Re: Code coverage report for groff 1.24.0pre -- thanks, gnulib!
"G. Branden Robinson" <[email protected]>
| Newsgroups | gmane.comp.lib.gnulib.bugs,gmane.comp.printing.groff.general |
|---|---|
| Message-ID | <20260117020635.prwntubjkg6fzfbv@illithid> |
Hi Simon, At 2026-01-17T02:37:58+0100, Simon Josefsson wrote: > "G. Branden Robinson" <[email protected]> writes: > > In groff, we don't use GNUMakefiles--we try to use portable Make, > > and have largely succeeded. > > Out of curiosity, how do you know that? What kind of non-GNU make > implementations do you test and support? I've built with Debian's bmake, which I _think_ is the same one (though likely a different release of) the one used by NetBSD and OpenBSD. ii bmake 20200710-14+deb11u1 amd64 NetBSD make ...but not lately. I should give it another go... Okay, bmake (this version anyway), absurdly doesn't accept "-j10" as option, claiming that the option argument "isn't an integer". Hmm, and I see I've gotten myself into trouble with font/devpdf/symbolsl.{afm,pfb}. A clash between BSD and GNU ideas of the meaning of VPATH, I think. This rings a bell. I'll need to do an in-tree build. However, kludging through it with: cd build bmake bmake ./font/devpdf/symbolsl.afm bmake ./font/devpdf/symbolsl.pfb bmake (The leading "./" is CRUCIAL.) ...the build runs to completion and exits with status zero. I haven't tested an in-tree build in a while, and this prompts me to do so--thanks! If I could just get my "automated upload registration" processed, I'd be tagging and signing groff 1.24.0.rc1 _really_ soon. > I tried some time ago to find some well-maintained make to do > alternative testing with, but I gave up because none worked on any > modern GNU/Linux distribution, or did intentionally not target POSIX > compliance and broke easily. Maybe things have changed since then, so > pointers welcome. If you can live with in-tree builds, they may have. I'm a little irked by that, and by BSD make's refusal to implement pattern rules because either they crawl with GNU cooties, or because clear semantics are not worth typing two more "%"s and a ": "--this asks too much of the descendants of the slackjawed disciples Thompson won for himself during his 1975 sabbatical. I wrote up my grievance at length. https://cgit.git.savannah.gnu.org/cgit/groff.git/tree/doc/doc.am?id=a18efe62807b77caf18da08ab2095bd35577277d#n306 That was almost enough to drive me to dynamite any claim of support for any Make but GNU's...but not quite. When I "git blamed" gnulib's "top/maint.mk" to see how I'd need to update the copyright notice in groff's "Makefile.am", I saw your name next to most of this coverage stuff. Thank you! Regards, Branden
signature.asc
(application/pgp-signature, 833 B)
-----BEGIN PGP SIGNATURE----- iQIzBAABCAAdFiEEh3PWHWjjDgcrENwa0Z6cfXEmbc4FAmlq7qMACgkQ0Z6cfXEm bc6KdxAAisUAfcfmYQRITHsI+QqT0aGCBWTch/CLgpWYP1zeLoPNosao6P5mvDlo PVgA4oX5sAy2hx08cRqSwxkyShWagztIMwBXF4NinVj3b7Uktic+IcPh75zUbaGQ F1uMVQCMlPjL8X00/srWwPHGKOCN2kK9ctF7JcBx2VVoMFH1LtV/pjsTQKBPmLxN JHzeLC0JZFBbiBI0J1q5/2iltmlbn9PywNFwOiSfdxmO+Bd211dtnujqoc0SS4VM XX2ywuUUwXI6ia3mx6OSTs1bFEqA/R7Zz9CHfuOT/uW8dNdua1ySXWZm7JUuiHTd hecBjmnrE52U1x4ZsmqIls2H7XSKJmE1w+SET8f7s9BT61Ep/V1KYAel84RKeGYy qnT/E7grPcAkp6aPMlvVwDrILtKqCwDgiYAm80VjdENvEF5mjHOQWpZW1PX7uPGk E2YzmbP8+kj+V7jTftht3Frx5XclHy6nQsKC+3HKjHoygLCdrwzi/GjmIfEeN+gL 0ZxorysqFONupGpZyjXl17ODvK3FR7VQAA7jrSCmlAnA4GWPMl6P1AY0lK5gpeQi URS8OC48nIO7u2R4g3FnDRdFBRf8KvlYYASYGTC4lVZu2xiIm16bE92UZhT61mtG YTZXC6Arm2SKU23idsxTsFNc62U0fb4cevsJ0r28+yyKMyYe0yc= =pn8H -----END PGP SIGNATURE-----