aboutsummaryrefslogtreecommitdiff
path: root/scripts
diff options
context:
space:
mode:
authorDaniel Eklöf <daniel@ekloef.se>2020-05-02 16:25:15 +0000
committerDaniel Eklöf <daniel@ekloef.se>2020-05-02 16:25:15 +0000
commita24a3520c7e0230b0b3b853db0cb2871427d1ab4 (patch)
tree2c0f43dec853c1edd42dadf30c1d1538a490fcda /scripts
parentreadme: utf8proc is no longer a dependency (diff)
script: generate-unicode-precompose: generate a complete header file
Diffstat (limited to 'scripts')
-rwxr-xr-xscripts/generate-unicode-precompose.sh29
1 files changed, 27 insertions, 2 deletions
diff --git a/scripts/generate-unicode-precompose.sh b/scripts/generate-unicode-precompose.sh
index 830e9379..49e12650 100755
--- a/scripts/generate-unicode-precompose.sh
+++ b/scripts/generate-unicode-precompose.sh
@@ -1,8 +1,33 @@
#!/usr/bin/sh
-cut - -d ";" -f 1,6 |
+unicodedata_txt="${1}"
+output="${2}"
+
+cat <<EOF > "${output}"
+#pragma once
+
+#include <wchar.h>
+
+static const struct {
+ wchar_t replacement;
+ wchar_t base;
+ wchar_t comb;
+} precompose_table[] = {
+EOF
+
+# extract canonical decomposition data from UnicodeData.txt,
+# - pad hex values to 5 digits,
+# - sort numerically on base character, then combining character,
+# - then reduce to 4 digits again where possible
+#
+# "borrowed" from xterm/unicode/make-precompose.sh
+
+cut "${unicodedata_txt}" -d ";" -f 1,6 |
grep ";[0-9,A-F]" | grep " " |
sed -e "s/ /, 0x/;s/^/{ 0x/;s/;/, 0x/;s/$/},/" |
sed -e "s,0x\(....\)\([^0-9A-Fa-f]\),0x0\1\2,g" |
(sort -k 3 || sort +2) |
- sed -e "s,0x0\(...[0-9A-Fa-f]\),0x\1,g"
+ sed -e "s,0x0\(...[0-9A-Fa-f]\),0x\1,g" |
+ sed 's/^/ /' >> "${output}"
+
+echo "};" >> "${output}"