github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-24 07:35:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

box[patch]: fix bugs in docs (#25699)

Browse Source
This commit is contained in:
Scott Hurrey 2024-08-23 15:36:23 -04:00 committed by GitHub
parent 64ace25eb8
commit 92abf62292
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
6 changed files with 195 additions and 78 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
docs/docs/integrations/document_loaders/box.ipynb
View File

@ -15,7 +15,7 @@
"source": [
"# BoxLoader\n",
"\n",
"This notebook provides a quick overview for getting started with Box [document loader](/docs/integrations/document_loaders/). For detailed documentation of all BoxLoader features and configurations head to the [API reference](https://python.langchain.com/v0.2/api_reference/community/document_loaders/langchain_community.document_loaders.langchain_box_loader.BoxLoader.html).\n",
"This notebook provides a quick overview for getting started with Box [document loader](/docs/integrations/document_loaders/). For detailed documentation of all BoxLoader features and configurations head to the [API reference](https://python.langchain.com/v0.2/api_reference/box/document_loaders/langchain_box.document_loaders.box.BoxLoader.html).\n",
"\n",
"\n",
"## Overview\n",
@ -34,7 +34,7 @@
"\n",
"| Class | Package | Local | Serializable | JS support|\n",
"| :--- | :--- | :---: | :---: | :---: |\n",
"| [BoxLoader](https://python.langchain.com/v0.2/api_reference/box/document_loaders/langchain_box.document_loaders.langchain_boxloader.BoxLoader.html) | [langchain_box](https://python.langchain.com/v0.2/api_reference/box/index.html) | ✅ | ❌ | ❌ | \n",
"| [BoxLoader](https://python.langchain.com/v0.2/api_reference/box/document_loaders/langchain_box.document_loaders.box.BoxLoader.html) | [langchain_box](https://python.langchain.com/v0.2/api_reference/box/index.html) | ✅ | ❌ | ❌ | \n",
"### Loader features\n",
"| Source | Document Lazy Loading | Async Support\n",
"| :---: | :---: | :---: | \n",
@ -244,7 +244,7 @@
"source": [
"## API reference\n",
"\n",
"For detailed documentation of all BoxLoader features and configurations head to the API reference: https://python.langchain.com/v0.2/api_reference/box/document_loaders/langchain_box.document_loaders.langchain_box_loader.BoxLoader.html\n",
"For detailed documentation of all BoxLoader features and configurations head to the [API reference](https://python.langchain.com/v0.2/api_reference/box/document_loaders/langchain_box.document_loaders.box.BoxLoader.html)\n",
"\n",
"\n",
"## Help\n",

10
docs/docs/integrations/providers/box.mdx
View File

@ -9,8 +9,8 @@ In this package, we make available a number of ways to include Box content in yo
### Installation and setup
```text
%pip install -U langchain-box
```bash
pip install -U langchain-box
```
@ -31,12 +31,6 @@ In order to integrate with Box, you need a few things:
The good news is if you are using a free developer account, you are the admin.
