The first problem that was introduced in a276d5160c
was a linking error:
ld: cannot find -licui18n
ld: cannot find -licuuc
ld: cannot find -licudata
So I added icu to the buildInputs.
The second problem was that the interpreter wasn't patched in
share/filters, apparently this is only needed when building with
autotools:
make[3]: Entering directory '/build/inkscape-0.92.3/share/filters'
./i18n.py ./filters.svg > ./filters.svg.h
./i18n.py: /usr/bin/env: bad interpreter: No such file or directory
A similar error also occurs for share/palettes, share/patterns,
share/symbols and share/templates, so I added patching the interpreter
there as well.
Switching to autotools in Inkscape is a very bad idea, because upstream
currently still has their own autotools files in the 0.92.x tree but
master already has them removed, see this commit:
https://gitlab.com/inkscape/inkscape/commit/e471a664f923f517b68071f2e33fbb6ce070f8b7
However for the sake of trying to not break Inkscape on Darwin again,
I tried to keep the fixes minimal and not went back to CMake.
I did however mark the stuff that's unneeded for CMake, so that we can
avoid forgetting to remove that crap once we get back to CMake.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @matthewbauer
On GNU/Linux the build references these files, so let's fetch them from
the Chromium repository. I haven't checked whether they are heavily
patched or whether we can use the version from LLVM, but when looking at
the changes, they do seem to divert a bit from upstream LLVM.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @matthewbauer, @stesie
The tarball from upstream seems to be generated on the fly, so the
output is not deterministic and using fetchzip makes this more reliable
as we have a recursively hashed output path without any of the
non-determinisms in tarballs.
Unfortunately, the build still fails on NixOS systems, because we need a
few more stuff in the build tree.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @matthewbauer, @stesie
Will Dietz