github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-09-08 22:42:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

community[patch]: Add classifier_url argument in PebbloSafeLoader and documentation update. (#21030)

Browse Source 
Description: Add classifier_url argument in PebbloSafeLoader.
Documentation: Updated PebbloSafeLoader documentation with above change
and new links for pebblo github pages.

---------

Signed-off-by: Rahul Tripathi <rauhl.psit.ec@gmail.com>
Co-authored-by: Rahul Tripathi <rauhl.psit.ec@gmail.com>
This commit is contained in:
Rahul Triptahi
2024-04-30 03:11:09 +05:30
committed by GitHub
parent 08d08d7c83
commit c172611647
2 changed files with 9 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
libs/community/langchain_community/document_loaders/pebblo.py
View File

@@ -45,6 +45,7 @@ class PebbloSafeLoader(BaseLoader):
description: str = "",
api_key: Optional[str] = None,
load_semantic: bool = False,
classifier_url: Optional[str] = None,
):
if not name or not isinstance(name, str):
raise NameError("Must specify a valid name.")
@@ -63,6 +64,7 @@ class PebbloSafeLoader(BaseLoader):
self.source_type = get_loader_type(loader_name)
self.source_path_size = self.get_source_size(self.source_path)
self.source_aggregate_size = 0
self.classifier_url = classifier_url or CLASSIFIER_URL
self.loader_details = {
"loader": loader_name,
"source_path": self.source_path,
@@ -210,7 +212,7 @@ class PebbloSafeLoader(BaseLoader):
self.source_aggregate_size
)
payload = Doc(**payload).dict(exclude_unset=True)
load_doc_url = f"{CLASSIFIER_URL}{LOADER_DOC_URL}"
load_doc_url = f"{self.classifier_url}{LOADER_DOC_URL}"
classified_docs = []
try:
pebblo_resp = requests.post(
@@ -296,7 +298,7 @@ class PebbloSafeLoader(BaseLoader):
"Content-Type": "application/json",
}
payload = self.app.dict(exclude_unset=True)
app_discover_url = f"{CLASSIFIER_URL}{APP_DISCOVER_URL}"
app_discover_url = f"{self.classifier_url}{APP_DISCOVER_URL}"
try:
pebblo_resp = requests.post(
app_discover_url, headers=headers, json=payload, timeout=20