[Authorize your app](https://developer.box.com/guides/authorization/custom-app-approval/#manual-approval)
## Installation
```bash
pip install -U langchain-box
```
## Authentication
The `box-langchain` package offers some flexibility to authentication. The

6
docs/docs/integrations/retrievers/box.ipynb
View File

@ -17,7 +17,7 @@
"source": [
"# BoxRetriever\n",
"\n",
"This will help you getting started with the Box [retriever](/docs/concepts/#retrievers). For detailed documentation of all BoxRetriever features and configurations head to the [API reference](https://python.langchain.com/v0.2/api_reference/box/retrievers/langchain_box.retrievers.Box.BoxRetriever.html).\n",
"This will help you getting started with the Box [retriever](/docs/concepts/#retrievers). For detailed documentation of all BoxRetriever features and configurations head to the [API reference](https://python.langchain.com/v0.2/api_reference/box/retrievers/langchain_box.retrievers.box.BoxRetriever.html).\n",
"\n",
"# Overview\n",
"\n",
@ -35,7 +35,7 @@
"\n",
"| Retriever | Self-host | Cloud offering | Package |\n",
"| :--- | :--- | :---: | :---: |\n",
"[BoxRetriever](https://python.langchain.com/v0.2/api_reference/langchain-box/retrievers/langchain-box.retrievers.langchain_box.BoxRetriever.html) | ❌ | ✅ | langchain-box |\n",
"[BoxRetriever](https://python.langchain.com/v0.2/api_reference/box/retrievers/langchain_box.retrievers.box.BoxRetriever.html) | ❌ | ✅ | langchain-box |\n",
"\n",
"## Setup\n",
"\n",
@ -290,7 +290,7 @@
"source": [
"## API reference\n",
"\n",
"For detailed documentation of all BoxRetriever features and configurations head to the [API reference](https://python.langchain.com/v0.2/api_reference/box/retrievers/langchain_box.retrievers.Box.BoxRetriever.html).\n",
"For detailed documentation of all BoxRetriever features and configurations head to the [API reference](https://python.langchain.com/v0.2/api_reference/box/retrievers/langchain_box.retrievers.box.BoxRetriever.html).\n",
"\n",
"\n",
"## Help\n",

81
libs/partners/box/langchain_box/document_loaders/box.py
View File

@ -21,11 +21,12 @@ class BoxLoader(BaseLoader, BaseModel):
Plus plan or above. The free developer account does not have access to Box AI.
In addition, using the Box AI API requires a few prerequisite steps:
* Your administrator must enable the Box AI API
* You must enable the `Manage AI` scope in your app in the developer console.
* Your administratormust install and enable your application.
Setup:
* Your administrator must enable the Box AI API
* You must enable the ``Manage AI`` scope in your app in the developer console.
* Your administrator must install and enable your application.
**Setup**:
Install ``langchain-box`` and set environment variable ``BOX_DEVELOPER_TOKEN``.
.. code-block:: bash
@ -34,7 +35,7 @@ class BoxLoader(BaseLoader, BaseModel):
export BOX_DEVELOPER_TOKEN="your-api-key"
This loader returns ``Document `` objects built from text representations of files
This loader returns ``Document`` objects built from text representations of files
in Box. It will skip any document without a text representation available. You can
provide either a ``List[str]`` containing Box file IDS, or you can provide a
``str`` contining a Box folder ID. If providing a folder ID, you can also enable
@ -46,19 +47,44 @@ class BoxLoader(BaseLoader, BaseModel):
recommend never getting all files from folder 0 recursively. Folder ID 0 is your
root folder.
Instantiate:
**Instantiate**:
Initialization variables
variable | description | type | required
---+---+---
box_developer_token | token to use for auth. | string | no
box_auth | client id for you app. Used for CCG | string | no
box_file_ids | Array of Box file Ids to retrieve | array of strings | no
box_folder_id | Box folder id to retrieve | string | no
recursive | whether to return subfolders, default False | bool | no
.. list-table:: Initialization variables
:widths: 25 50 15 10
:header-rows: 1
Get files — this method requires you pass the ``box_file_ids`` parameter. This is a
``List[str]`` containing the file IDs you wish to index.
* - Variable
- Description
- Type
- Default
* - box_developer_token
- Token to use for auth.
- ``str``
- ``None``
* - box_auth
- client id for you app. Used for CCG
- ``langchain_box.utilities.BoxAuth``
- ``None``
* - box_file_ids
- client id for you app. Used for CCG
- ``List[str]``
- ``None``
* - box_folder_id
- client id for you app. Used for CCG
- ``str``
- ``None``
* - recursive
- client id for you app. Used for CCG
- ``Bool``
- ``False``
* - character_limit
- client id for you app. Used for CCG
- ``int``
- ``-1``
**Get files** — this method requires you pass the ``box_file_ids`` parameter.
This is a ``List[str]`` containing the file IDs you wish to index.
