tea/browsertrix
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

validation: ensure seed urls, and other url properties, are validated on POST by using pydantic HttpUrl type, fixes #277 (#278)

Browse Source
This commit is contained in:
Ilya Kreymer 2022-06-29 16:09:32 -07:00 committed by GitHub
parent 3fec2a9f82
commit 50c525853f
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
backend/btrixcloud/crawlconfigs.py
View File

@ -10,7 +10,7 @@ import re
from datetime import datetime from datetime import datetime
import pymongo import pymongo
from pydantic import BaseModel, UUID4, conint from pydantic import BaseModel, UUID4, conint, HttpUrl
from fastapi import APIRouter, Depends, HTTPException from fastapi import APIRouter, Depends, HTTPException
from .users import User from .users import User
@ -35,12 +35,12 @@ class ScopeType(str, Enum):
class Seed(BaseModel): class Seed(BaseModel):
"""Crawl seed""" """Crawl seed"""
url: str url: HttpUrl
scopeType: Optional[ScopeType] = ScopeType.PREFIX scopeType: Optional[ScopeType] = ScopeType.PREFIX
include: Union[str, List[str], None] include: Union[str, List[str], None]
exclude: Union[str, List[str], None] exclude: Union[str, List[str], None]
sitemap: Union[bool, str, None] sitemap: Union[bool, HttpUrl, None]
allowHash: Optional[bool] allowHash: Optional[bool]
depth: Optional[int] depth: Optional[int]
@ -49,7 +49,7 @@ class Seed(BaseModel):
class RawCrawlConfig(BaseModel): class RawCrawlConfig(BaseModel):
"""Base Crawl Config""" """Base Crawl Config"""
seeds: List[Union[str, Seed]] seeds: List[Union[HttpUrl, Seed]]
scopeType: Optional[ScopeType] = ScopeType.PREFIX scopeType: Optional[ScopeType] = ScopeType.PREFIX