[bug#40322] [PATCH v3] gnu: Add Xplanet.

[Eric Bavier <bavier@posteo.net>]

On 31.03.2020 02:57, R Veera Kumar wrote:
> +       (patches
> +        (search-patches
> +         "xplanet-1.3.1-remove-null-comparison.patch"
> +         "xplanet-1.3.1-libdisplay_DisplayOutput.cpp.patch"
> +         "xplanet-1.3.1-libimage_gif.c.patch"
> +         "xplanet-1.3.1-readConfig-fixclang.cpp.patch"

Is this patch necessary if we're building with gcc?

> +         "xplanet-1.3.1-xpUtil-Add2017LeapSecond.cpp.patch"))))
> +    (build-system gnu-build-system)
> +    (native-inputs
> +     `(("pkg-config" ,pkg-config)))
> +    (inputs
> +     `(("libx11" ,libx11)
> +       ("libxscrnsaver" ,libxscrnsaver)
> +       ("libxext" ,libxext)
> +       ("libice" ,libice)
> +       ("freetype" ,freetype)
> +       ("pango" ,pango)
> +       ("giflib" ,giflib)
> +       ("libjpeg", libjpeg)
> +       ("libpng" ,libpng)
> +       ("libtiff" ,libtiff)
> +       ("zlib" ,zlib)))
> +    (arguments
> +     `(#:configure-flags
> +       (list
> +        "--without-pnm"       ;; no proper pnm library in guix

The "netpbm" package provides a pnm library.

> diff --git
> a/gnu/packages/patches/xplanet-1.3.1-remove-null-comparison.patch
> b/gnu/packages/patches/xplanet-1.3.1-remove-null-comparison.patch
> new file mode 100644
> index 0000000000..002701ed0c
> --- /dev/null
> +++ b/gnu/packages/patches/xplanet-1.3.1-remove-null-comparison.patch
> @@ -0,0 +1,161 @@
> +Origin: Gentoo Harri Nieminen 2017-02-28
> +Url: 
> https://gitweb.gentoo.org/repo/gentoo.git/log/x11-misc/xplanet/files/
> ++xplanet-1.3.1-remove-null-comparison.patch
> +
> +Index: src/libannotate/addArcs.cpp
> +===================================================================
> +diff --git a/src/libannotate/addArcs.cpp b/src/libannotate/addArcs.cpp
> +--- a/src/libannotate/addArcs.cpp	(revision 206)
> ++++ b/src/libannotate/addArcs.cpp	(revision 207)
> +@@ -258,7 +258,7 @@
> +         {
> +             ifstream inFile(arcFile.c_str());
> +             char *line = new char[MAX_LINE_LENGTH];
> +-            while (inFile.getline (line, MAX_LINE_LENGTH, '\n') != 
> NULL)
> ++            while (inFile.getline (line, MAX_LINE_LENGTH, '\n'))

I'm worried this might not be correct.  The "getline" function always 
returns it's first parameter, which, I think, can never evaluate to a 
falsy value.  The safer option would be to call ".eof()" on the result, 
and would match the intentions.  See e.g. 
https://notabug.org/bavier/guix-bavier/src/master/bavier/patches/xplanet-cxx11-eof.patch

-- 
`~Eric