[bug#60358] [PATCH v7 4/5] gnu: Add gnulib.

[Vivien Kraus]

* gnu/packages/build-tools.scm (gnulib): New variable.
(gnulib-checkout): New function. It returns a package with a specific commit
of gnulib.
---
 gnu/packages/build-tools.scm | 220 +++++++++++++++++++++++++++++++++++
 1 file changed, 220 insertions(+)

diff --git a/gnu/packages/build-tools.scm b/gnu/packages/build-tools.scm
index 6c1350c44f..d2eeb88db6 100644
--- a/gnu/packages/build-tools.scm
+++ b/gnu/packages/build-tools.scm
@@ -32,27 +32,36 @@
 ;;; along with GNU Guix.  If not, see <http://www.gnu.org/licenses/>.
 
 (define-module (gnu packages build-tools)
+  #:use-module (ice-9 optargs)
   #:use-module ((guix licenses) #:prefix license:)
   #:use-module (guix utils)
   #:use-module (guix packages)
   #:use-module (guix gexp)
   #:use-module (guix download)
   #:use-module (guix git-download)
+  #:use-module (guix git)
   #:use-module (guix build-system cmake)
+  #:use-module (guix modules)
   #:use-module (gnu packages)
   #:use-module (gnu packages adns)
+  #:use-module (gnu packages autotools)
   #:use-module (gnu packages base)
   #:use-module (gnu packages bash)
   #:use-module (gnu packages check)
+  #:use-module (gnu packages code)
   #:use-module (gnu packages compression)
   #:use-module (gnu packages cpp)
+  #:use-module (gnu packages cppi)
   #:use-module (gnu packages elf)
+  #:use-module (gnu packages gcc)
   #:use-module (gnu packages linux)
+  #:use-module (gnu packages lisp)
   #:use-module (gnu packages logging)
   #:use-module (gnu packages lua)
   #:use-module (gnu packages ninja)
   #:use-module (gnu packages package-management)
   #:use-module (gnu packages pcre)
+  #:use-module (gnu packages perl)
   #:use-module (gnu packages pkg-config)
   #:use-module (gnu packages pretty-print)
   #:use-module (gnu packages protobuf)
@@ -65,6 +74,7 @@ (define-module (gnu packages build-tools)
   #:use-module (gnu packages rpc)
   #:use-module (gnu packages sqlite)
   #:use-module (gnu packages tls)
+  #:use-module (gnu packages unicode)
   #:use-module (gnu packages version-control)
   #:use-module (guix build-system gnu)
   #:use-module (guix build-system python))
@@ -803,3 +813,213 @@ (define-public genie
 same settings to multiple projects.  It supports generating projects using GNU
 Makefiles, JSON Compilation Database, and experimentally Ninja.")
       (license license:bsd-3))))
+
+(define*-public (gnulib-checkout #:key
+                                 version
+                                 (revision "1")
+                                 commit
+                                 hash)
+  "Return as a package the exact gnulib checkout."
+  (package
+    (name "gnulib")
+    (version (git-version version revision commit))
+    (source
+     (origin
+       (method git-fetch)
+       (uri (git-reference
+             (url "https://git.savannah.gnu.org/git/gnulib.git";)
+             (commit commit)))
+       (file-name (git-file-name name version))
+       (sha256 hash)
+       (snippet
+        (with-imported-modules (source-module-closure '((guix build utils)))
+          #~(begin
+              (use-modules (guix build utils)
+                           (ice-9 ftw)
+                           (ice-9 rdelim))
+              ;; .c, .h and .gperf files whose first line is /* DO NOT EDIT!
+              ;; GENERATED AUTOMATICALLY! */ are generated automatically based
+              ;; on the unicode database. Since we replace the unicode
+              ;; database with our own, we need to regenerate them. So, they
+              ;; are removed from the source. They are sprinkled all over the
+              ;; place unfortunately, so we can’t exclude whole directories.
+              (let ((count-purged
+                     (file-system-fold
+                      ;; enter?
