Closed agzam closed 5 months ago
Thanks for reporting this. The idea was to always keep one request spec per Org heading - currently all of Verb works like this. In your case, you could add two level-3 headings for each request spec. Would that work with your workflow, or does it break something else?
I think you're missing the point. I understand that probably, initially, you built it with the assumption that it primarily will be used within a hierarchy of headings. However, verb-send-request-on-point
incidentally (or by design) also works with source blocks.
And that may mislead people to think that it works for any source block, after all the command is called -on-point
, not -on-point-within-a-heading
. That is exactly what happened to me. I spent some time scratching my head, wondering why data for two different endpoints looks the same.
I think it would be wise to make it either fail with a message or make it do the right thing. This is exactly what I'm trying to do in https://github.com/federicotdn/verb/pull/55
Of course, I'm not arguing for the correctness of that approach (you may suggest doing it differently), but fundamentally, I think, it is a problem worth fixing.
Finally fixed this! You were right; additionally, I had already more or less planned for this scenario (sending requests embedded in Babel blocks using verb-send-request-on-point), only that my solution was to simply take the first block. See: https://github.com/federicotdn/verb/commit/b878d52ebc8214b2aedd59165aa89f8d918ed95c
verb-send-request-on-point
works for both - tree-structured and source-block approaches. But it doesn't seem to correctly detect the source block at point, when there are multiple within the same heading, i.e.,when the cursor is at the second source code, running
verb-send-request-on-point
should work for the second one, currently, it sends the request for the first source block. That basically makes problematic to keep multiple source blocks within the same heading.verb-export-request-on-point-curl
, as it looks like, also has the same issue, it doesn't correctly detect the source block at point.