.. code-block:: python
@ -71,7 +97,7 @@ class BoxLoader(BaseLoader, BaseModel):
character_limit=10000 # Optional. Defaults to no limit
)
Get files in a folder — this method requires you pass the ``box_folder_id``
**Get files in a folder** — this method requires you pass the ``box_folder_id``
parameter. This is a ``str`` containing the folder ID you wish to index.
.. code-block:: python
@ -85,7 +111,7 @@ class BoxLoader(BaseLoader, BaseModel):
recursive=False # Optional. return entire tree, defaults to False
)
Load:
**Load**:
.. code-block:: python
docs = loader.load()
@ -96,11 +122,11 @@ class BoxLoader(BaseLoader, BaseModel):
Document(metadata={'source': 'https://dl.boxcloud.com/api/2.0/
internal_files/1514555423624/versions/1663171610024/representations
/extracted_text/content/', 'title': 'Invoice-A5555_txt'},
page_content='Vendor: AstroTech Solutions\nInvoice Number: A5555\n\nLine
Items:\n - Gravitational Wave Detector Kit: $800\n - Exoplanet
Terrarium: $120\nTotal: $920')
page_content='Vendor: AstroTech Solutions\\nInvoice Number: A5555\\n\\nLine
Items:\\n - Gravitational Wave Detector Kit: $800\\n - Exoplanet
Terrarium: $120\\nTotal: $920')
Lazy load:
**Lazy load**:
.. code-block:: python
docs = []
@ -116,16 +142,19 @@ class BoxLoader(BaseLoader, BaseModel):
Document(metadata={'source': 'https://dl.boxcloud.com/api/2.0/
internal_files/1514555423624/versions/1663171610024/representations
/extracted_text/content/', 'title': 'Invoice-A5555_txt'},
page_content='Vendor: AstroTech Solutions\nInvoice Number: A5555\n\nLine
Items:\n - Gravitational Wave Detector Kit: $800\n - Exoplanet
Terrarium: $120\nTotal: $920')
page_content='Vendor: AstroTech Solutions\\nInvoice Number: A5555\\n\\nLine
Items:\\n - Gravitational Wave Detector Kit: $800\\n - Exoplanet
Terrarium: $120\\nTotal: $920')
"""
box_developer_token: Optional[str] = None
"""String containing the Box Developer Token generated in the developer console"""
box_auth: Optional[BoxAuth] = None
"""Configured langchain_box.utilities.BoxAuth object"""
"""Configured
`BoxAuth <https://python.langchain.com/v0.2/api_reference/box/utilities/langchain_box.utilities.box.BoxAuth.html>`_
object"""
box_file_ids: Optional[List[str]] = None
"""List[str] containing Box file ids"""

7
libs/partners/box/langchain_box/retrievers/box.py
View File

@ -32,6 +32,7 @@ class BoxRetriever(BaseRetriever):
Instantiate:
To use search:
.. code-block:: python
from langchain_box.retrievers import BoxRetriever
@ -39,6 +40,7 @@ class BoxRetriever(BaseRetriever):
retriever = BoxRetriever()
To use Box AI:
.. code-block:: python
from langchain_box.retrievers import BoxRetriever
@ -114,10 +116,13 @@ class BoxRetriever(BaseRetriever):
"""String containing the Box Developer Token generated in the developer console"""
box_auth: Optional[BoxAuth] = None
"""Configured langchain_box.utilities.BoxAuth object"""
"""Configured
`BoxAuth <https://python.langchain.com/v0.2/api_reference/box/utilities/langchain_box.utilities.box.BoxAuth.html>`_
object"""
box_file_ids: Optional[List[str]] = None
"""List[str] containing Box file ids"""
character_limit: Optional[int] = -1
"""character_limit is an int that caps the number of characters to
return per document."""

