jyapayne/langflow
1
0
Fork 0
Code Issues Pull requests Projects Releases 1 Packages Wiki Activity Actions

added firecrawl integration

Browse source
This commit is contained in:
rafaelsideguide 2024-06-25 17:10:29 -03:00
commit d4c18d3c00
10 changed files with 291 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

81
src/backend/base/langflow/components/langchain_utilities/FirecrawlCrawlApi.py Normal file
View file

@ -0,0 +1,81 @@
from typing import Optional
from firecrawl.firecrawl import FirecrawlApp
from langflow.custom import CustomComponent
from langflow.schema import Data
import uuid
class FirecrawlCrawlApi(CustomComponent):
display_name: str = "FirecrawlCrawlApi"
description: str = "Firecrawl Crawl API."
output_types: list[str] = ["Document"]
documentation: str = "https://docs.firecrawl.dev/api-reference/endpoint/crawl"
field_config = {
"api_key": {
"display_name": "API Key",
"field_type": "str",
"required": True,
"password": True,
"info": "The API key to use Firecrawl API.",
},
"url": {
"display_name": "URL",
"field_type": "str",
"required": True,
"info": "The base URL to start crawling from.",
},
"timeout": {
"display_name": "Timeout",
"field_type": "int",
"info": "The timeout in milliseconds.",
},
"crawlerOptions": {
"display_name": "Crawler Options",
"info": "Options for the crawler behavior.",
},
"pageOptions": {
"display_name": "Page Options",
"info": "The page options to send with the request.",
},
"idempotency_key": {
"display_name": "Idempotency Key",
"field_type": "str",
"info": "Optional idempotency key to ensure unique requests.",
},
}
def build(
self,
api_key: str,
url: str,
timeout: Optional[int] = 30000,
crawlerOptions: Optional[Data] = None,
pageOptions: Optional[Data] = None,
idempotency_key: Optional[str] = None,
) -> Data:
if crawlerOptions:
crawler_options_dict = crawlerOptions.__dict__['data']['text']
else:
crawler_options_dict = {}
if pageOptions:
page_options_dict = pageOptions.__dict__['data']['text']
else:
page_options_dict = {}
if not idempotency_key:
idempotency_key = str(uuid.uuid4())
app = FirecrawlApp(api_key=api_key)
crawl_result = app.crawl_url(
url,
{
"crawlerOptions": crawler_options_dict,
"pageOptions": page_options_dict,
},
True,
int(timeout / 1000),
idempotency_key
)
records = Data(data={"results": crawl_result})
return records

72
src/backend/base/langflow/components/langchain_utilities/FirecrawlScrapeApi.py Normal file
View file

@ -0,0 +1,72 @@
from typing import Optional
from firecrawl.firecrawl import FirecrawlApp
from langflow.custom import CustomComponent
from langflow.schema import Data
from langflow.services.database.models.base import orjson_dumps
import json
class FirecrawlScrapeApi(CustomComponent):
display_name: str = "FirecrawlScrapeApi"
description: str = "Firecrawl Scrape API."
output_types: list[str] = ["Document"]
documentation: str = "https://docs.firecrawl.dev/api-reference/endpoint/scrape"
field_config = {
"api_key": {
"display_name": "API Key",
"field_type": "str",
"required": True,
"password": True,
"info": "The API key to use Firecrawl API.",
},
"url": {
"display_name": "URL",
"field_type": "str",
"required": True,
"info": "The URL to scrape.",
},
"timeout": {
"display_name": "Timeout",
"info": "Timeout in milliseconds for the request.",
"field_type": "int",
"default_value": 10000,
},
"pageOptions": {
"display_name": "Page Options",
"info": "The page options to send with the request.",
},
"extractorOptions": {
"display_name": "Extractor Options",
"info": "The extractor options to send with the request.",
},
}
def build(
self,
api_key: str,
url: str,
timeout: Optional[int] = 10000,
pageOptions: Optional[Data] = None,
extractorOptions: Optional[Data] = None,
) -> Data:
if extractorOptions:
extractor_options_dict = extractorOptions.__dict__['data']['text']
else:
extractor_options_dict = {}
if pageOptions:
page_options_dict = pageOptions.__dict__['data']['text']
else:
page_options_dict = {}
app = FirecrawlApp(api_key=api_key)
results = app.scrape_url(
url,
{
"timeout": str(timeout),
"extractorOptions": extractor_options_dict,
"pageOptions": page_options_dict,
},
)
record = Data(data=results)
return record

16
src/backend/base/poetry.lock generated
View file

@ -739,6 +739,20 @@ typer = ">=0.12.3"
[package.extras]
standard = ["fastapi", "uvicorn[standard] (>=0.15.0)"]
[[package]]
name = "firecrawl-py"
version = "0.0.16"
description = "Python SDK for Firecrawl API"
optional = false
python-versions = ">=3.8"
files = [
{file = "firecrawl_py-0.0.16-py3-none-any.whl", hash = "sha256:9024f483b501852a6b9c4e6cdfc9e8dde452d922afac357080bb278a0c9c2a26"},
{file = "firecrawl_py-0.0.16.tar.gz", hash = "sha256:6c662fa0a549bc7f5c0acb704baba6731869ca0451094034264dfc1b4eb086e4"},
]
[package.dependencies]
requests = "*"
[[package]]
name = "frozenlist"
version = "1.4.1"
@ -3232,4 +3246,4 @@ local = []
[metadata]
lock-version = "2.0"
python-versions = ">=3.10,<3.13"
content-hash = "4f566531a8539ddc81cb91a7e7f9b723c84679f0af5bb8619f7b02f9ffc6cfaa"
content-hash = "7e46144d27c633214f00e73e496c0e4d56db1fb47032a21861677ec275b79d86"

1
src/backend/base/pyproject.toml
View file

@ -64,6 +64,7 @@ pyperclip = "^1.8.2"
uncurl = "^0.0.11"
sentry-sdk = {extras = ["fastapi", "loguru"], version = "^2.5.1"}
chardet = "^5.2.0"
firecrawl-py = "^0.0.16"
[tool.poetry.extras]