this currently requires modifying LangChain.\n",
|
||||
"\n",
|
||||
"The language used for parsing can be configured, along with the minimum number of\n",
|
||||
"lines required to activate the splitting based on syntax.\n",
|
||||
"\n",
|
||||
"If a language is not explicitly specified, `LanguageParser` will infer one from\n",
|
||||
"filename extensions, if present."
|
||||
]
|
||||
},
|
||||
{
|
||||
@@ -19,7 +47,7 @@
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"%pip install --upgrade --quiet esprima"
|
||||
"%pip install -qU esprima esprima tree_sitter tree_sitter_languages"
|
||||
]
|
||||
},
|
||||
{
|
||||
@@ -395,6 +423,33 @@
|
||||
"source": [
|
||||
"print(\"\\n\\n--8<--\\n\\n\".join([document.page_content for document in result]))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Adding Languages using Tree-sitter Template\n",
|
||||
"\n",
|
||||
"Expanding language support using the Tree-Sitter template involves a few essential steps:\n",
|
||||
"\n",
|
||||
"1. **Creating a New Language File**:\n",
|
||||
" - Begin by creating a new file in the designated directory (langchain/libs/community/langchain_community/document_loaders/parsers/language).\n",
|
||||
" - Model this file based on the structure and parsing logic of existing language files like **`cpp.py`**.\n",
|
||||
" - You will also need to create a file in the langchain directory (langchain/libs/langchain/langchain/document_loaders/parsers/language).\n",
|
||||
"2. **Parsing Language Specifics**:\n",
|
||||
" - Mimic the structure used in the **`cpp.py`** file, adapting it to suit the language you are incorporating.\n",
|
||||
" - The primary alteration involves adjusting the chunk query array to suit the syntax and structure of the language you are parsing.\n",
|
||||
"3. **Testing the Language Parser**:\n",
|
||||
" - For thorough validation, generate a test file specific to the new language. Create **`test_language.py`** in the designated directory(langchain/libs/community/tests/unit_tests/document_loaders/parsers/language).\n",
|
||||
" - Follow the example set by **`test_cpp.py`** to establish fundamental tests for the parsed elements in the new language.\n",
|
||||
"4. **Integration into the Parser and Text Splitter**:\n",
|
||||
" - Incorporate your new language within the **`language_parser.py`** file. Ensure to update LANGUAGE_EXTENSIONS and LANGUAGE_SEGMENTERS along with the docstring for LanguageParser to recognize and handle the added language.\n",
|
||||
" - Also, confirm that your language is included in **`text_splitter.py`** in class Language for proper parsing.\n",
|
||||
"\n",
|
||||
"By following these steps and ensuring comprehensive testing and integration, you'll successfully extend language support using the Tree-Sitter template.\n",
|
||||
"\n",
|
||||
"Best of luck!"
|
||||
]
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
@@ -413,7 +468,7 @@
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.9.16"
|
||||
"version": "3.11.5"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
|
||||
Reference in New Issue
Block a user