ignore the patch anyway. it's far too invasive.

also, it *did* compile but it was messed up when i ran it. it compiled normal firefox!

i'm going to re-do it in a much more conservative way in a future attempt.

what i will do in a future re-attempt is:

* remove mozconfig from librewolf, and only use CONFIGURE_ARGS. keep it the same as firefox, but with librewolf hardening added

* possibly add hardening to the firefox port (in CONFIGURE_ARGS). based on the removed mozconfig from librewolf

* yes, move more things into www/mozilla (no www/mozilla-browser). perhaps conditional items in www/mozilla, e.g. check whether it is a browser type, and act there, rather than having www/mozilla-browser


my main concern is not simply to make librewolf maintenance easier, but to generally simplify all mozilla maintenance in openbsd. assimilating tor-browser into this newly generalised framework would also be nice, but i'll look into that much later and only with the blessing of the tor-browser port maintainer.

the other concern is this:

with a more generalised porting framework specifically for mozilla browsers, we could add a lot more mozilla-based browsers to openbsd quite easily. there are already several spins of librewolf for example. e.g. librewolf-hellfire



Am 02.09.26 um 08:56 schrieb Landry Breuil:
Le Wed, Sep 02, 2026 at 08:12:53AM +0100, Leah Rowe a écrit :
once i'm done unifying firefox/firefox-esr/librewolf, i'm probably going to
target tor-browser as well.

OpenBSD has like five different versions of the same port. It seems to me
that all will be more maintainable overall if they all use the same setup,
with each individual port only altering what it needs. It's the same concept
behind www/mozilla, but for browsers, hence my addition (in the WIP patch)
of www/mozilla-browser
if you want to factorize things, move them to www/mozilla. that module
evolved a lot over time and maybe needs a refactoring, i agree things
would be more maintainable overall, but i wont have the time/resources
to dive into this before months (eg at a hackathon where you can spend
several consecutive hours testing all aspects), so dont hope any
review/feedback from me on such a refactoring, unless you want to take
maintainership of all the consumer ports.

my current workflow to maintain
firefox/thunderbird/firefox-esr/(seamonkey) works for me and i
understand it, and i'd like to keep it this way :)

--
Company director, Minifree Ltd
Registered in England, No. 9361826 | VAT No. GB202190462
Registered Office: 19 Hilton Road, Canvey Island, Essex SS8 9QA, UK

Attachment: OpenPGP_0x5C654067D383B1FF.asc
Description: OpenPGP public key

Attachment: OpenPGP_signature.asc
Description: OpenPGP digital signature

Reply via email to