defon another front, in #SeaMonkey, Sompi compained that the links to nightly buiklds and langpacks on this page seamonkey-project.org/dev/nightly didn;t work. I added redirect intdex.html pages on my website to tgtrdh to the appropiate locain on stchive.seamonkey-project.org

this should really be altered on the seamonky-project webite to just go directly to the correct place on aerchive and avoid these redirects.

alternatively could just have one link to wg9s.com/seamonkey-253-nightly and change the text to say it has builds, langpacks, and source tarballs.

what do you think?

wron channel, sorry

frg: how do you accept contribution to SeaMonkey web site repository gitlab.com/seamonkey-project/website ?

not sure now but you can send me a patch via a pm

iss this for 2.53?

if for comm-central not so much

only updates to seamonkey-project.org/dev/nightly to link directly your wg9s.com/seamonkey-253-nightly

WG9s: why do you not update it yourself and publish the website from the repo ?

bet thivest is to file a bug and attach the patch

the ieas is to not be dependent as mush on my website being up now that we have an offial setmonkey-project.org site

as you mention here:

so only need to have backup servers and redunancy on the official site so my webserver is not single point of failure still point of failure for builds but once build and posted to archive a more officially supported site

will be fising the links to go direactly to the officai site but we have some issues hoingf on preventing updateing the sebsite so i made the incorrect links work via redireects on my website

ircbot.comm-central.org:8080/seamonkey/20240227#c225700 the info in dev/nightly contains outdated links which can easily be updated with a standard GIT pull request followed by a merge

WG9s: as I clear 2 year old moonchild crap off my older server there is a bunch of space could you use a 10-20gbs as a mirror

guest_: I am just leaving the channel now. yhank you so much for being a complete asshole!

"standard git pull request" isn't a thing, is it?

AFAIK "pull requests" are a workflow mostly specific to github's web interface

gitlab does support forking a repos and also pull request - this is not a github specific workflow

I did say "mostly". It's far from "standard git"

with pull & merge request you can keep the master repo clean from all experimental commits

agree but definitively a good idea

hm, yeah, but last I checked these things didn't really map well to the usual git workflow

WG9s: why have you been so friendly to me ?

with git and mercurial you can export patches, and then attach to bugzilla or otherwise share them (but usually it'd go into bugzilla)

with git you'll reach more developers than with hg / mercurial

I'll have to see the data on that

that sounds more like something I'd see on news.ycombinator.com :-P

in reality, both are DVCS systems, so there isn't even a so big difference in how the "clone a remote, work locally" thing works

many larger company are switching to Git repository - is this not a big sign about the future of hg/mercurial ?

I can't quite wrap my mind around the concept some people would be turned away because things are in the wrong kind of DVCS

guest_: are switching to git from what?

and how can you be sure about their reasons to switch?

and also, more importantly: that something is used, does not mean it's the only acceptable choice, or the only one that works

SourceSafe, SVN, TFVC

this could easily be all emacs v. vim again

sure but for the time beeing more developers are common to git than to anything difference as long as they're not living in a mainframe env

^different

if SM project team is comitted to hg/mercurial why are they up-to-date repos maintained on gitlab.com ? reaching a wider audience ?

guest_: Please stop whatever this is. SeaMoneky's development status quo is a bit, unorthadox at the moment but it also happens to be working for them for now. WG9s was more than kind offering to personally accept your patches rather than basically the same proceedure it has always been.. File a bug attach a patch and satisfy the project requirements.

glad he is gone. i am boack

back

next time being as asshole will result in a kick/ban

it might be just somebody with very strong ideas about there being only one software way to do things

I meant to at some point ask them what is the github or gitlab equivalent to mercurial queues

njsg i don;t think there is one

is this the new communication style in SeaMoney devs ? ircbot.comm-central.org:8080/seamonkey/20240227#c225743 WG9s next time being as asshole will result in a kick/ban

no wonder that SM suffers from developer support

have a nice day

njsg: lol i don't think there is one mq is not a thing in git and rebase ain't that

no but wow someone is trying hard to get under people's skin for the log-look-and-see-i-told-you-so-ness i dunno why they think it will improve anything especially with the damned embodyment of it literally not working sitting in here.

there are people who seem to truly think like this and believe that if their project doesn't, I don't know, switch VCS to github/gitlab, or their documentation to a Wiki instance, or their textual support message system to discourse, that "people won't want to participate"

I think that, grabbing this example, while git might be more popular, that does not preclude people knowing and/or using other systems

sure, there will also be people who *will* turn something down because it's not on github or on facebook

now the way this conversation went, it could have gone more the "how do I contribute" route and less the "you should do things in this great way instead of whatever" route

but maybe that's just me

and - for anyone reading the channel log - the mq thing was intended to bring up a topic where, AFAIK, git and hg are visibly different in the feature set, and that could count as one reason not to use git

amd the reaon hq is so slow for mozilla is the completely obnoxious amount of history they try to keep in the repo should not heep so mych and have a seperate history thing

using git would make little difference here

whn you clone mozilla central you get all the change shitory since when changed form cvs to mercurail for the repository

so like mozilla2.0 time

nice to have all that history, but I don;t need it in my local repo

well he was specifically pushing at you, I have noticed several people doing that over time..

njsg: hi

well at least it would seem Guest_ was not Tobin

no dude i have no desire to upset you

i want you to be decidedly NOT upset actually

he started out normal asked how to contirbute and tried to anser and then it all wentvery sideways

There would be no real way to do what is currently done for 2.53 on git alone I understand that now .. doesn't mean I can't turn around and git cinnabar me up a git repo off it ;)

you know, if i care to do that

robin well funny thing is heptapod is a mercural front end for a git backend so ther must be git repos there somewhare

heptapod is just git trpo with hg commands

sounds like cinnabar

but on the forge side?

so you can hg clone and get a mercural repo but the real backend repo on heptapod is really git

as far as i know

I think because frg's instructions work and the fact you are still doing patch by request or patch by bugzilla what repo system you are using doesn't matter much beyond that

If I ever get my own bug tracker working I am basically gonna make pull requests just create a bug and apply the patch file and then close the PR with a link to the bug

attach*

that is what I think I will do personally for my stuffs

I have never been a fan of the pull request concept

i like the concept mozilla had of module owners and you have to get an r+ form someone responsible for the module you are patching, but, unfortunately this has somehow degenerated into you r+ mine and i will r+ yours.

and no one is actually doing any reviews

I also think i don;t like that all the review process is now in phbricator and not within the bug itself

less transparent

an less consistent with open source prinsiples

well no one is gonna add the phabricator functions to bugzilla

lol

the only person actaully doing seamonkey patches that hs a phabricator account is IanN so, just me neing me, I call him mr Phabulous (a Blues Brothers reference)

so when we need a mozilla-central thing changed we have to get Mr Phabulous to do the bug

and less easy to use, possibly more like "pull requests" and other github/gitlab workflow

njsg: it IS basically a pull request just sent from cli

direct to mozilla

code changes, code comments, diffs, etc

phabricator can basically BE the forge

replacing bugzilla

bbl