GNOME.org
 

[gucharmap] unicode: Remove generated sources

commit 8abd62c462b9a45c5ed182b35fbb27e3242cb1b0
Author: Christian Persch <chpe gnome org>
Date:   Thu Mar 3 19:01:01 2016 +0100

    unicode: Remove generated sources
    
    Instead of putting the generated sources in git and the tarballs,
    generate them at build time from data either provided in a directory
    with --with-unicode-data=DIRECTORY or downloaded (and checksum-
    checked) from unicode.org. The only generated file to remain in
    git is unicode-i18n.h dummy file which contains the translatable
    strings, for extraction by gettext.

 configure.ac                                |   50 +-
 gucharmap/Makefile.am                       |   74 +-
 gucharmap/download-unicode-files.sh         |   23 -
 gucharmap/gen-guch-unicode-tables.pl        |   36 +-
 gucharmap/gucharmap-script-codepoint-list.c |    2 +
 gucharmap/gucharmap-unicode-info.c          |    2 +
 gucharmap/unicode-blocks.h                  |  551 -
 gucharmap/unicode-categories.h              | 2978 -
 gucharmap/unicode-names.h                   |58398 ----------
 gucharmap/unicode-nameslist.h               |34642 ------
 gucharmap/unicode-scripts.h                 | 2153 -
 gucharmap/unicode-unihan.h                  |166062 ---------------------------
 gucharmap/unicode-versions.h                | 1396 -
 gucharmap/unicode.sha512sums                |    6 +
 14 files changed, 120 insertions(+), 266253 deletions(-)
---
diff --git a/configure.ac b/configure.ac
index e0980c9..321dc4f 100644
--- a/configure.ac
+++ b/configure.ac
@@ -3,7 +3,7 @@
 #
 # This program is free software; you can redistribute it and/or modify it
 # under the terms of the GNU General Public License as published by the
-# Free Software Foundation; either version 2 of the License, or (at your
+# Free Software Foundation; either version 3 of the License, or (at your
 # option) any later version.
 #
 # This program is distributed in the hope that it will be useful, but
@@ -148,6 +148,7 @@ fi
 
 # Some utilities
 AC_PROG_LN_S
+AC_PROG_SED
 
 AC_ARG_VAR([GLIB_COMPILE_RESOURCES],[the glib-compile-resources programme])
 AC_PATH_PROG([GLIB_COMPILE_RESOURCES],[glib-compile-resources],[false])
@@ -155,28 +156,55 @@ if test "$GLIB_COMPILE_RESOURCES" = "false"; then
     AC_MSG_ERROR([glib-compile-resources not found])
 fi
 
-# AC_PATH_PROG([GTK_BUILDER_CONVERT],[gtk-builder-convert],[false])
-# if test "$GTK_BUILDER_CONVERT" = "false"; then
-#   AC_MSG_ERROR([gtk-builder-convert not found])
-# fi
-
 AC_ARG_VAR([XMLLINT],[the xmllint programme])
 AC_PATH_PROG([XMLLINT],[xmllint],[false])
 if test "$XMLLINT" = "false"; then
   AC_MSG_ERROR([xmllint not found])
 fi
 
-# AC_PATH_PROG([HEXDUMP],[hexdump],[false])
-# if test "$HEXDUMP" = "false"; then
-#   AC_MSG_ERROR([hexdump not found])
-# fi
-
 AC_ARG_VAR([DESKTOP_FILE_VALIDATE],[the desktop-file-validate programme])
 AC_PATH_PROG([DESKTOP_FILE_VALIDATE],[desktop-file-validate],[])
 if test -z "$DESKTOP_FILE_VALIDATE"; then
   AC_MSG_ERROR([desktop-file-validate not found])
 fi
 
