[rust] Support multiple names for Firefox binary in Linux (#17695)#17732
Conversation
PR Summary by Qodo[rust] Detect Firefox binaries with alternate Linux names
AI Description
Diagram
High-Level Assessment
Files changed (1)
|
Code Review by Qodo
Context used✅ Tickets:
🎫 2.48 doesn't trigger javascript in link's href on click() 🎫 Instance ChroneDriver Error: ConnectFailure✅ Compliance rules (platform):
14 rules 1. Linux channel path bypass
|
|
Thanks for working on this @bonigarcia! I ran selenium-manager-linux on a system with both
|
f225b48 to
eb2b83b
Compare
|
Code review by qodo was updated up to the latest commit eb2b83b |
|
Thanks for testing it, @kevinoid. Indeed, the implementation was not fully correct. I tried to fix it. Can you please have a try with the new binary? |
|
Thanks for the updates @bonigarcia! This version works correctly for
|
|
@kevinoid. Thanks for testing, it. But I am unable to reproduce your problems. I don't understand a couple of parts of your execution. First, these traces: For some reason, Selenium Manager is not able to keep the Firefox version at that point. That Second, in all of your executions, there is a couple of the following traces: Which means that Selenium Manager tries to create the metadata file twice. Which makes no sense to me, since the first time the metadata file does not exist, it is created by Selenium Manager. So it should be there the second time this file is read (which happens a few ms after). I don't have an explanation for that. Do you have some special settings in your cache folder? |
|
Here's my avoid-browser-download = true
avoid-stats = trueIs there anything else I can do to help investigate the behavior we are observing? I'd be happy to run other debug builds with more logging, send |
|
Code review by qodo was updated up to the latest commit 70d00d1 |
|
@kevinoid Indeed, the use of In any case, two of your tests are going to continue failing: Regarding the first one, it makes senses that this execution fail since you are requesting to use Firefox (stable, by default), and it is not installed in your system (it seem it installed beta, dev, etc.) but not stable. And you are specifying not download browsers. So it makes sense a failure to me. The second is a bit different since you are specifying a fixed version but you are avoiding browser download. Even when the unstable might correspond to that version, Selenium Manager does not check every single possibility. |
My apologies for not checking it sooner! Thanks for adding support for this configuration as well.
It appears to be behaving as expected, as far as I can see:
|
|
Code review by qodo was updated up to the latest commit 87bce21 |
🔗 Related Issues
Fixes #17695.
💥 What does this PR do?
This PR makes SM to support different Firefox names (
firefox-esr,firefox-beta,firefox-devedition,firefox-trunk) in Linux.🔧 Implementation Notes
🤖 AI assistance
💡 Additional Considerations
🔄 Types of changes