search

gildas-lormeau / single-file-cli

CLI tool for saving a faithful copy of a complete web page in a single HTML file (based on SingleFile)
GNU Affero General Public License v3.0
644 stars 63 forks source link

Not working: --browser-load-max-time (no timeout, browser running "forever") #134

Open emp-00 opened 1 month ago

emp-00 commented 1 month ago

Apparently, in the current release 2.0.68 (not sure if already since a longer time) there's an issue with the browser loading timeout which does not work as expected. I'm on Win11 and using thorium browser (chromium derivative, running fine with single-file since half a year).

Symptom: a) single-file.exe executed in directory C:\Browser\Thorium_Portable\ with below command-line ==> single-file.exe and two thorium.exe processes are running forever (more than 5 minutes) and network activity indicates that the browser is still running and probably downloading b) same as above, additionally using "--browser-load-max-time=10000" in the command line (expecting 10 seconds timeout) ==> same as above, running forever

Expected behavior: Reference: "--browser-load-max-time: Maximum delay of time to wait for page loading in ms (default: 60000)" a) expecting timeout after 60 seconds terminating single-file.exe and the thorium.exe child processes with error logging b) expecting timeout after 10 seconds terminating single-file.exe and the thorium.exe child processes with error logging

Am I misinterpreting single-file's functionality or is this a bug? Thanks for looking into this!

command-line: single-file.exe --browser-executable-path="C:\Browser\Thorium_Portable\bin\thorium.exe" --browser-arg="--user-data-dir=C:\Browser\Thorium_Portable\user_data" --browser-width=640 --browser-height=1200 --block-fonts --compress-HTML=false --error-file="test_ERROR-Log.txt" --filename-conflict-action=overwrite --urls-file="test_URL-List.txt" --filename-template="test_{url-pathname-flat}{url-search}.html"

test_URL-List.txt (these are large-file download URL's, just for testing "long loading pages" for testing the timeout functionality):

https://link.testfile.org/30MB
https://testfile.org/files-5GB
https://mp4-download.com/8k-4-MP4
gildas-lormeau commented 1 month ago

That was not working because the URLs are not HTML pages and trigger downloads in the browser instead of loading a page. I fixed this particular issue in the version I've just published (downloads are now denied). In other cases, I was not able to reproduce the issue related to --browser-load-max-time.

emp-00 commented 1 month ago

Alright - got it. Actually I only used those url's in order to test single-file with pages that take a long time to load (see https://github.com/gildas-lormeau/single-file-cli/issues/135 --> I just downloaded and started testing your new revision, will take some time to collect sufficient data in my logs).

Reason for testing with those url's: My target url's are only sometimes slow in response causing "my" timeout issues. If anybody has a good url in the open internet for testing "slow response / slow loading / webpages" let me know! This would speed up my testing of the other issue above.... :-)