Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fontforge: 20190317 -> 20190413 #60537

Merged
merged 1 commit into from May 1, 2019
Merged

Conversation

r-ryantm
Copy link
Contributor

Semi-automatic update generated by https://github.com/ryantm/nixpkgs-update tools. This update was made based on information from https://repology.org/metapackage/fontforge/versions.

meta.description for fontforge is: '"A font editor"'.

Release on GitHub

Compare changes on GitHub

Checks done (click to expand)
Rebuild report (if merged into master) (click to expand)

113 total rebuild path(s)

44 package rebuild(s)

44 x86_64-linux rebuild(s)
36 i686-linux rebuild(s)
16 x86_64-darwin rebuild(s)
17 aarch64-linux rebuild(s)

First fifty rebuilds by attrpath
asciidoc-full
asciidoc-full-with-plugins
btrbk
creep
denemo
diffoscope
disorderfs
emojione
fontforge
fontforge-fonttools
fontforge-gtk
frescobaldi
liberation-sans-narrow
libreoffice-fresh
libreoffice-fresh-unwrapped
lilypond
lilypond-unstable
lilypond-with-fonts
pdf2htmlEX
pipelight
playonlinux
python27Packages.fontforge
python37Packages.fontforge
ricty
scfbuild
solfege
tlwg
twemoji-color-font
udiskie
wine
wine-staging
winePackages.base
winePackages.full
winePackages.minimal
winePackages.stable
winePackages.staging
winePackages.unstable
wineWowPackages.base
wineWowPackages.full
wineWowPackages.minimal
wineWowPackages.stable
wineWowPackages.staging
wineWowPackages.unstable
winetricks

Instructions to test this update (click to expand)

Either download from Cachix:

nix-store -r /nix/store/07pznd600wfxq2zbmk8mi2fxnmalviml-fontforge-20190413 \
  --option binary-caches 'https://cache.nixos.org/ https://r-ryantm.cachix.org/' \
  --option trusted-public-keys '
  r-ryantm.cachix.org-1:gkUbLkouDAyvBdpBX0JOdIiD2/DP1ldF3Z3Y6Gqcc4c=
  cache.nixos.org-1:6NCHdD59X431o0gWypbMrAURkbJ16ZPMQFGspcDShjY=
  '

(r-ryantm's Cachix cache is only trusted for this store-path realization.)

Or, build yourself:

nix-build -A fontforge https://github.com/r-ryantm/nixpkgs/archive/e8da6b08adfdfedf5f4beae067f6cfc72f947f28.tar.gz

After you've downloaded or built it, look at the files and if there are any, run the binaries:

ls -la /nix/store/07pznd600wfxq2zbmk8mi2fxnmalviml-fontforge-20190413
ls -la /nix/store/07pznd600wfxq2zbmk8mi2fxnmalviml-fontforge-20190413/bin

Semi-automatic update generated by
https://github.com/ryantm/nixpkgs-update tools. This update was made
based on information from
https://repology.org/metapackage/fontforge/versions
@dtzWill
Copy link
Member

dtzWill commented May 10, 2019

Do we have any sense as to whether the differences are:

  1. significant/observable in terms of actually different glyphs/etc, vs "noise"? (does diffoscope support fonts? :D)
  2. reproducible for a given fontforge version?
  3. assuming differences cause the fonts to behave/render differently, are these changes ... intended? desirable?

Thankfully I think (2) is mostly resolved by previous efforts, and probably would have caused enough breakage for occasional non-cache-using builders that we would've noticed by now.

Answering (1) is probably not too hard, but making sense of it enough for (3)... not sure how to tackle that.

I'm not entirely sure about our policy to "build from source" for packages of fonts or themes (twemoji-color) -- are we wasting resources generating and storing data that's not what upstream intended nor what our users expect?
Without going too much into it, I'm not against it at all but these sorts of things do seem different than programs for a few reasons... just wondering if this is the right approach, how we might make decisions about these changes, and so on.

I suppose most immediately we should try to at least informally identify which fonts are possibly changed, and hopefully other "generated" files that might similarly be a bit overstating their fixed-output-ness ;). The most common example I can think of offhand would be themes, which commonly use imagemagick to render a variety of sizes or whatnot. I don't think we ever mark them as FO, but I'd have to check.

In summary: aren't we lying re;marking these FO, and can we do better than that? This sort of thing blocks font upgrade all the time, for example this came up recently and basically stalled our update of the font: #57604 .

(of course in the immediate future we can just fix the hashes.. if they aren't already....)

@7c6f434c
Copy link
Member

7c6f434c commented May 10, 2019 via email

@r-ryantm r-ryantm deleted the auto-update/fontforge branch May 11, 2019 02:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants