Sergey Fedorov

Results 877 comments of Sergey Fedorov

I believe this can be merged.

So `colm` builds fine now on buildbots (without html docs), while `ragel` fails strangely on: ``` /usr/bin/clang++ -DHAVE_CONFIG_H -I. -I../aapl -I../colm/include -I/opt/local/include/aapl -I/opt/local/include -I/opt/local/include -isysroot/Library/Developer/CommandLineTools/SDKs/MacOSX11.sdk -pipe -Os -stdlib=libc++ -isysroot/Library/Developer/CommandLineTools/SDKs/MacOSX11.sdk -arch...

> are we doomed to have @barracuda156 add this to 10,000 ports to support rosetta? This is added by the muniversal PG I believe, but that is not being used...

> what is wrong with building the docs exactly? ``` Making install in colm make[2]: Entering directory `/opt/local/var/macports/build/_opt_PPCRosettaPorts_lang_colm/colm/work/colm-0.14.7/doc/colm' b html5 -a icons -a iconsdir=/usr/share/asciidoc/icons -a max-width=800px -a source-highlighter=pygments -o 0_00_welcome.html...

> I can't change base for this, but if there is suitable motivation, it might be done. @kencu This is by the way genuinely Rosetta-specific problem (on 10.6 PPC the...

> > @kencu This is by the way genuinely Rosetta-specific problem (on 10.6 PPC the triple is set correctly). > > It's not really even wrong. The machine you're building...

@kencu Turns out that the error `No rule to make target /opt/local/lib/libfsm.la, needed by ragel` has been reported to upstream ages ago, but never fixed and issue closed (link above)....

Update from 10.5.8. `colm` builds fine for `ppc`: ``` 36-249:~ svacchanda$ port -v installed colm The following ports are currently installed: colm @0.14.7_0 (active) requested_variants='-universal' platform='darwin 9' archs='ppc' date='2022-10-08T02:03:56+0800' ```...

@essandess `ragel` 7.0.4 builds fine on 10.5.8 too: ``` 36-249:~ svacchanda$ port -v installed ragel The following ports are currently installed: ragel @7.0.4_0 (active) requested_variants='-universal' platform='darwin 9' archs='ppc' date='2022-10-08T02:12:23+0800' ```...

Updated to latest-commit versions from Github, rebased upon master. Let’s see if this works. UPD. Well, no, same story :(