Timerix/utf8proc
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

give up on Unifont for charwidth data (#150)

Browse Source 
* fix CHARBOUND option for non-characters

* give up on unifont for charwidth computation
This commit is contained in:
Steven G. Johnson
2019-03-30 16:05:50 -04:00
committed by GitHub
parent 94395db282
commit fd4d8a3454
3 changed files with 4885 additions and 5205 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
data/Makefile
View File

@@ -1,7 +1,7 @@
# Unicode data generation rules. Except for the test data files, most
# users will not use these Makefile rules, which are primarily to re-generate
# unicode_data.c when we get a new Unicode version or charwidth data; they
# require ruby, fontforge, and julia to be installed.
# require ruby and julia to be installed.
# programs
CURL=curl
@@ -9,7 +9,6 @@ RUBY=ruby
PERL=perl
MAKE=make
JULIA=julia
FONTFORGE=fontforge
CURLFLAGS = --retry 5 --location
.PHONY: clean
@@ -19,19 +18,7 @@ CURLFLAGS = --retry 5 --location
utf8proc_data.c.new: data_generator.rb UnicodeData.txt GraphemeBreakProperty.txt DerivedCoreProperties.txt CompositionExclusions.txt CaseFolding.txt CharWidths.txt emoji-data.txt
$(RUBY) data_generator.rb < UnicodeData.txt > $@
# GNU Unifont version for font metric calculations:
UNIFONT_VERSION=12.0.01
unifont.ttf:
$(CURL) $(CURLFLAGS) -o $@ $(URLCACHE)https://mirrors.kernel.org/gnu/unifont/unifont-$(UNIFONT_VERSION)/unifont-$(UNIFONT_VERSION).ttf
unifont_upper.ttf:
$(CURL) $(CURLFLAGS) -o $@ $(URLCACHE)https://mirrors.kernel.org/gnu/unifont/unifont-$(UNIFONT_VERSION)/unifont_upper-$(UNIFONT_VERSION).ttf
%.sfd: %.ttf
$(FONTFORGE) -lang=ff -c "Open(\"$<\");Save(\"$@\");Quit(0);"
CharWidths.txt: charwidths.jl unifont.sfd unifont_upper.sfd EastAsianWidth.txt
CharWidths.txt: charwidths.jl EastAsianWidth.txt
$(JULIA) charwidths.jl > $@
# Unicode data version
@@ -65,5 +52,5 @@ emoji-data.txt:
$(CURL) $(CURLFLAGS) -o $@ -O $(URLCACHE)http://unicode.org/Public/emoji/`echo $(UNICODE_VERSION) | cut -d. -f1-2`/emoji-data.txt
clean:
rm -f UnicodeData.txt EastAsianWidth.txt GraphemeBreakProperty.txt DerivedCoreProperties.txt CompositionExclusions.txt CaseFolding.txt NormalizationTest.txt GraphemeBreakTest.txt CharWidths.txt unifont*.ttf unifont*.sfd emoji-data.txt
rm -f UnicodeData.txt EastAsianWidth.txt GraphemeBreakProperty.txt DerivedCoreProperties.txt CompositionExclusions.txt CaseFolding.txt NormalizationTest.txt GraphemeBreakTest.txt CharWidths.txt emoji-data.txt
rm -f utf8proc_data.c.new