Conversation
…che#3981) Collects and simplify all path exclusion logic into a single class configured in web.xml and makes our reliance on jetty's DefaultServlet (or whatever is configured with the name 'default') for serving static content explicit rather than an implicit feature one has to figure out (or already know about). (cherry picked from commit 58e11d8)
|
It all seems suitable to backport thru 10x. Why a PR for the backport? The one reason for a PR dedicated to a backport is that it's not straight-forward and you want peer review on those particular parts, and you commented on those particular parts so the reviewer knows where to look. |
I find it easier to always push the same buttons. Also it gives me a second pass where I can look to see that I'm not publishing anything I didn't mean to... cheap safety for me. |
|
I agree, I like to do backport PR for non trivial code, where backport differs slightly from original commit or just to get a confirmation with all the CI tests running. |
|
The commit message of this backport has no information other than a JIRA reference, which is very inconvenient when scanning the commit log. When I cherry-pick in my IDE; I get everything, plus the cherry-pick source hash. |
3 participants
https://issues.apache.org/jira/browse/SOLR-18041
Not sure if everything that we do for SOLR-18040 (the parent ticket) will be suitable for 10.1+, but This seems simple enough that it should go through.