163
libs/partners/box/langchain_box/utilities/box.py
View File

@ -134,52 +134,63 @@ class BoxAuthType(Enum):
an enum to tell BoxLoader how you wish to autheticate your Box connection.
Options are:
TOKEN - Use a developer token generated from the Box Deevloper Token.
Only recommended for development.
Provide `box_developer_token`.
Provide ``box_developer_token``.
CCG - Client Credentials Grant.
provide `box_client_id`, `box_client_secret`,
and `box_enterprise_id` or optionally `box_user_id`.
provide ``box_client_id`, ``box_client_secret`,
and ``box_enterprise_id`` or optionally `box_user_id`.
JWT - Use JWT for authentication. Config should be stored on the file
system accessible to your app.
provide `box_jwt_path`. Optionally, provide `box_user_id` to
provide ``box_jwt_path``. Optionally, provide ``box_user_id`` to
act as a specific user
"""
TOKEN = "token"
"""Use a developer token or a token retrieved from box-sdk-gen"""
"""Use a developer token or a token retrieved from ``box-sdk-gen``"""
CCG = "ccg"
"""Use `client_credentials` type grant"""
"""Use ``client_credentials`` type grant"""
JWT = "jwt"
"""Use JWT bearer token auth"""
class BoxAuth(BaseModel):
"""BoxAuth.
"""**BoxAuth.**
The ``box-langchain`` package offers some flexibility to authentication. The
most basic authentication method is by using a developer token. This can be
found in the `Box developer console <https://account.box.com/developers/console>`_
on the configuration screen. This token is purposely short-lived (1 hour) and is
intended for development. With this token, you can add it to your environment as
``BOX_DEVELOPER_TOKEN``, you can pass it directly to the loader, or you can use the
``BoxAuth`` authentication helper class.
`BoxAuth` supports the following authentication methods:
* Token — either a developer token or any token generated through the Box SDK
* JWT with a service account
* JWT with a specified user
* CCG with a service account
* CCG with a specified user
* **Token** — either a developer token or any token generated through the Box SDK
* **JWT** with a service account
* **JWT** with a specified user
* **CCG** with a service account
* **CCG** with a specified user
.. note::
If using JWT authentication, you will need to download the configuration from the
Box developer console after generating your public/private key pair. Place this
file in your application directory structure somewhere. You will use the path to
this file when using the `BoxAuth` helper class.
If using JWT authentication, you will need to download the configuration from
the Box developer console after generating your public/private key pair. Place
this file in your application directory structure somewhere. You will use the
path to this file when using the ``BoxAuth`` helper class. If you wish to use
OAuth2 with the authorization_code flow, please use ``BoxAuthType.TOKEN`` with
the token you have acquired.
For more information, learn about how to
[set up a Box application](https://developer.box.com/guides/getting-started/first-application/),
`set up a Box application <https://developer.box.com/guides/getting-started/first-application/>`_,
and check out the
[Box authentication guide](https://developer.box.com/guides/authentication/select/)
`Box authentication guide <https://developer.box.com/guides/authentication/select/>`_
for more about our different authentication options.
Simple implementation
Simple implementation:
To instantiate, you must provide a ``langchain_box.utilities.BoxAuthType``.
@ -187,18 +198,24 @@ class BoxAuth(BaseModel):
Box connection.
Options are:
TOKEN - Use a developer token generated from the Box Deevloper Token.
Only recommended for development.
Provide `box_developer_token`.
Provide ``box_developer_token``.
CCG - Client Credentials Grant.
provide `box_client_id`, `box_client_secret`,
and `box_enterprise_id` or optionally `box_user_id`.
provide ``box_client_id``, ``box_client_secret``,
and ``box_enterprise_id`` or optionally ``box_user_id``.
