Re: [PATCH] Fix ob-plantuml over TRAMP

[Ihor Radchenko]

guillaume@gbuisson.com writes:

> It seems to me that there is a redundant check which uses `file-exists-p` 
> even if using TRAMP, the plantuml jar is located on the remote server, so 
> it fails with an error even though the command to actually render the 
> diagrams works fine.

I disagree that the check is redundant. It is still a good idea to check
if plantuml jar exists on server itself. Another question is how to make
`file-exists-p' work with TRAMP.

> The patch I propose involves keeping this check but moving it so it only 
> generates a message instead of failing the whole generation process,
> doing so allows to restore the functionality and I can finally generate my 
> sequence diagrams over TRAMP.

Your patch will break legitimate check for people running plantuml
locally. A better way would be checking default-directory and if we are
in a remote server, using appropriate check working over TRAMP.

Best,
Ihor