[bug#28004] Chromium

[ng0]

ng0 transcribed 5.5K bytes:
> ng0 transcribed 3.7K bytes:
> > ng0 transcribed 3.4K bytes:
> > > Marius Bakke transcribed 2.4K bytes:
> > > > ng0 <address@hidden> writes:
> > > > 
> > > > > Hi Marius,
> > > > >
> > > > > Marius Bakke transcribed 43K bytes:
> > > > >> Hello Guix!
> > > > >> 
> > > > >> Attached is a patch for Chromium, a popular web browser.
> > > > >
> > > > > Nice! I've been using this from your branch for a while now,
> > > > > works just fine :)
> > > > > Is this not affected by the chromium discussion which happened
> > > > > a while back? Can we include this? I'm all for this, because I
> > > > > mainly use it for websites where firefox/icecat doesn't work so
> > > > > well, and building it locally takes a very long time.
> > > > 
> > > > I believe this is within the Free System Distribution Guidelines. DRM
> > > > ("Widevine") is disabled at build time, and the Web Store is
> > > > non-functional without the end user explicitly enabling it.
> > > > 
> > > > There are some grey areas though. The browser may interact with certain
> > > > non-free APIs (apart from regular browser duties) such as translation or
> > > > prediction services. These features are optional, but some are enabled
> > > > by default, and difficult to maintain patches for (I've tried).
> > > > 
> > > > However, I have verified that it does not send any unsolicited requests
> > > > with the current command-line options, apart from the very first launch
> > > > which spawns a login prompt (help wanted!). Without either of those
> > > > flags the browser "calls home" every time it starts.
> > > > 
> > > > >> Note that I cannot guarantee timely delivery of security updates. 
> > > > >> Major
> > > > >> version upgrades are hugely painful, and almost always contain many
> > > > >> high-severity fixes. Should we mention that in the description?
> > > > >> 
> > > > >> Happy for any feedback.
> > > > >> 
> > > > >
> > > > > Shouldn't you mention defines in addition to the define-public aswell,
> > > > > or don't we do that?
> > > > 
> > > > Not for new files (modules), typically. I don't think Magit can fill out
> > > > those variable names (by pressing C on the hunks) either ;-) But it
> > > > should probably go in web-browsers.scm anyway.
> > > 
> > > Unless someone else is already building this, I'm giving it a spin.
> > > 
> > > I guess you changed some things since the version of yours I have in
> > > here: 
> > > https://gitlab.com/ng0_guix/packages/blob/master/ng0/packages/chromium.scm
> > > so I have to rebuild it.
> > > It might take a while because I'm offloading to something much slower
> > > but which doesn't care about heat as much as a this one ;)
> > 
> > Patch itself LGTM, I'm now waiting on the build to finish in the
> > next couple of hours.
> 
> x86_64 architecture, builds fails at this point:
> 
> [6247/27388] STAMP obj/mojo/common/common.stamp
> [6248/27388] ACTION 
> //net/http:generate_transport_security_state(//build/toolchain/linux:x64)
> FAILED: gen/net/http/transport_security_state_static.h
> python ../../build/gn_run_binary.py transport_security_state_generator 
> ../../net/http/transport_security_state_static.json 
> ../../net/http/transport_security_state_static.pins 
> ../../net/http/transport_security_state_static.template 
> gen/net/http/transport_security_state_static.h
> ./transport_security_state_generator: error while loading shared libraries: 
> libglib-2.0.so.0: cannot open shared object file: No such file or directory
> transport_security_state_generator failed with exit code 127
> [6249/27388] AR obj/sandbox/linux/libsandbox_services.a
> ninja: build stopped: subcommand failed.
> phase `build' failed after 1777.2 seconds
> builder for 
> `/gnu/store/2afpy542vywbmk093dd1kzlfx74s2460-chromium-60.0.3112.90.drv' 
> failed with exit code 1
> @ build-failed 
> /gnu/store/2afpy542vywbmk093dd1kzlfx74s2460-chromium-60.0.3112.90.drv - 1 
> builder for 
> `/gnu/store/2afpy542vywbmk093dd1kzlfx74s2460-chromium-60.0.3112.90.drv' 
> failed with exit code 1
> derivation 
> '/gnu/store/2afpy542vywbmk093dd1kzlfx74s2460-chromium-60.0.3112.90.drv' 
> offloaded to '192.168.1.179' failed: build of 
> `/gnu/store/2afpy542vywbmk093dd1kzlfx74s2460-chromium-60.0.3112.90.drv' failed
> @ build-failed 
> /gnu/store/2afpy542vywbmk093dd1kzlfx74s2460-chromium-60.0.3112.90.drv - 1 
> builder for 
> `/gnu/store/2afpy542vywbmk093dd1kzlfx74s2460-chromium-60.0.3112.90.drv' 
> failed with exit code 100
> guix build: error: build failed: build of 
> `/gnu/store/2afpy542vywbmk093dd1kzlfx74s2460-chromium-60.0.3112.90.drv' failed
> 
> Have you experienced this before?

As efraim pointed out I missed the part where you wrote that
it is for core-updates. I just assumed it worked like it is
on master because what I had locally (chromium 58) works on
master).

Someone else must test it then.
-- 
ng0
GnuPG: A88C8ADD129828D7EAC02E52E22F9BBFEE348588
GnuPG: https://n0is.noblogs.org/my-keys
https://www.infotropique.org https://krosos.org

signature.asc
Description: PGP signature