RE: new cvs committer, some ideas
"Welch, Brent" <[email protected]> Wed, 18 Jun 2008 23:39:46 -0700
| Newsgroups | gmane.mail.exmh.devel |
|---|---|
| Message-ID | <[email protected]> |
I'm OK with install(dir,mumble) for the variable names - sorry for the stray reference to exmh() -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Alexander Zangerl Sent: Wednesday, June 18, 2008 6:26 PM To: [email protected] Subject: Re: new cvs committer, some ideas On Wed, 18 Jun 2008 09:28:43 MST, "Welch, Brent" writes: >The one true way to fix this is to make sure that the paths are >associated with exmh() variables, and those are set in the main exmh, >exmh-bg, exmh-async scripts. i agree, and my existing code already does exactly that. (with the one minor difference that i used a separate install() var for the directories rather than messing with exmh(). i'll change that to suitably named keys for exmh() before committing it.) >Frankly, I'm not familiar with how the package maintainers >co-opt/work-around the built in installation schemes. in some cases you're better off not knowing...it very much depends on how large the spiritual gap between upstream's view of the universe and yours is. the exmh installer is "average" in that regard, not trivial to convince to work but nowhere near being a major nightmare. >Just don't mess it up :-) i don't plan to, and mostly i manage to stick to my plans :-) regards az -- + Alexander Zangerl + DSA 42BD645D + (RSA 5B586291) > hence the AIX boxen of increasing power [...] That phrasing reminded me of "Rodents of Unusual Size". -- Mike and Greg Andrews