+                      (lambda (name stat result)
+                        #t)
+                      ;; leaf
+                      (lambda (name stat result)
+                        (if (or (string-suffix? ".c" name)
+                                (string-suffix? ".h" name)
+                                (string-suffix? ".gperf" name))
+                            (call-with-input-file name
+                              (lambda (port)
+                                (let ((first-line (read-line port)))
+                                  (if (equal?
+                                       first-line
+                                       "/* DO NOT EDIT! GENERATED 
AUTOMATICALLY! */")
+                                      (begin
+                                        (delete-file name)
+                                        (1+ result))
+                                      result))))
+                            result))
+                      ;; down
+                      (lambda (name stat result)
+                        result)
+                      ;; up
+                      (lambda (name stat result)
+                        result)
+                      ;; skip
+                      (lambda (name stat result)
+                        (error "No directory should be spared"))
+                      ;; error
+                      (lambda (name stat errno result)
+                        (error "A file is inaccessible"))
+                      ;; Initial value
+                      0
+                      (getcwd))))
+                (unless (eqv? count-purged 332)
+                  (format (current-error-port) "There were ~s files purged.\n" 
count-purged)
+                  (error "Please check the number of automatically generated 
files.")))
+              ;; Other files are copied from UCD.
+              (for-each delete-file
+                        '("tests/unigbrk/GraphemeBreakTest.txt"
+                          "tests/uninorm/NormalizationTest.txt"
+                          "tests/uniname/UnicodeData.txt"
+                          "tests/uniname/NameAliases.txt"
+                          ;; FIXME: tests/uniname/HangulSyllableNames.txt
+                          ;; seems like a UCD file but it is not distributed
+                          ;; with UCD.
+                          "tests/uniwbrk/WordBreakTest.txt")))))))
+    (build-system gnu-build-system)
+    (arguments
+     (list
+      #:phases
+      #~(modify-phases %standard-phases
+          (add-after 'unpack 'fix-tests
+            (lambda _
+              (substitute* "Makefile"
+                (("-f maint.mk syntax-check")
+                 "_gl-Makefile=yes -f maint.mk syntax-check"))
+              (invoke "git" "init")
+              (invoke "git" "config" "user.name" "Guix")
+              (invoke "git" "config" "user.email" "guix@localhost")
+              (invoke "git" "add" ".")
+              (invoke "git" "commit" "-m"
+                      "Syntax checks are only run against committed files, so 
commit everything.")))
+          (add-after 'fix-tests 'disable-failing-tests
+            (lambda _
+              (substitute* "cfg.mk"
+                (("local-checks-to-skip =")
+                 "local-checks-to-skip = \\
+  sc_Wundef_boolean \\
+  sc_file_system \\
+  sc_indent \\
+  sc_keep_gnulib_texi_files_mostly_ascii \\
+  sc_prohibit_assert_without_use \\
+  sc_prohibit_close_stream_without_use \\
+  sc_prohibit_defined_have_decl_tests \\
+  sc_prohibit_doubled_word \\
+  sc_prohibit_empty_lines_at_EOF \\
+  sc_prohibit_intprops_without_use \\
+  sc_prohibit_openat_without_use \\
+  sc_prohibit_test_minus_ao \\
+  sc_unportable_grep_q"))
+              (substitute* "Makefile"
+                (("sc_check_sym_list")
+                 "disabled_check_sym_list")
+                (("sc_cpp_indent_check")
+                 "disabled_cpp_indent_check")
+                (("sc_check_copyright")
+                 "disabled_check_copyright")
+                (("sc_prohibit_AC_LIBOBJ_in_m4")
+                 "disabled_prohibit_AC_LIBOBJ_in_m4")
+                (("sc_prefer_ac_check_funcs_once")
+                 "disabled_prefer_ac_check_funcs_once")
+                (("sc_prohibit_leading_TABs")
+                 "disabled_prohibit_leading_TABs"))))
+          (delete 'configure)
+          (add-after 'unpack 'regenerate-unicode
+            (lambda* (#:key inputs #:allow-other-keys)
+              (with-directory-excursion "lib"
+                ;; See the compile-command buffer-local variable in
+                ;; lib/gen-uni-tables.c
+                (invoke #+(file-append gcc "/bin/gcc") "-O" "-Wall" 
"gen-uni-tables.c"
+                        "-Iunictype" "-o" "gen-uni-tables")
+                (apply invoke
+                       "./gen-uni-tables"
+                       `(,@(map (lambda (f)
+                                  (search-input-file inputs f))
+                                '("share/ucd/UnicodeData.txt"
+                                  "share/ucd/PropList.txt"
+                                  "share/ucd/DerivedCoreProperties.txt"
+                                  "share/ucd/emoji/emoji-data.txt"
+                                  "share/ucd/ArabicShaping.txt"
+                                  "share/ucd/Scripts.txt"
+                                  "share/ucd/Blocks.txt"
+                                  "share/ucd/PropList-3.0.1.txt"
+                                  "share/ucd/EastAsianWidth.txt"
+                                  "share/ucd/LineBreak.txt"
+                                  "share/ucd/auxiliary/WordBreakProperty.txt"
+                                  
"share/ucd/auxiliary/GraphemeBreakProperty.txt"
+                                  "share/ucd/CompositionExclusions.txt"
+                                  "share/ucd/SpecialCasing.txt"
+                                  "share/ucd/CaseFolding.txt"))
+                         ;; This is the version of the UCD used, it should be
+                         ;; the same as the native-input.
