Check in script used to create encoding tables.

1 files changed, 49 insertions, 0 deletions

diff --git a/scripts/makeencoding.py b/scripts/makeencoding.py
new file mode 100644
index 00000000..054be9c0
--- /dev/null
+++ b/scripts/makeencoding.py
@@ -0,0 +1,49 @@
+# Convert unicode mapping table to C arrays mapping glyph names and unicode values.
+#
+# ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MISC/KOI8-U.TXT
+# ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-7.TXT
+#
+
+glyphs = {}
+for line in open("scripts/glyphlist.txt").readlines():
+	if line[0] != '#':
+		n, u = line.rstrip().split(';')
+		if len(u) == 4:
+			u = int(u, base=16)
+			glyphs[u] = n
+
+def load_table(fn):
+	table = [0] * 256
+	for line in open(fn).readlines():
+		if line[0] != '#':
+			line = line.split()
+			c = int(line[0][2:], base=16)
+			u = int(line[1][2:], base=16)
+			table[c] = u
+	return table
+
+def dump_table(name, table):
+	print "const char *pdf_glyph_name_from_%s[%d] = {" % (name, len(table))
+	for u in table:
+		if u in glyphs:
+			print '"%s",' % glyphs[u]
+		else:
+			print '_notdef,'
+	print "};"
+	print
+	print "static const struct { unsigned short u, c; } %s_from_unicode[] = {" % name
+	rev = []
+	i = 0
+	for u in table:
+		if u in glyphs:
+			if u >= 128:
+				rev += ['{0x%04x,%d},' % (u, i)]
+		i = i + 1
+	rev.sort()
+	for s in rev:
+		print s
+	print "};"
+	print
+
+dump_table("koi8u", load_table("scripts/KOI8-U.TXT"))
+dump_table("iso8859_7", load_table("scripts/8859-7.TXT"))