Fetch UnicodeData.txt from unicode.org if not present to make the build fully automated.

.gitignore

@@ -12,6 +12,7 @@
 
 # build output:
 
+data/
 src/libucd.la
 tests/printucddata
 

Makefile.am

@@ -54,14 +54,21 @@ EXTRA_DIST += ChangeLog
 
 ############################# Unicode Character Database ######################
 
-src/case.cpp: tools/case.py tools/ucd.py
-	tools/case.py ${UCD_ROOTDIR} > $@
+UCD_VERSION=6.2.0
+UCD_ROOTDIR=data/ucd
 
-src/categories.cpp: tools/categories.py tools/ucd.py
-	tools/categories.py ${UCD_ROOTDIR} > $@
+data/ucd/UnicodeData.txt:
+	mkdir -pv data/ucd
+	wget -O $@ http://www.unicode.org/Public/${UCD_VERSION}/ucd/UnicodeData.txt
 
 ############################# libucd ##########################################
 
+src/case.cpp: tools/case.py tools/ucd.py data/ucd/UnicodeData.txt
+	tools/case.py ${UCD_ROOTDIR} ${UCD_VERSION} > $@
+
+src/categories.cpp: tools/categories.py tools/ucd.py data/ucd/UnicodeData.txt
+	tools/categories.py ${UCD_ROOTDIR} ${UCD_VERSION} > $@
+
 libucd_includedir = $(includedir)/ucd
 libucd_include_HEADERS = \
 	src/include/ucd/ucd.h

src/case.cpp

@@ -27,7 +27,7 @@
 
 using namespace ucd;
 
-// Unicode Character Data 6.2
+// Unicode Character Data 6.2.0
 
 struct case_conversion_entry
 {

src/categories.cpp

@@ -27,7 +27,7 @@
 
 using namespace ucd;
 
-// Unicode Character Data 6.2
+// Unicode Character Data 6.2.0
 
 static const uint8_t categories_Cn[256] =
 {

tools/case.py

@@ -22,7 +22,7 @@ import sys
 import ucd
 
 ucd_rootdir = sys.argv[1]
-ucd_version = ucd_rootdir.split('-')[-1]
+ucd_version = sys.argv[2]
 
 unicode_chars = {}
 null = ucd.CodePoint('0000')

tools/categories.py

@@ -22,7 +22,7 @@ import sys
 import ucd
 
 ucd_rootdir = sys.argv[1]
-ucd_version = ucd_rootdir.split('-')[-1]
+ucd_version = sys.argv[2]
 
 unicode_chars = {}
 for data in ucd.parse_ucd_data(ucd_rootdir, 'UnicodeData'):