Remove unwanted libraries and lighten package (#1391)

mem0ai · Jul 8, 2024 · 1a5d0d2 · 1a5d0d2
1 parent ebbf90f
commit 1a5d0d2
Show file tree

Hide file tree

Showing 28 changed files with 238 additions and 1,544 deletions.
diff --git a/Makefile b/Makefile
@@ -9,9 +9,13 @@ PROJECT_NAME := embedchain
 install:
 	poetry install
 
+# TODO: use a more efficient way to install these packages
 install_all:
 	poetry install --all-extras
-	poetry run pip install pinecone-text pinecone-client langchain-anthropic "unstructured[local-inference, all-docs]" ollama langchain_together==0.1.3 langchain_cohere==0.1.5 deepgram-sdk==3.2.7 langchain-huggingface psutil
+	poetry run pip install pinecone-text pinecone-client langchain-anthropic "unstructured[local-inference, all-docs]" ollama langchain_together==0.1.3 \
+		langchain_cohere==0.1.5 deepgram-sdk==3.2.7 langchain-huggingface psutil clarifai==10.0.1 flask==2.3.3 twilio==8.5.0 fastapi-poe==0.0.16 discord==2.3.2 \
+	 	slack-sdk==3.21.3 huggingface_hub==0.23.0 gitpython==3.1.38 yt_dlp==2023.11.14 PyGithub==1.59.1 feedparser==6.0.10 newspaper3k==0.2.8 listparser==0.19 \
+	 	modal==0.56.4329 dropbox==11.36.2 boto3==1.34.20 youtube-transcript-api==0.6.1 pytube==15.0.0 beautifulsoup4==4.12.3
 
 install_es:
 	poetry install --extras elasticsearch

diff --git a/docs/examples/poe_bot.mdx b/docs/examples/poe_bot.mdx
@@ -7,7 +7,7 @@ title: '🔮 Poe Bot'
 1. Install embedchain python package:
 
 ```bash
-pip install --upgrade "embedchain[poe]"
+pip install fastapi-poe==0.0.16 
 ```
 
 2. Create a free account on [Poe](https://www.poe.com?utm_source=embedchain).

diff --git a/embedchain/bots/discord.py b/embedchain/bots/discord.py
@@ -12,8 +12,7 @@
     from discord.ext import commands
 except ModuleNotFoundError:
     raise ModuleNotFoundError(
-        "The required dependencies for Discord are not installed."
-        'Please install with `pip install "embedchain[discord]"`'
+        "The required dependencies for Discord are not installed." "Please install with `pip install discord==2.3.2`"
     ) from None
 
 

diff --git a/embedchain/bots/poe.py b/embedchain/bots/poe.py
@@ -11,7 +11,7 @@
     from fastapi_poe import PoeBot, run
 except ModuleNotFoundError:
     raise ModuleNotFoundError(
-        "The required dependencies for Poe are not installed." 'Please install with `pip install "embedchain[poe]"`'
+        "The required dependencies for Poe are not installed." "Please install with `pip install fastapi-poe==0.0.16`"
     ) from None
 
 

diff --git a/embedchain/bots/slack.py b/embedchain/bots/slack.py
@@ -15,7 +15,7 @@
 except ModuleNotFoundError:
     raise ModuleNotFoundError(
         "The required dependencies for Slack are not installed."
-        'Please install with `pip install --upgrade "embedchain[slack]"`'
+        "Please install with `pip install slack-sdk==3.21.3 flask==2.3.3`"
     ) from None
 