+# Unicode data
+
+AC_MSG_CHECKING([for Unicode data])
+AC_ARG_WITH([unicode-data],
+        [AS_HELP_STRING([--with-unicode-data=none],[Specify absolute directory where the Unicode data files 
reside. If 'none', downloads the files to the build directory.])],
+        [unicodedatadir="$(readlink -f $withval)"
+         with_provided_unicode_data="yes (from $unicodedatadir)"],
+        [unicodedatadir='${builddir}'
+         with_provided_unicode_data=no])
+AC_MSG_RESULT([$with_provided_unicode_data])
+
+AM_CONDITIONAL([WITH_UNICODE_DATA],[test "$with_provided_unicode_data" != "no"])
+AC_SUBST([unicodedatadir])
+
+# If we don't have a directory where we can find Unicode data in, we need to download it.
+# For that, we need some utilities
+
+# Note! Not using UNZIP as variable name here since unzip itself
+# uses the UNZIP env var and interprets it as its options.
+AC_ARG_VAR([PROG_UNZIP],[the unzip programme])
+AC_PATH_PROG([PROG_UNZIP],[unzip],[false])
+if test "$PROG_UNZIP" = "false"; then
+    AC_MSG_ERROR([unzip not found])
+fi
+
+AC_ARG_VAR([WGET],[the wget programme])
+AC_PATH_PROG([WGET],[wget],[false])
+if test "$with_provided_unicode_data" = "no" -a "$WGET" = "false"; then
+    AC_MSG_ERROR([wget not found])
+fi
+
+AC_ARG_VAR([SHA512SUM],[the sha512sum programme])
+AC_PATH_PROG([SHA512SUM],[sha512sum],[false])
+if test "$with_provided_unicode_data" = "no" -a "$WGET" = "false"; then
+  AC_MSG_ERROR([sha512sum not found])
+fi
+
 # ****
 # i18n
 # ****
diff --git a/gucharmap/Makefile.am b/gucharmap/Makefile.am
index ac2282d..6d5758c 100644
--- a/gucharmap/Makefile.am
+++ b/gucharmap/Makefile.am
@@ -1,9 +1,9 @@
 ## Copyright © 2003  Noah Levitt
-## Copyright © 2008  Christian Persch
+## Copyright © 2008, 2016  Christian Persch
 ##
 ## This program is free software; you can redistribute it and/or modify it
 ## under the terms of the GNU General Public License as published by the
-## Free Software Foundation; either version 2 of the License, or (at your
+## Free Software Foundation; either version 3 of the License, or (at your
 ## option) any later version.
 ##
 ## This program is distributed in the hope that it will be useful, but
@@ -27,6 +27,7 @@ BUILT_SOURCES = \
        gucharmap-resources.h \
        gucharmap-type-builtins.c \
        gucharmap-type-builtins.h \
+       $(unicode_sources) \
        $(NULL)
 
 libgucharmap GUCHARMAP_LIBRARY_SUFFIX_U@_la_SOURCES = \
@@ -56,13 +57,6 @@ libgucharmap GUCHARMAP_LIBRARY_SUFFIX_U@_la_SOURCES = \
        gucharmap-script-codepoint-list.h \
        gucharmap-unicode-info.c \
        gucharmap-unicode-info.h \
-       unicode-blocks.h \
-       unicode-categories.h \
-       unicode-names.h \
-       unicode-nameslist.h \
-       unicode-scripts.h \
-       unicode-unihan.h \
-       unicode-versions.h \
        $(NULL)
 
 nodist_libgucharmap GUCHARMAP_LIBRARY_SUFFIX_U@_la_SOURCES = $(BUILT_SOURCES)
@@ -206,12 +200,54 @@ gucharmap-type-builtins.c: gucharmap-type-builtins.c.template $(libgucharmapincl
        && (cmp -s xgen-gtbc gucharmap-type-builtins.c || cp xgen-gtbc gucharmap-type-builtins.c ) \
        && rm -f xgen-gtbc
 
-# printoptions.h: print.glade
-#      $(AM_V_GEN) $(GTK_BUILDER_CONVERT) --root custom_tab_container $< - | \
-#              $(XMLLINT) --noblanks - | \
-#              $(HEXDUMP) -v -e '"" 1/1 "0x%02x,""\n"' > xgen-$(@F) \
-#      && (cmp -s xgen-$(@F) $@ || cp -f xgen-$(@F) $@) \
-#      && rm -f xgen-$(@F)
+# Unicode data
+
+WGETFLAGS = --timestamping --timeout=300 --tries=8 --no-directories --directory-prefix="$(builddir)"
+
+unicode_version = 8.0.0
+
+unicode_files = \
+       Blocks.txt \
+       DerivedAge.txt \
+       NamesList.txt \
+       Scripts.txt \
+       UnicodeData.txt \
+       Unihan.zip \
+       $(NULL)
+
+unicode_sources = \
+       unicode-blocks.h \
+       unicode-categories.h \
+       unicode-names.h \
+       unicode-nameslist.h \
+       unicode-scripts.h \
+       unicode-unihan.h \
+       unicode-versions.h \
+       $(NULL)
+
+stamp-download-unicode-data:
+if WITH_UNICODE_DATA
+       @echo timestamp > $(@F)
+else
+       $(MAKE) download-unicode-data && echo timestamp > $(@F)
+endif
+
+download-unicode-data:
+       for f in $(unicode_files); do \
+               [ -e $(builddir)/$${f} ] || $(WGET) $(WGETFLAGS) 
"http://www.unicode.org/Public/$(unicode_version)/ucd/$${f}"; \
+       done && \
+       $(SHA512SUM) --check $(srcdir)/unicode.sha512sums
+
+unicode.sha512sums: stamp-download-unicode-data
+       $(AM_V_GEN)$(SHA512SUM) $(addprefix $(unicodedatadir)/,$(unicode_files)) | $(SED) 
's,$(unicodedatadir)/,,g' > $(srcdir)/$(@F)
+
+stamp-unicode-data:
+       PROG_UNZIP=$(PROG_UNZIP) $(srcdir)/gen-guch-unicode-tables.pl $(unicode_version) $(unicodedatadir) && 
\
+       echo timestamp > $(@F)
+
+$(unicode_sources): stamp-unicode-data
+
+# Dist & cleanup
 