JWT - Use JWT for authentication. Config should be stored on the file
system accessible to your app.
provide `box_jwt_path`. Optionally, provide `box_user_id` to
provide ``box_jwt_path``. Optionally, provide ``box_user_id`` to
act as a specific user
**Examples**:
**Token**
.. code-block:: python
from langchain_box.document_loaders import BoxLoader
from langchain_box.utilities import BoxAuth, BoxAuthType
@ -212,37 +229,109 @@ class BoxAuth(BaseModel):
...
)
To see examples for each supported authentication methodology, visit the
[Box providers](/docs/integrations/providers/box) page. If you want to
use OAuth 2.0 `authorization_code` flow, use
[box-sdk-gen](https://github.com/box/box-python-sdk-gen) SDK, get your
token, and use `BoxAuthType.TOKEN` type.
**JWT with a service account**
.. code-block:: python
from langchain_box.document_loaders import BoxLoader
from langchain_box.utilities import BoxAuth, BoxAuthType
auth = BoxAuth(
auth_type=BoxAuthType.JWT,
box_jwt_path=box_jwt_path
)
loader = BoxLoader(
box_auth=auth,
...
)
**JWT with a specified user**
.. code-block:: python
from langchain_box.document_loaders import BoxLoader
from langchain_box.utilities import BoxAuth, BoxAuthType
auth = BoxAuth(
auth_type=BoxAuthType.JWT,
box_jwt_path=box_jwt_path,
box_user_id=box_user_id
)
loader = BoxLoader(
box_auth=auth,
...
)
**CCG with a service account**
.. code-block:: python
from langchain_box.document_loaders import BoxLoader
from langchain_box.utilities import BoxAuth, BoxAuthType
auth = BoxAuth(
auth_type=BoxAuthType.CCG,
box_client_id=box_client_id,
box_client_secret=box_client_secret,
box_enterprise_id=box_enterprise_id
)
loader = BoxLoader(
box_auth=auth,
...
)
**CCG with a specified user**
.. code-block:: python
from langchain_box.document_loaders import BoxLoader
from langchain_box.utilities import BoxAuth, BoxAuthType
auth = BoxAuth(
auth_type=BoxAuthType.CCG,
box_client_id=box_client_id,
box_client_secret=box_client_secret,
box_user_id=box_user_id
)
loader = BoxLoader(
box_auth=auth,
...
)
"""
auth_type: BoxAuthType
"""langchain_box.utilities.BoxAuthType. Enum describing how to
"""``langchain_box.utilities.BoxAuthType``. Enum describing how to
authenticate against Box"""
box_developer_token: Optional[str] = None
""" If using BoxAuthType.TOKEN, provide your token here"""
""" If using ``BoxAuthType.TOKEN``, provide your token here"""
box_jwt_path: Optional[str] = None
"""If using BoxAuthType.JWT, provide local path to your
"""If using ``BoxAuthType.JWT``, provide local path to your
JWT configuration file"""
box_client_id: Optional[str] = None
"""If using BoxAuthType.CCG, provide your app's client ID"""
"""If using ``BoxAuthType.CCG``, provide your app's client ID"""
box_client_secret: Optional[str] = None
"""If using BoxAuthType.CCG, provide your app's client secret"""
"""If using ``BoxAuthType.CCG``, provide your app's client secret"""
box_enterprise_id: Optional[str] = None
"""If using BoxAuthType.CCG, provide your enterprise ID.
Only required if you are not sending `box_user_id`"""
"""If using ``BoxAuthType.CCG``, provide your enterprise ID.
Only required if you are not sending ``box_user_id``"""
box_user_id: Optional[str] = None
"""If using BoxAuthType.CCG or BoxAuthType.JWT, providing
`box_user_id` will act on behalf of a specific user"""
"""If using ``BoxAuthType.CCG`` or ``BoxAuthType.JWT``, providing
``box_user_id`` will act on behalf of a specific user"""
_box_client: Optional[box_sdk_gen.BoxClient] = None
_custom_header: Dict = dict({"x-box-ai-library": "langchain"})