Marcus Read

4.2.4 released as required (checked manually), although the [workflow failed](https://github.com/gerrymanoim/exchange_calendars/actions/runs/3469495398/jobs/5796555469) on the subsequent 'Install and import' job. The analogous job to check the upload to the test repo ran fine,...

Workflow failed for [release 4.2.5](https://github.com/gerrymanoim/exchange_calendars/actions/runs/4025867183/jobs/6919638847) for the same reason as 4.2.4 (although release was ok). Workflow passed for [release 4.2.6](https://github.com/gerrymanoim/exchange_calendars/actions/runs/4649600491/jobs/8228048478) 🤷‍♂️ .

[Workflow](https://github.com/gerrymanoim/exchange_calendars/actions/runs/4932403341/jobs/8815384600) passed for release 4.2.7. [Workflow](https://github.com/gerrymanoim/exchange_calendars/actions/runs/5078270647/jobs/9122567973) failed for release 4.2.8 on trying to install the newly published version (although release fine, checked manually). Appears that the workflow is failing in...

[Workflow](https://github.com/gerrymanoim/exchange_calendars/actions/runs/6098877213/job/16549724308) failed for release 4.3 in same manner as it failed on 4.2.8. (Again, the actual release was fine - checked manually.) Going to add an echo to identify if...

I have the same [workflow](https://github.com/maread99/valimp/blob/0.2/.github/workflows/release.yml) running in the [valimp](https://github.com/maread99/valimp) library and was able to identify [here](https://github.com/maread99/valimp/actions/runs/6984860994/job/19008325978) that it's failing on the `((i++))`. @gerrymanoim, my knowledge of bash is pretty poor,...

Thanks @gerrymanoim, looks like that could well be it. According to [the ref you included](https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#jobsjob_idstepsshell) *If bash is not found in the path, this is treated as sh.* Ahead of...

I've explicitly defined bash as the shell (#348). 4.5.1 released fine although doesn't appear to have entered the loop so will have to hold off judgement on whether that's done...

4.5.2 did enter the loop and the [workflow failed](https://github.com/gerrymanoim/exchange_calendars/actions/runs/7730390167/job/21075703506), regardless of the shell being defined as bash. (NB published to PyPI fine, just failed on the subsequent install and import...

`let i++` doesn't make any difference. [Failed](https://github.com/maread99/market_prices/actions/runs/7757679768/job/21157705858) on a market-prices release that I'd tried it on. I'm pretty much out of ideas. Going to add a pre-emptive 5 second sleep...

Ran fine with the changes for the 4.5.3 release. Didn't enter the loop although with the pre-emptive 5s sleep between publishing and installing I guess there's a reasonable chance it...