fix: packer validate unsupported type error
#13245
Conversation
`packer validate` would output the same error message four times per unsupported root block type found in a template (e.g., 'src' instead of 'source'). This behavior was due to a function being called four times for each file on each stage of the parsing.
|
|
|
Thank you for your submission! We require that all contributors sign our Contributor License Agreement ("CLA") before we can accept the contribution. Read and sign the agreement Learn more about why HashiCorp requires a CLA and what the CLA includes Have you signed the CLA already but the status is still pending? Recheck it. |
There was a problem hiding this comment.
Hey @gstvcruz,
Thanks for the PR! I have to say though, this kinda conflicts with PR #12607, which has been stale for a while, but reworks how parsing works for HCL2 templates.
The approach is slightly different as the other PR processes files once, and within the files all blocks are processed in one phase, compared to the current approach which relies on the file being parsed for every type of top-level block we're looking for.
That said, your approach seems functional (I tested it quickly on a couple templates I had lying around), and is definitely less intrusive than the other, so I can envision a path where yours is merged first in order to silence those extra error diags, then we can focus our efforts on the other PR to refactor how parsing is processed.
I don't necessarily advocate for one over the other, and while biased as I submitted the other one, I think we should discuss which one we should merge eventually, or both even maybe.
We'll discuss this in January since most of us are out of office for the holidays now, in the meantime thanks again for reporting the issue and tackling it, much appreciated!
|
Hey @lbajolet-hashicorp, thanks for your reply! I believe refactoring the parsing process would be a more effective solution since the additional error diags are just a side effect of the current implementation. By refactoring the parsing, this bug would be resolved entirely. That said, the changes I proposed could act as a temporary measure to address the issue until the parsing refactor is merged. I’d like to start working on the parser refactoring. Does the PR you mentioned provide enough information for me to get started? Also, why has it gone stale? Could you provide some context, please? Thanks for you time! |
|
Hi again @gstvcruz, Regarding the PR, it got stale mostly because we got sidetracked working on other things, this wasn't a high-priority issue to fix, plus there were a couple of comments on that that I'm not sure were addressed completely, so it needs discussions and a solid rebase as there are conflicts at the moment. I do agree with you though, your PR is definitely simpler and would fix the issue in the meantime, so as I mentioned, I am not opposed to merging it in time of our next release, I'll bring this up when people are back from holidays, and we can decide what to do with it then. Thanks for the response! |
packer validatewould output the same error message four times per unsupported root block type found in a template (e.g., 'src' instead of 'source'). This behavior was due to a function being called four times for each file on each stage of the parsing.Closes #13241