+                         #$(package-version ucd-next)))
+                (invoke "clisp" "-C" "uniname/gen-uninames.lisp"
+                        (search-input-file inputs "share/ucd/UnicodeData.txt")
+                        "uniname/uninames.h"
+                        (search-input-file inputs "share/ucd/NameAliases.txt"))
+                (copy-file (search-input-file inputs 
"share/ucd/NameAliases.txt")
+                           "../tests/uniname/NameAliases.txt")
+                (copy-file (search-input-file inputs 
"share/ucd/UnicodeData.txt")
+                           "../tests/uniname/UnicodeData.txt")
+                (copy-file (search-input-file inputs 
"share/ucd/NormalizationTest.txt")
+                           "../tests/uninorm/NormalizationTest.txt")
+                (copy-file (search-input-file inputs 
"share/ucd/auxiliary/GraphemeBreakTest.txt")
+                           "../tests/unigbrk/GraphemeBreakTest.txt")
+                (copy-file (search-input-file inputs 
"share/ucd/auxiliary/WordBreakTest.txt")
+                           "../tests/uniwbrk/WordBreakTest.txt")
+                (delete-file "gen-uni-tables"))))
+          (replace 'install
+            (lambda _
+              (install-file "gnulib-tool"
+                            (string-append #$output "/bin"))
+              (delete-file-recursively ".git")
+              (copy-recursively "." (string-append #$output 
"/src/gnulib/")))))))
+    (inputs ;; Shebangs for some auxiliary build files.
+     (list python perl clisp))
+    (native-inputs
+     (list
+      python perl clisp
+      ;; Unicode data:
+      ucd-next ;; If you change it, also change #$(package-version ucd-next)
+               ;; in the regenerate-unicode phase.
+      ucd3.0-update1
+      ;; Programs for the tests:
+      cppi indent git autoconf))
+    (home-page "https://www.gnu.org/software/gnulib/";)
+    (synopsis "Source files to share among distributions")
+    (description
+     "Gnulib is a central location for common infrastructure needed by GNU
+packages.  It provides a wide variety of functionality, e.g., portability
+across many systems, working with Unicode strings, cryptographic computation,
+and much more.  The code is intended to be shared at the level of source
+files, rather than being a standalone library that is distributed, built, and
+installed.  The included @command{gnulib-tool} script helps with using Gnulib
+code in other packages.  Gnulib also includes copies of licensing and
+maintenance-related files, for convenience.")
+    (native-search-paths
+     (list (search-path-specification
+            (variable "GNULIB_SRCDIR")
+            (files (list "src/gnulib")))))
+    (license (list license:lgpl2.0+ license:gpl3+))))
+
+(define-public gnulib
+  (gnulib-checkout
+   #:version "2022-12-28"
+   #:commit "14a7b0ce5462c90ce86d97bf952185ec2500d341"
+   #:hash (base32 "0fz25ccdlxf4xp37rjsl1fslc4g0x12qpvadz4qw0cq18dvx5kbx")))
-- 
2.38.1