 

diff --git a/embedchain/bots/whatsapp.py b/embedchain/bots/whatsapp.py
@@ -20,7 +20,7 @@ def __init__(self):
         except ModuleNotFoundError:
             raise ModuleNotFoundError(
                 "The required dependencies for WhatsApp are not installed. "
-                'Please install with `pip install --upgrade "embedchain[whatsapp]"`'
+                "Please install with `pip install twilio==8.5.0 flask==2.3.3`"
             ) from None
         super().__init__()
 

diff --git a/embedchain/deployment/modal.com/app.py b/embedchain/deployment/modal.com/app.py
@@ -8,7 +8,18 @@
 
 image = Image.debian_slim().pip_install(
     "embedchain",
-    "embedchain[dataloaders]",
+    "lanchain_community==0.2.6",
+    "youtube-transcript-api==0.6.1",
+    "pytube==15.0.0",
+    "beautifulsoup4==4.12.3",
+    "slack-sdk==3.21.3",
+    "huggingface_hub==0.23.0",
+    "gitpython==3.1.38",
+    "yt_dlp==2023.11.14",
+    "PyGithub==1.59.1",
+    "feedparser==6.0.10",
+    "newspaper3k==0.2.8",
+    "listparser==0.19",
 )
 
 stub = Stub(

diff --git a/embedchain/llm/aws_bedrock.py b/embedchain/llm/aws_bedrock.py
@@ -23,7 +23,7 @@ def _get_answer(self, prompt: str, config: BaseLlmConfig) -> str:
         except ModuleNotFoundError:
             raise ModuleNotFoundError(
                 "The required dependencies for AWSBedrock are not installed."
-                'Please install with `pip install --upgrade "embedchain[aws-bedrock]"`'
+                "Please install with `pip install boto3==1.34.20`"
             ) from None
 
         self.boto_client = boto3.client("bedrock-runtime", "us-west-2" or os.environ.get("AWS_REGION"))
@@ -38,8 +38,7 @@ def _get_answer(self, prompt: str, config: BaseLlmConfig) -> str:
         }
 
         if config.stream:
-            from langchain.callbacks.streaming_stdout import \
-                StreamingStdOutCallbackHandler
+            from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
 
             callbacks = [StreamingStdOutCallbackHandler()]
             llm = Bedrock(**kwargs, streaming=config.stream, callbacks=callbacks)

diff --git a/embedchain/llm/base.py b/embedchain/llm/base.py
@@ -180,7 +180,7 @@ def access_search_and_get_results(input_query: str):
             from langchain.tools import DuckDuckGoSearchRun
         except ImportError:
             raise ImportError(
-                'Searching requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
+                "Searching requires extra dependencies. Install with `pip install duckduckgo-search==6.1.5`"
             ) from None
         search = DuckDuckGoSearchRun()
         logger.info(f"Access search to get answers for {input_query}")

diff --git a/embedchain/llm/clarifai.py b/embedchain/llm/clarifai.py
@@ -24,7 +24,7 @@ def _get_answer(prompt: str, config: BaseLlmConfig) -> str:
         except ModuleNotFoundError:
             raise ModuleNotFoundError(
                 "The required dependencies for Clarifai are not installed."
-                'Please install with `pip install --upgrade "embedchain[clarifai]"`'
+                "Please install with `pip install clarifai==10.0.1`"
             ) from None
 
         model_name = config.model

diff --git a/embedchain/llm/huggingface.py b/embedchain/llm/huggingface.py
@@ -22,7 +22,7 @@ def __init__(self, config: Optional[BaseLlmConfig] = None):
         except ModuleNotFoundError:
             raise ModuleNotFoundError(
                 "The required dependencies for HuggingFaceHub are not installed."
-                'Please install with `pip install --upgrade "embedchain[huggingface-hub]"`'
+                "Please install with `pip install huggingface-hub==0.23.0`"
             ) from None
 
         super().__init__(config=config)

diff --git a/embedchain/loaders/beehiiv.py b/embedchain/loaders/beehiiv.py
@@ -24,7 +24,7 @@ def load_data(self, url: str):
             from bs4.builder import ParserRejectedMarkup
         except ImportError:
             raise ImportError(
-                'Beehiiv requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
+                "Beehiiv requires extra dependencies. Install with `pip install beautifulsoup4==4.12.3`"
             ) from None
 
         if not url.endswith("sitemap.xml"):

diff --git a/embedchain/loaders/docs_site_loader.py b/embedchain/loaders/docs_site_loader.py
@@ -8,7 +8,7 @@
     from bs4 import BeautifulSoup
 except ImportError:
     raise ImportError(
-        'DocsSite requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
+        "DocsSite requires extra dependencies. Install with `pip install beautifulsoup4==4.12.3`"
     ) from None
 
