On Tue, Jul 30, 2013 at 10:21:08PM +0200, Paul Wise wrote: > On Tue, Jul 30, 2013 at 8:17 PM, chrysn wrote: > > > given that the game only displays its own static strings and usually > > ships the font file, i'd assume upstream will stay with sdl ttf, but i > > asked anyway. > > The main advantage of fontconfig is not having to hard-code font > pathnames at build time nor in git. If you add pango/QuesoGLC then you > get fallback on secondary fonts, which is mainly useful for l10n. zeno (the hyperrogue author) has shown interest in pango sdl, but won't convert on short term. (currently, there's no i18n either). as a quick fix, i've added a fontconfig patch following the example of chromium-bsu as you suggested. i don't understand everything it's doing with its patterns, but checked with the fontconfig to avoid the worst pitfalls. as i assume that fontconfig will do just that kind of fallbacks, i've changed the depends from ttf-bitstream-vera to ttf-bitstream-vera | fonts-dejavu-core. > > my impression was that it's a mixture of copy/pasted and one-shot import > > from somewhere else, possibly an ad-hoc process -- but i've asked for > > clarification here too. much better -- as a hidden feature, the game can be used as a hyperbolic vector editor, which spits out c++ code. this, along with the other hidden features, has been documented in the man page, with an additional note in README.source concerning preferred forms of modification. speaking of preferred forms of modification: i've obtained a more original file the icon was created from; it used to be a screenshot, but modifications were whose sources are lost (as it is common with this kind of files). i added the most original version to the debian package (with a include-binaries override), which is now shipped as .png, and the xpm is generated from it. > > the build process is a single gcc invocation. --parallel wouldn't do > > anything, would it? > > Right now no, point taken and added, but i'll have to check carefully anyways if the upstream build system changes much. > > i've already spotted another zero pointer dereference > > (03-worm-segfault.patch), which might be indicative of a general > > [...] > > Sounds reasonable. upstream has been very responsive and supportive; all the non-debian specific patches so far have been included in a 3.8 version that is just being released (and is already merged into the package). from my point of view, the package is in a releaseable state again (and uploaded to mentors, see urls below). please review again and upload if you don't find remaining issues. thanks chrysn urls: mentors overview: http://mentors.debian.net/package/hyperrogue dsc: dget -x http://mentors.debian.net/debian/pool/main/h/hyperrogue/hyperrogue_3.8+dfsg-1.dsc git: git://anonscm.debian.org/pkg-games/hyperrogue.git -- To use raw power is to make yourself infinitely vulnerable to greater powers. -- Bene Gesserit axiom
Attachment:
signature.asc
Description: Digital signature