Re: bug: Error handling in source blocks.

[<tomas@tuxteam.de>]

[-- Attachment #1: Type: text/plain, Size: 984 bytes --]

On Wed, Aug 11, 2021 at 06:30:26AM +1000, Tim Cross wrote:

>                                      [...] For example, in an
> interpreted language, you could have errors due to problems with the
> interpreter, you could have errors in the code or you could have a code
> block which legitimately returns an error [...]

Yes, your error might be my nugget. For one illustrative example, I
once cobbled up a small shell tutorial. Babel seemed like the right
thing -- but I needed script snippets erroring out to show off.

Needless to say, Babel freaked out, not only on non-zero exit codes,
but also on any hint of activity on stderr.

I cheated myself out with preambles and postambles, which redirected
stderr to stdout and lied about exit codes.

This is, of course, horribly ugly, because it's language-specific
(you'd have to do it for C or Scheme or Perl or...), although the
concepts of exit code and stderr are OS wide. We could do better :)

Just one data point.

Cheers
 - t

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]