 EXTRA_DIST = \
        gucharmap-marshal.list \
@@ -221,13 +257,21 @@ EXTRA_DIST = \
        download-unicode-files.sh \
        gen-guch-unicode-tables.pl \
         gucharmap.gresource.xml.in \
+       unicode.sha512sums \
+       unicode-strings.h \
        $(NULL)
 
 CLEANFILES = \
        stamp-gucharmap-type-builtins.h \
+       stamp-download-unicode-data \
+       stamp-unicode-data \
        $(BUILT_SOURCES) \
        $(NULL)
 
+DISTCLEANFILES = \
+       $(unicode_files) \
+       $(NULL)
+
 # introspection
 
 if HAVE_INTROSPECTION
diff --git a/gucharmap/gen-guch-unicode-tables.pl b/gucharmap/gen-guch-unicode-tables.pl
index 69d856c..f2b5abb 100755
--- a/gucharmap/gen-guch-unicode-tables.pl
+++ b/gucharmap/gen-guch-unicode-tables.pl
@@ -17,17 +17,9 @@
 # NOTE! Some code copied from glib/glib/gen-unicode-tables.pl; keep in sync!
 
 use strict;
-use vars ('$UNZIP');
 
-# if these things aren't in your path you can put full paths to them here
-$UNZIP = 'unzip';
-
-sub process_unicode_data_txt ($);
-sub process_unihan_zip ($);
-sub process_nameslist_txt ($);
-sub process_blocks_txt ($);
-sub process_scripts_txt ($);
-sub process_versions_txt ($);
+use Env qw($PROG_UNZIP);
+$PROG_UNZIP = "unzip" unless (defined $PROG_UNZIP);
 
 $| = 1;  # flush stdout buffer
 
@@ -94,7 +86,7 @@ exit;
 
 #------------------------#
 
-sub process_unicode_data_txt ($)
+sub process_unicode_data_txt
 {
     my ($unicodedata_txt) = @_;
 
@@ -114,7 +106,6 @@ sub process_unicode_data_txt ($)
     print $out "#define UNICODE_NAMES_H\n\n";
 
     print $out "#include <glib.h>\n\n";
-    print $out "#include <glib/gi18n-lib.h>\n\n";
 
     my @unicode_pairs;
     my %names;
@@ -304,14 +295,14 @@ EOT
 #------------------------#
 
 # XXX should do kFrequency too
-sub process_unihan_zip ($)
+sub process_unihan_zip
 {
     my ($unihan_zip) = @_;
 
-    open (my $unihan, "$UNZIP -c $unihan_zip |") or die;
-    open (my $out, "> unicode-unihan.h") or die;
+    print "processing $unihan_zip.";
 
-    print "processing $unihan_zip";
+    open (my $unihan, "$PROG_UNZIP -c '$unihan_zip' |") or die;
+    open (my $out, "> unicode-unihan.h") or die;
 
     print $out "/* unicode-unihan.h */\n";
     print $out "/* THIS IS A GENERATED FILE. CHANGES WILL BE OVERWRITTEN. */\n";
@@ -530,7 +521,7 @@ sub print_names_list
     print $out "};\n\n";
 }
 
-sub process_nameslist_txt ($)
+sub process_nameslist_txt
 {
     my ($nameslist_txt) = @_;
 
@@ -736,7 +727,7 @@ sub read_blocks_txt
     close ($blocks_file);
 }
 
-sub process_blocks_txt ($)
+sub process_blocks_txt
 {
     my ($blocks_txt) = @_;
 
