Overview
Request 545194 revoked
Remove as we still want to drop fortune unless reimplemented using iconv instead of recode.
- Created by scarabeus_iv
- In state revoked
- Open review for Base:System / fortune
- Open review for repo-checker
- Open review for factory-staging
Request History
scarabeus_iv created request
Remove as we still want to drop fortune unless reimplemented using iconv instead of recode.
licensedigger accepted review
ok
factory-auto added fortune as a reviewer
Submission for fortune by someone who is not maintainer in the devel project (Base:System). Please review
factory-auto added repo-checker as a reviewer
Is this delete request safe?
factory-auto accepted review
ok
dimstar_suse set openSUSE:Factory:Staging:M as a staging project
Being evaluated by staging project "openSUSE:Factory:Staging:M"
dimstar_suse accepted review
Picked openSUSE:Factory:Staging:M
WernerFink declined review
IMHO this is more like an attack against free speech
WernerFink declined request
IMHO this is more like an attack against free speech
scarabeus_iv reopened request
maxlin_factory accepted review
Removing from openSUSE:Factory:Staging:M, re-evaluation needed
maxlin_factory added factory-staging as a reviewer
Requesting new staging review
dimstar_suse declined request
Guerilla drop of building packages with a clear counter by the maintainer are not acceptable. Work as a team to find a solution, then act accordingly
scarabeus_iv revoked request
this drop is not acceptable
Dear Werner, should I then translate it as you volunteering to maintain recode? Also, there are alternative implementations for fortune...
It looks like the upstream used to depend on iconv, but switched to recode (by debian maintainer???). See changelogs in fortune-mod_1.99.1-4.diff.gz.
And there's "bool No_recode = FALSE;" in the code. I simply guess we could change it to TRUE to disable recode dependence. By this maybe some characters cannot be shown correctly but at least it work in most case.
Not gonna help you. main() contains lines as follows
outer = recode_new_outer(true); request = recode_new_request (outer);
and that means, you have to link to recode. No runtime option will fix that.
Tom, your crusade against that unmaintained recode package is honorable but just filing delete requests can be considered offensive. I'm sure fortune can be migrated to a different library. Writing that code is probably easier than having the discussion here :)
wait
Are there actually any bugs/CVEs open against recode? I like it e.g. for its guessing of input (recode ..to_charset), for converting line endings (/LF), for it's "charsets" of e.g. "tex" and "html"...
$ echo "ä" | recode ..tex \"a $ echo "ä" | recode ..h4 ä
Do that with iconv!
I can't find real non-fixed bugs against recode in bugzilla. Why "should" recode be dropped? Just because it hasn't changed in a while? How about it's just solid as a rock? One might check updating to the 3.7 beta2, but besides that?
Hey, if I had written "hello.c" 25 years ago, would it be dropped now only because it's "unmaintained"? Yes, upstream of recode seems dead. But unless anything crops up, why drop it? Gentoo, whose maintainers are quite "trigger-happy" about dropping unmaintained stuff, like qt4 currently (and dropping anything depending on it) are merrily keeping recode (with patches, all available on github) so far.
Just my 2¢.
I can make you the maintainer of Base:System/recode, if you want.