 

diff --git a/embedchain/loaders/docx_file.py b/embedchain/loaders/docx_file.py
@@ -3,9 +3,7 @@
 try:
     from langchain_community.document_loaders import Docx2txtLoader
 except ImportError:
-    raise ImportError(
-        'Docx file requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
-    ) from None
+    raise ImportError("Docx file requires extra dependencies. Install with `pip install docx2txt==0.8`") from None
 from embedchain.helpers.json_serializable import register_deserializable
 from embedchain.loaders.base_loader import BaseLoader
 

diff --git a/embedchain/loaders/dropbox.py b/embedchain/loaders/dropbox.py
@@ -17,9 +17,7 @@ def __init__(self):
         try:
             from dropbox import Dropbox, exceptions
         except ImportError:
-            raise ImportError(
-                'Dropbox requires extra dependencies. Install with `pip install --upgrade "embedchain[dropbox]"`'
-            )
+            raise ImportError("Dropbox requires extra dependencies. Install with `pip install dropbox==11.36.2`")
 
         try:
             dbx = Dropbox(access_token)

diff --git a/embedchain/loaders/github.py b/embedchain/loaders/github.py
@@ -30,7 +30,8 @@ def __init__(self, config: Optional[dict[str, Any]] = None):
             from github import Github
         except ImportError as e:
             raise ValueError(
-                "GithubLoader requires extra dependencies. Install with `pip install --upgrade 'embedchain[github]'`"
+                "GithubLoader requires extra dependencies. \
+                  Install with `pip install gitpython==3.1.38 PyGithub==1.59.1`"
             ) from e
 
         self.config = config

diff --git a/embedchain/loaders/pdf_file.py b/embedchain/loaders/pdf_file.py
@@ -1,11 +1,6 @@
 import hashlib
 
-try:
-    from langchain_community.document_loaders import PyPDFLoader
-except ImportError:
-    raise ImportError(
-        'PDF File requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
-    ) from None
+from langchain_community.document_loaders import PyPDFLoader
 from embedchain.helpers.json_serializable import register_deserializable
 from embedchain.loaders.base_loader import BaseLoader
 from embedchain.utils.misc import clean_string

diff --git a/embedchain/loaders/rss_feed.py b/embedchain/loaders/rss_feed.py
@@ -28,12 +28,11 @@ def serialize_metadata(metadata):
     @staticmethod
     def get_rss_content(url: str):
         try:
-            from langchain_community.document_loaders import \
-                RSSFeedLoader as LangchainRSSFeedLoader
+            from langchain_community.document_loaders import RSSFeedLoader as LangchainRSSFeedLoader
         except ImportError:
             raise ImportError(
                 """RSSFeedLoader file requires extra dependencies.
-                Install with `pip install --upgrade "embedchain[rss_feed]"`"""
+                Install with `pip install feedparser==6.0.10 newspaper3k==0.2.8 listparser==0.19`"""
             ) from None
 
         output = []

diff --git a/embedchain/loaders/sitemap.py b/embedchain/loaders/sitemap.py
@@ -12,7 +12,7 @@
     from bs4.builder import ParserRejectedMarkup
 except ImportError:
     raise ImportError(
-        'Sitemap requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
+        "Sitemap requires extra dependencies. Install with `pip install beautifulsoup4==4.12.3`"
     ) from None
 
 from embedchain.helpers.json_serializable import register_deserializable

diff --git a/embedchain/loaders/substack.py b/embedchain/loaders/substack.py
@@ -24,7 +24,7 @@ def load_data(self, url: str):
             from bs4.builder import ParserRejectedMarkup
         except ImportError:
             raise ImportError(
-                'Substack requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
+                "Substack requires extra dependencies. Install with `pip install beautifulsoup4==4.12.3`"
             ) from None
 
         if not url.endswith("sitemap.xml"):

