Git-Mirrors/brozzler
1
0
Fork 0
mirror of https://github.com/internetarchive/brozzler.git synced 2025-04-20 23:56:34 -04:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Merge pull request #257 from vbanos/screenshot-on-successful-capture

Browse Source 
Thanks, @vbanos!
This commit is contained in:
Barbara Miller 2023-07-17 15:59:12 -07:00 committed by GitHub
commit 740019cc18
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 9 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
brozzler/browser.py
View File

@ -503,9 +503,8 @@ class Browser:
page_url)
self.navigate_to_page(page_url, timeout=page_timeout)
# If the target page HTTP status is 4xx/5xx, there is no point
# in running behaviors, outlink and hashtag extraction as we
# didn't get a valid page. Screenshot should run because i
# may be useful to have a picture of the error page.
# in running behaviors, screenshot, outlink and hashtag
# extraction as we didn't get a valid page.
# This is only enabled with option `simpler404`.
run_behaviors = True
if simpler404 and (self.websock_thread.page_status is None or
@ -519,7 +518,13 @@ class Browser:
self.run_behavior(behavior_script, timeout=behavior_timeout)
final_page_url = self.url()
if on_screenshot:
self._try_screenshot(on_screenshot, screenshot_full_page)
if simpler404:
if self.websock_thread.page_status and \
self.websock_thread.page_status < 400:
self._try_screenshot(on_screenshot, screenshot_full_page)
else:
self._try_screenshot(on_screenshot, screenshot_full_page)
if not run_behaviors or skip_extract_outlinks:
outlinks = []
else: