Automate browser based workflows with AI
Python aiapiautomationbrowserbrowser-automationcomputergptllmplaywrightpowerautomatepuppeteerpythonrpaseleniumvisionworkflow
Normalize duplicate www URL prefixes (#6386)
Suchintan1796016a826b
diff --git a/skyvern/forge/sdk/schemas/task_generations.py b/skyvern/forge/sdk/schemas/task_generations.pyindex 54342ebd3..6968b9bcf 100644--- a/skyvern/forge/sdk/schemas/task_generations.py+++ b/skyvern/forge/sdk/schemas/task_generations.py@@ -1,7 +1,9 @@from datetime import datetimefrom typing import Any-from pydantic import BaseModel, ConfigDict, Field+from pydantic import BaseModel, ConfigDict, Field, field_validator++from skyvern.utils.url_validators import collapse_duplicate_www_prefixclass TaskGenerationBase(BaseModel):@@ -20,6 +22,13 @@ class TaskGenerationBase(BaseModel):llm_response: str | None = Nonesuggested_title: str | None = None+ @field_validator("url")+ @classmethod+ def normalize_url(cls, url: str | None) -> str | None:+ if not url:+ return url+ return collapse_duplicate_www_prefix(url)+class TaskGeneration(TaskGenerationBase):task_generation_id: strdiff --git a/skyvern/utils/url_validators.py b/skyvern/utils/url_validators.pyindex 592771661..7db2a565a 100644--- a/skyvern/utils/url_validators.py+++ b/skyvern/utils/url_validators.py@@ -25,6 +25,24 @@ def strip_query_params(url: str) -> str:return f"{parsed.scheme}://{host}{port_str}{parsed.path}"+def collapse_duplicate_www_prefix(url: str) -> str:+ try:+ parts = urlsplit(url)+ except ValueError:+ return url++ if not parts.netloc:+ return url++ userinfo, separator, host_port = parts.netloc.rpartition("@")+ if not host_port.lower().startswith("www.www."):+ return url++ host_port = host_port[4:]+ netloc = f"{userinfo}{separator}{host_port}" if separator else host_port+ return urlunsplit((parts.scheme, netloc, parts.path, parts.query, parts.fragment))++def prepend_scheme_and_validate_url(url: str) -> str:if not url:return url@@ -37,6 +55,8 @@ def prepend_scheme_and_validate_url(url: str) -> str:if not parsed_url.scheme:url = f"https://{url}"+ url = collapse_duplicate_www_prefix(url)+try:HttpUrl(url)except ValidationError: