I Sometimes wonder if the errors are deliberately ...
# pact-php
l
I Sometimes wonder if the errors are deliberately obtuse... https://github.com/pact-foundation/pact-php/runs/6570723077?check_suite_focus=true
Added the new test.pactflow.io url to this now instead of test.pact.dius.com.au Makes me think this is not run often enough.
y
Ahh yes, the url was decommissioned on the 2nd May https://docs.pactflow.io/notices/2022-03-28-pact-dius-com-au-decommission and the last changes were 29 days ago to master, so that slipped the net. Comms and co-ordination across the repos for maintainers is high on my list at the moment, as well of ideas to build robust and similar workflows across the libraries to ensure that when changes are made, they are appropriately tested and filtered through. The process is quite slick on some repos which are heavily used, but others are maintained often by single maintainers, so the extra eyes and feedback of new contributors and maintainers alike is really valuable. Appreciate the extra effort going in to make debugging easier to distinguish in future. Will co-ordinate with the maintainers and key contributors and see what we can rustle up, ensuring that we do it in the open so others can get involved
Obtuse errors frustrate me, I watched a video where someone mentioned one of the best things they did was every time they got an error and it was just garbage on their screen, they would clean up the logging to make it easier to grok for others. Might have been a podcast
m
Yes, that is strange. It’s hard to know where/why that error is occurring.
can you reproduce locally?
l
test.pact.dius.com.au won't work while decommissioned Although it is failed in pactflow branch, https://github.com/Lewiscowles1986/pact-php/actions/runs/2380311568 shows this same code passing. This seems to be due to race conditions.
Also while I might abstractly agree on improving logging, if you know where the logs are coming from. a large part of the suggested approach creates extra line-count in PR's and patches, which are then harder to read, review, and inevitably take more time to author, test; and with a sample size of 1 (me) might just be opinion pieces.
large issue capturing. It can be closed if it is an opinion piece https://github.com/pact-foundation/pact-php/issues/250