[wpe-webkit] Unable to build stable (semi-colon gotcha?)

[Andrea Giammarchi]

Uh ... I see (u), that explains better what happened, thanks for
investigating.

Since we are here, and if you don't mind, I've built cairo-glesv2 and
wpewebkit, and I'm using wpebackend-fdo and cog, where cog is built
directly from git
https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=cog-wpe-gl

I can bootstrap WPE on Weston without issues, but all I see is dozen
horizontal bars that actually would show otherwise what's in the page, but
the "barring" makes it impossible to use ... this, at least, in Raspberry
Pi 4 and its armv7l stack.

y any chance you have an idea what could've gone wrong? All builds looked
just fine, and installed fine too.

Thanks.


On Mon, Jan 13, 2020 at 5:13 PM Andres Gomez <tanty at igalia.com> wrote:

> Hi,
>
> it seems icu 65.1 "introduced" this regression:
> https://bugs.webkit.org/show_bug.cgi?id=202600
>
> FWIW, the previous code was like it since 2008. Hence, this is not
> something introduced in a "stable" release but rather a conflict
> between versions.
>
> FWIW, it was already added into the cherry-pick list for the next
> stable version, 2.26.3:
> https://trac.webkit.org/wiki/WebKitGTK/2.26.x
>
> Br.
>
>
> On Mon, 2020-01-13 at 17:28 +0200, Andres Gomez wrote:
> > Hi,
> >
> > quickly answering:
> >
> > U16_NEXT is a preprocessor macro. In other words, it is replaced by the
> > actual code in that stage of the compilation.
> >
> > If you check the definition of this macro at:
> >
> https://git.webkit.org/?p=WebKit.git;a=blob;f=Source/WTF/icu/unicode/utf16.h;h=bdd88a8b9c4f00ea0df272ac02f6967ef61089b6;hb=HEAD#l264
> >
> > You will realize that it is replaced by a block of code (starting and
> > closed with brackets).
> >
> > In other words, although it would be good to have the semicolon there,
> > it is not really needed and it should not lead to a build failure,
> > AFAIS.
> >
> > So, are you sure you don't have a different problem building the code?
> >
> > Br.
> >
> > On Mon, 2020-01-13 at 14:26 +0100, Andrea Giammarchi wrote:
> > > Happy new year everyone!
> > >
> > > Quick question: while I see that current code has semi colons at the
> right place, it's not the first file I need to mokey patch from the stable
> 2.26.2 archive.
> > >
> > > As example, the Document.cpp file has 3 lines where U16_NEXT... misses
> the semi-colon at the end, resulting in build failures.
> > >
> > > My question is:
> > > how is it possible that a stable release doesn't build out of the box?
> > > do I understand correctly source files from WebKit or WPE fork are not
> linted?
> > > Specially the latter point, would be quite surprising to me ... as I'm
> automating ways to build WPEWebKit for various architectures, and dealing
> with files that cannot be used and break the build down the process is not
> really expected/convenient.
> > >
> > > Thanks for any sort of clarification.
> > >
> > > Best Regards
> > >  _______________________________________________
> > > Do not post admin requests to the list. They will be ignored.
> > > webkit-wpe mailing list      (webkit-wpe at lists.webkit.org)
> > > Help/Unsubscribe/Update your Subscription:
> > > https://lists.webkit.org/mailman//listinfo/webkit-wpe
> >
> >  _______________________________________________
> > Do not post admin requests to the list. They will be ignored.
> > webkit-wpe mailing list      (webkit-wpe at lists.webkit.org)
> > Help/Unsubscribe/Update your Subscription:
> > https://lists.webkit.org/mailman//listinfo/webkit-wpe
> --
> Br,
>
> Tanty
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-wpe/attachments/20200113/b0951d75/attachment-0001.htm>