Closed beszel closed 5 years ago
I agree this works correctly, but the commit history for this PR is difficult to make sense of. In particular, I note that "Run full Puppeteer #101" appears twice in the commit history, and I bet a bunch of other commits do as well. It looks like the duplicate commits are all from the popup
branch.
I don't want to introduce duplicate commits to the master branch, if we can help it. I think you should close this PR, cherry pick the commits relevant to this feature into a new branch based on master
, and make a new PR from the new branch.
Does this make sense to you? Or should we meet and figure it out together?
On thinking it over, I tried this at the console:
git checkout master git merge --squash grey-icon
The actual changes are in few enough files I'm okay with a squash merge. That means it's not necessary to clean up the commit history in this branch.
I'm a bit surprised not to see end-to-end tests of this feature. Is this still work in progress? Or did you decide that testing isn't feasible?
All the duplicate commits are actually left from branching off of popup
(#109), and you'll see duplicate commits there too. I then asked you and Professor Stratton about how to prevent that when rebasing. I didn't make a big deal of it since we're squashing anyway.
I think I assumed that end-to-end testing isn't feasible here, but I didn't look into it! My reasoning was that the icon isn't part of any DOM that Puppeteer can access. I'll look into it and see what I can do.
I see that #109 wasn't squashed, so master
now has duplicate commits. I wish I had said something while it was in review!
Whoops, I wish I had noticed that too! I may look into whether we can do something about that. Live and learn...
On Tue, Nov 19, 2019 at 8:40 AM Ian Hawkins notifications@github.com wrote:
I see that #109 https://github.com/glam-lab/degender-the-web/pull/109 wasn't squashed, so master now has duplicate commits. I wish I had said something while it was in review!
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/glam-lab/degender-the-web/pull/113?email_source=notifications&email_token=AEXJIU5YBKAPVNVPNNFISVLQUQJGDA5CNFSM4JNTL6A2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEEO3MGA#issuecomment-555595288, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEXJIU5E6GTGDAOAG2RMZODQUQJGDANCNFSM4JNTL6AQ .
-- Janet Davis (she/her or they/them) Associate Professor and Microsoft Chair of Computer Science, Whitman College 509-527-5758, http://cs.whitman.edu/~davisj/ Schedule an appointment: http://meetme.so/JanetDavis
It looks like there's no way to check which version of the icon the user is seeing on a particular page, so end-to-end tests on this feature aren't feasible. The closest thing is the existing tests for popup status messages.
I don't think anything needs to be added to the manual test page.
For manual testing as part of this code review, I reccommend this comment to see how the icon should behave on different pages.
Thanks, Ian, this makes sense. For future reference, can you add the gist of that comment to the manual test page?
On Sat, Nov 23, 2019 at 4:56 PM Ian Hawkins notifications@github.com wrote:
It looks like there's no way to check which version of the icon the user is seeing on a particular page, so end-to-end tests on this feature aren't feasible. The closest thing is the existing tests for popup status messages.
I don't think anything needs to be added to the manual test page.
For manual testing as part of this code review, I reccommend this comment https://github.com/glam-lab/degender-the-web/issues/96#issuecomment-551963448 to see how the icon should behave on different pages.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/glam-lab/degender-the-web/pull/113?email_source=notifications&email_token=AEXJIU75W25BUXGCSIAJH3DQVHGLLA5CNFSM4JNTL6A2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEFABESA#issuecomment-557847112, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEXJIU6NOTXE2DGJSCWMLU3QVHGLLANCNFSM4JNTL6AQ .
-- Janet Davis (she/her or they/them) Associate Professor and Microsoft Chair of Computer Science, Whitman College 509-527-5758, http://cs.whitman.edu/~davisj/ Schedule an appointment: http://meetme.so/JanetDavis
Cleaning up master
and rebasing took care of the duplicate commits here. I am making additional notes on the manual tests page per your comments on #96.
Merged from the command line.
Resolves #96