klbr/khoj
1
0
Fork 0
mirror of https://github.com/khoaliber/khoj.git synced 2026-03-06 05:39:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

Make back action in browser environment use goto to avoid timeouts

Browse Source 
For some reason the page.go_back() action in playwright had a much
higher propensity to timeout. Use goto instead to reduce these page
traversal timeouts.

This requires tracking navigation history.
This commit is contained in:
Debanjum
2025-05-08 08:26:47 -06:00
parent fd139d4708
commit 38bcba2f4b
1 changed files with 17 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
src/khoj/processor/operator/operator_environment_browser.py
View File

@@ -27,6 +27,7 @@ class BrowserEnvironment(Environment):
self.height: int = 768 self.height: int = 768
self.visited_urls: Set[str] = set() self.visited_urls: Set[str] = set()
self.excluded_urls = {"about:blank", "https://duckduckgo.com", "https://www.bing.com", "https://www.google.com"} self.excluded_urls = {"about:blank", "https://duckduckgo.com", "https://www.bing.com", "https://www.google.com"}
self.navigation_history: list[str] = []
async def start(self, width: int = 1024, height: int = 768) -> None: async def start(self, width: int = 1024, height: int = 768) -> None:
self.width = width self.width = width
@@ -48,7 +49,13 @@ class BrowserEnvironment(Environment):
# Define a handler for page load events to capture URLs # Define a handler for page load events to capture URLs
async def handle_load(loaded_page: Page): async def handle_load(loaded_page: Page):
url = loaded_page.url url = loaded_page.url
if url and url not in self.excluded_urls and url not in self.visited_urls: if not url:
return
if not self.navigation_history or self.navigation_history[-1] != url:
self.navigation_history.append(url)
if url not in self.excluded_urls and url not in self.visited_urls:
logger.debug(f"Page loaded: {url}") logger.debug(f"Page loaded: {url}")
self.visited_urls.add(url) self.visited_urls.add(url)
@@ -252,9 +259,15 @@ class BrowserEnvironment(Environment):
logger.debug(f"Action: {action.type} to {url}") logger.debug(f"Action: {action.type} to {url}")
case "back": case "back":
await self.page.go_back() if len(self.navigation_history) > 1:
output = "Navigated back" self.navigation_history.pop()
logger.debug(f"Action: {action.type}") previous_url = self.navigation_history[-1]
await self.page.goto(previous_url)
output = f"Navigated back to {previous_url}"
else:
output = "No previous URL to navigate back"
previous_url = "about:blank"
logger.debug(f"Action: {action.type} to {previous_url}")
case _: case _:
error = f"Unrecognized action type: {action.type}" error = f"Unrecognized action type: {action.type}"