@@ -753,7 +744,6 @@ sub process_blocks_txt ($)
     print $out "#define UNICODE_BLOCKS_H\n\n";
 
     print $out "#include <glib.h>\n";
-    print $out "#include <glib/gi18n-lib.h>\n\n";
 
     my @blocks;
     read_blocks_txt ($blocks_txt, \ blocks);
@@ -843,7 +833,7 @@ sub read_scripts_txt
     $scripts->{"Common"} = 1; 
 }
 
-sub process_scripts_txt ($)
+sub process_scripts_txt
 {
     my ($scripts_txt) = @_;
 
@@ -865,7 +855,6 @@ sub process_scripts_txt ($)
     print $out "#define UNICODE_SCRIPTS_H\n\n";
 
     print $out "#include <glib.h>\n";
-    print $out "#include <glib/gi18n-lib.h>\n\n";
 
     print $out "typedef struct _UnicodeScript UnicodeScript;\n\n";
 
@@ -915,7 +904,7 @@ sub process_scripts_txt ($)
 
 #------------------------#
 
-sub  process_translatable_strings
+sub process_translatable_strings
 {
     my ($blocks_txt, $scripts_txt) = @_;
 
@@ -956,7 +945,7 @@ sub  process_translatable_strings
 
 #------------------------#
 
-sub process_versions_txt ($)
+sub process_versions_txt
 {
     my ($versions_txt) = @_;
 
@@ -1010,7 +999,6 @@ sub process_versions_txt ($)
     print $out "#define UNICODE_VERSIONS_H\n\n";
 
     print $out "#include <glib.h>\n";
-    print $out "#include <glib/gi18n-lib.h>\n\n";
 
     print $out "typedef struct {\n";
     print $out "  gunichar start;\n";
diff --git a/gucharmap/gucharmap-script-codepoint-list.c b/gucharmap/gucharmap-script-codepoint-list.c
index 3197b2f..4c09c0b 100644
--- a/gucharmap/gucharmap-script-codepoint-list.c
+++ b/gucharmap/gucharmap-script-codepoint-list.c
@@ -20,6 +20,8 @@
 #include <glib.h>
 #include <string.h>
 
+#include <glib/gi18n-lib.h>
+
 #include "gucharmap.h"
 #include "gucharmap-private.h"
 
diff --git a/gucharmap/gucharmap-unicode-info.c b/gucharmap/gucharmap-unicode-info.c
index 80dd6b4..c6eafa5 100644
--- a/gucharmap/gucharmap-unicode-info.c
+++ b/gucharmap/gucharmap-unicode-info.c
@@ -21,6 +21,8 @@
 #include <gtk/gtk.h>
 #include <string.h>
 
+#include <glib/gi18n-lib.h>
+
 #include "gucharmap.h"
 #include "gucharmap-private.h"
 
diff --git a/gucharmap/unicode.sha512sums b/gucharmap/unicode.sha512sums
new file mode 100644
index 0000000..dd8e650
--- /dev/null
+++ b/gucharmap/unicode.sha512sums
@@ -0,0 +1,6 @@
+7d11590e5678a06b51d9fab787fa4912120ea5e8dc8f89b2cddda9472dadd52514a74b0d3672f47e38f1ed3691e0599cdf6134cd4bbc7b5121977b57e8c842ef
  Blocks.txt
+9939b52a5ee51d4b6548199892f3a890324b793dd5fe81c70c8de977f737e3c91b9fc20b8b2241bb5b613d3946a06d9b0a6432cfa3b70432158ad0a78b40b032
  DerivedAge.txt
+7f9fb0349f64e8c84d1d2fa3416d561b30744d6f1fcb7d01efe1bc93dca2fe7f58148918270bfeb92551a1e08c7fefba3208af259bf35d8653a1c7772b8da259
  NamesList.txt
+c444218e9eb1090e0edf24f7dbafa9ff44031fa1b2124bf9133faa2022101a1e2f0cc56a5f014c3cb37fdc672fb65f14e67bc816ada082cdb256055c7009a9d2
  Scripts.txt
+34b46357ead07ec499746cac355e1fd6f3a1feea0fb786a28fe28985dd264323fba0ebf9ee6f7ab29bf412da191992aa3f6b41a394f817bcd1193f7eb06c1c91
  UnicodeData.txt
+4ab0f816041801720f626c264d5c9e4ded3f729c2de4b837428c9748b84950f1433243cdf8fb41ae58cbb32f401cebd547bf0eff268ff7549d5632c7baacd137
  Unihan.zip
[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]