pdf: Additional entries in Mac OS Roman encoding not in MacRomanEncoding.

When encoding truetype fonts via the mac roman cmap table, we should be
using the additional entries introduced in PDF 1.5, which are different
from the standard MacRomanEncoding table in the appendix.

5 files changed, 39 insertions, 16 deletions

diff --git a/include/mupdf/pdf/font.h b/include/mupdf/pdf/font.h
index d2421ab0..63177061 100644
--- a/include/mupdf/pdf/font.h
+++ b/include/mupdf/pdf/font.h
@@ -18,8 +18,8 @@ enum
 	PDF_FD_FORCE_BOLD = 1 << 18
 };
 
-void pdf_load_encoding(char **estrings, char *encoding);
-int pdf_lookup_agl(char *name);
+void pdf_load_encoding(const char **estrings, char *encoding);
+int pdf_lookup_agl(const char *name);
 const char **pdf_lookup_agl_duplicates(int ucs);
 
 extern const unsigned short pdf_doc_encoding[256];
@@ -99,7 +99,7 @@ void pdf_end_vmtx(fz_context *ctx, pdf_font_desc *font);
 pdf_hmtx pdf_lookup_hmtx(fz_context *ctx, pdf_font_desc *font, int cid);
 pdf_vmtx pdf_lookup_vmtx(fz_context *ctx, pdf_font_desc *font, int cid);
 
-void pdf_load_to_unicode(fz_context *ctx, pdf_document *doc, pdf_font_desc *font, char **strings, char *collection, pdf_obj *cmapstm);
+void pdf_load_to_unicode(fz_context *ctx, pdf_document *doc, pdf_font_desc *font, const char **strings, char *collection, pdf_obj *cmapstm);
 
 int pdf_font_cid_to_gid(fz_context *ctx, pdf_font_desc *fontdesc, int cid);
 
diff --git a/source/pdf/pdf-encoding.c b/source/pdf/pdf-encoding.c
index d0a4bb88..25843c12 100644
--- a/source/pdf/pdf-encoding.c
+++ b/source/pdf/pdf-encoding.c
@@ -4,19 +4,19 @@
 #include "pdf-glyphlist.h"
 
 void
-pdf_load_encoding(char **estrings, char *encoding)
+pdf_load_encoding(const char **estrings, char *encoding)
 {
-	char **bstrings = NULL;
+	const char * const *bstrings = NULL;
 	int i;
 
 	if (!strcmp(encoding, "StandardEncoding"))
-		bstrings = (char**) pdf_standard;
+		bstrings = pdf_standard;
 	if (!strcmp(encoding, "MacRomanEncoding"))
-		bstrings = (char**) pdf_mac_roman;
+		bstrings = pdf_mac_roman;
 	if (!strcmp(encoding, "MacExpertEncoding"))
-		bstrings = (char**) pdf_mac_expert;
+		bstrings = pdf_mac_expert;
 	if (!strcmp(encoding, "WinAnsiEncoding"))
-		bstrings = (char**) pdf_win_ansi;
+		bstrings = pdf_win_ansi;
 
 	if (bstrings)
 		for (i = 0; i < 256; i++)
@@ -24,7 +24,7 @@ pdf_load_encoding(char **estrings, char *encoding)
 }
 
 int
-pdf_lookup_agl(char *name)
+pdf_lookup_agl(const char *name)
 {
 	char buf[64];
 	char *p;
diff --git a/source/pdf/pdf-font.c b/source/pdf/pdf-font.c
index a974c8d5..eaeac3c0 100644
--- a/source/pdf/pdf-font.c
+++ b/source/pdf/pdf-font.c
@@ -178,9 +178,9 @@ static int ft_char_index(FT_Face face, int cid)
 	return gid;
 }
 
-static int ft_name_index(FT_Face face, char *name)
+static int ft_name_index(FT_Face face, const char *name)
 {
-	int code = FT_Get_Name_Index(face, name);
+	int code = FT_Get_Name_Index(face, (char*)name);
 	if (code == 0)
 	{
 		int unicode = pdf_lookup_agl(name);
@@ -300,9 +300,32 @@ static int ft_width(fz_context *ctx, pdf_font_desc *fontdesc, int cid)
 	return adv * 1000 / ((FT_Face)fontdesc->font->ft_face)->units_per_EM;
 }
 
-static int lookup_mre_code(char *name)
+static const struct { int code; const char *name; } mre_diff_table[] =
+{
+	{ 173, "notequal" },
+	{ 176, "infinity" },
+	{ 178, "lessequal" },
+	{ 179, "greaterequal" },
+	{ 182, "partialdiff" },
+	{ 183, "summation" },
+	{ 184, "product" },
+	{ 185, "pi" },
+	{ 186, "integral" },
+	{ 189, "Omega" },
+	{ 195, "radical" },
+	{ 197, "approxequal" },
+	{ 198, "Delta" },
+	{ 215, "lozenge" },
+	{ 219, "Euro" },
+	{ 240, "apple" },
+};
+
+static int lookup_mre_code(const char *name)
 {
 	int i;
+	for (i = 0; i < nelem(mre_diff_table); ++i)
+		if (!strcmp(name, mre_diff_table[i].name))
+			return mre_diff_table[i].code;
 	for (i = 0; i < 256; i++)
 		if (pdf_mac_roman[i] && !strcmp(name, pdf_mac_roman[i]))
 			return i;
@@ -557,7 +580,7 @@ pdf_load_simple_font_by_name(fz_context *ctx, pdf_document *doc, pdf_obj *dict,
 	int kind;
 	int glyph;
 
-	char *estrings[256];
+	const char *estrings[256];
 	char ebuffer[256][32];
 	int i, k, n;
 	int fterr;
diff --git a/source/pdf/pdf-type3.c b/source/pdf/pdf-type3.c
index 677a26e5..3eeff1f8 100644
--- a/source/pdf/pdf-type3.c
+++ b/source/pdf/pdf-type3.c
@@ -19,7 +19,7 @@ pdf_font_desc *
 pdf_load_type3_font(fz_context *ctx, pdf_document *doc, pdf_obj *rdb, pdf_obj *dict)
 {
 	char buf[256];
-	char *estrings[256];
+	const char *estrings[256];
 	pdf_font_desc *fontdesc = NULL;
 	pdf_obj *encoding;
 	pdf_obj *widths;
diff --git a/source/pdf/pdf-unicode.c b/source/pdf/pdf-unicode.c
index 65bda460..fdff47ba 100644
--- a/source/pdf/pdf-unicode.c
+++ b/source/pdf/pdf-unicode.c
@@ -57,7 +57,7 @@ pdf_remap_cmap(fz_context *ctx, pdf_cmap *gid_from_cpt, pdf_cmap *ucs_from_cpt)
 
 void
 pdf_load_to_unicode(fz_context *ctx, pdf_document *doc, pdf_font_desc *font,
-	char **strings, char *collection, pdf_obj *cmapstm)
+	const char **strings, char *collection, pdf_obj *cmapstm)
 {
 	unsigned int cpt;