diff --git a/embedchain/loaders/web_page.py b/embedchain/loaders/web_page.py
@@ -7,7 +7,7 @@
     from bs4 import BeautifulSoup
 except ImportError:
     raise ImportError(
-        'Webpage requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
+        "Webpage requires extra dependencies. Install with `pip install beautifulsoup4==4.12.3`"
     ) from None
 
 from embedchain.helpers.json_serializable import register_deserializable

diff --git a/embedchain/loaders/youtube_channel.py b/embedchain/loaders/youtube_channel.py
@@ -18,7 +18,7 @@ def load_data(self, channel_name):
             import yt_dlp
         except ImportError as e:
             raise ValueError(
-                "YoutubeLoader requires extra dependencies. Install with `pip install --upgrade 'embedchain[youtube_channel]'`"  # noqa: E501
+                "YoutubeChannelLoader requires extra dependencies. Install with `pip install yt_dlp==2023.11.14 youtube-transcript-api==0.6.1`"  # noqa: E501
             ) from e
 
         data = []

diff --git a/embedchain/loaders/youtube_video.py b/embedchain/loaders/youtube_video.py
@@ -5,14 +5,12 @@
 try:
     from youtube_transcript_api import YouTubeTranscriptApi
 except ImportError:
-    raise ImportError('YouTube video requires extra dependencies. Install with `pip install youtube-transcript-api "`')
+    raise ImportError("YouTube video requires extra dependencies. Install with `pip install youtube-transcript-api`")
 try:
     from langchain_community.document_loaders import YoutubeLoader
     from langchain_community.document_loaders.youtube import _parse_video_id
 except ImportError:
-    raise ImportError(
-        'YouTube video requires extra dependencies. Install with `pip install --upgrade "embedchain[dataloaders]"`'
-    ) from None
+    raise ImportError("YouTube video requires extra dependencies. Install with `pip install pytube==15.0.0`") from None
 from embedchain.helpers.json_serializable import register_deserializable
 from embedchain.loaders.base_loader import BaseLoader
 from embedchain.utils.misc import clean_string

diff --git a/examples/api_server/requirements.txt b/examples/api_server/requirements.txt
@@ -1,2 +1,12 @@
 flask==2.3.2
-embedchain[dataloaders]==0.0.78
+youtube-transcript-api==0.6.1 
+pytube==15.0.0 
+beautifulsoup4==4.12.3
+slack-sdk==3.21.3
+huggingface_hub==0.23.0
+gitpython==3.1.38
+yt_dlp==2023.11.14
+PyGithub==1.59.1
+feedparser==6.0.10
+newspaper3k==0.2.8
+listparser==0.19
diff --git a/examples/rest-api/requirements.txt b/examples/rest-api/requirements.txt
@@ -1,6 +1,24 @@
 fastapi==0.104.0
 uvicorn==0.23.2
+streamlit==1.29.0
 embedchain==0.1.3
-embedchain[streamlit, community, opensource, elasticsearch, opensearch, poe, discord, slack, whatsapp, weaviate, pinecone, qdrant, images, huggingface_hub, cohere, together, milvus, dataloaders, vertexai, llama2, gmail, json]==0.1.3
+slack-sdk==3.21.3 
+flask==2.3.3
+fastapi-poe==0.0.16
+discord==2.3.2
+twilio==8.5.0
+huggingface-hub==0.17.3
+embedchain[community, opensource, elasticsearch, opensearch, weaviate, pinecone, qdrant, images, cohere, together, milvus, vertexai, llama2, gmail, json]==0.1.3
 sqlalchemy==2.0.22
 python-multipart==0.0.6
+youtube-transcript-api==0.6.1 
+pytube==15.0.0 
+beautifulsoup4==4.12.3
+slack-sdk==3.21.3
+huggingface_hub==0.23.0
+gitpython==3.1.38
+yt_dlp==2023.11.14
+PyGithub==1.59.1
+feedparser==6.0.10
+newspaper3k==0.2.8
+listparser==0.19
diff --git a/examples/slack_bot/requirements.txt b/examples/slack_bot/requirements.txt
@@ -1 +1,3 @@
-embedchain[slack, poe]==0.1.7
+slack-sdk==3.21.3 
+flask==2.3.3
+fastapi-poe==0.0.16