get TTFINFO_FONT_PSNAME

PR:		37246
Submitted by:	Statue <statue@softwareliberty.org>

1 files changed, 141 insertions, 0 deletions

diff --git a/chinese/ttfm/files/patch-ttfinfo.c b/chinese/ttfm/files/patch-ttfinfo.c
new file mode 100644
index 0000000..8c52e5a
--- /dev/null
+++ b/chinese/ttfm/files/patch-ttfinfo.c
@@ -0,0 +1,141 @@
+--- ttfinfo.c.orig	Fri Apr 19 18:30:44 2002
++++ ttfinfo.c	Fri Apr 19 18:31:02 2002
+@@ -112,6 +112,7 @@
+ static void  usage             (const char *name);
+ static const char *get_foundry (const TT_Char orig_code[4]);
+ static const char *get_name    (TT_Face face);
++static const char *get_psname  (TT_Face face);
+ static void  add_mapping (struct xlfdmapping **listhead,
+ 			  char *file_name,
+ 			  const char *foundry_name,
+@@ -684,6 +685,7 @@
+     TT_Face face;
+     TT_Face_Properties  properties;
+     const char *font_name;
++    const char *font_psname;
+     const char *foundry_name;
+     const char *weight_name;
+     const char *width_name;
+@@ -722,6 +724,7 @@
+     }
+ 
+     font_name    = get_name (face);
++    font_psname  = get_psname (face);
+     foundry_name = get_foundry (properties.os2->achVendID);
+     weight_name  = get_weight (path, &properties);
+     width_name   = get_width_name (path, &properties);
+@@ -729,6 +732,7 @@
+     /* basic fone information */
+     fprintf(outstream,"TTFINFO_FONT_FILE=\"%s\"\n",path);
+     fprintf(outstream,"TTFINFO_FONT_NAME=\"%s\"\n",font_name);
++    fprintf(outstream,"TTFINFO_FONT_PSNAME=\"%s\"\n",font_psname);
+     fprintf(outstream,"TTFINFO_FOUNDRY_NAME=\"%s\"\n",foundry_name);
+     fprintf(outstream,"TTFINFO_WEIGHT_NAME=\"%s\"\n",weight_name);
+     fprintf(outstream,"TTFINFO_WIDTH=\"%s\"\n",width_name);
+@@ -1206,7 +1210,7 @@
+        function returns as soon as an english name for the font is found,
+        so most of the time one loop is enough; however some fonts have only
+        non-english names in their font faminly entry, so the second loop
+-       to try the Postscript name as last ressort
++       to try the Postscript name as last resort
+     */
+     for (k = 0; k < 2; k++) {
+ 
+@@ -1276,6 +1280,97 @@
+ 		    }
+ 	    }
+ 	}
++    }
++ 
++    /* Not found */
++    return "unknown";
++}
++
++/* This get_psname() function is copied from get_name().  I probably should
++   try to merge the two functions somehow, but unfortunately, I do not have
++   the time right now.  :-)  Anthony Fok <anthony@thizlinux.com>
++*/
++
++const char *
++get_psname (TT_Face face)
++{
++    int i, j, k, n;
++
++    TT_UShort  platform;
++    TT_UShort  encoding;
++    TT_UShort  language;
++    TT_UShort  NameId;
++
++    /* get the number of name strings in this font */
++    if ((n = TT_Get_Name_Count (face)) < 0) {
++	return NULL;
++    }
++    /* printf("\nn=%d\n",n); */
++
++    /* for (k = 0; k < 2; k++) { */
++
++    k = 1;
++	/* now look at every entry */
++	for (i = 0; i < n; i++) {
++	    TT_Get_Name_ID (face, i, &platform, &encoding, &language, &NameId);
++	    /*	    printf("k=%d, face=%d, i=%d, platform=%d, encoding=%d, lang=%d, Id=%d\n",k ,face, i, platform, encoding, language, NameId); */
++
++	    if (NameId == TT_NAME_ID_PS_NAME) {
++
++		/* Try a Macintosh english name */
++		if ((platform == TT_PLATFORM_MACINTOSH) && 
++		    (language == TT_MAC_LANGID_ENGLISH))
++		    {
++			char*  string;
++			TT_UShort string_len;
++			static char  name_buffer[257];
++			int   name_len;
++			int   english=1;
++
++			TT_Get_Name_String (face, i, &string, &string_len);
++			string_len =  string_len < 257 ? string_len : 257;
++			name_len = 0;
++
++			for (j = 0; j < string_len; j++) {
++			    /* some font names claim to be in english but aren't */
++			    if ((string[j] & 0xff) > 0x7f) english=0;
++			    name_buffer[name_len++] = string[j];
++			}
++
++			name_buffer[name_len] = '\0';
++			if (english) return name_buffer;
++		    }
++		/* Try to find an Apple Unicode or Microsoft English name */
++		if ((platform == TT_PLATFORM_APPLE_UNICODE) ||
++		    ((platform == TT_PLATFORM_MICROSOFT) &&
++		     ((language & 0x00FF) ==
++		      (TT_MS_LANGID_ENGLISH_UNITED_STATES & 0x00FF))) ||
++		    (((platform == TT_PLATFORM_MICROSOFT) &&
++		      (encoding == TT_MS_ID_UNICODE_CS)) &&
++		     (NameId == TT_NAME_ID_PS_NAME)))
++		    {
++			/* The following code was inspired from Mark Leisher's */
++			/* ttf2bdf package                                     */
++			char*  string;
++			TT_UShort string_len;
++			static char  name_buffer[513];
++			int   name_len;
++
++			TT_Get_Name_String (face, i, &string, &string_len);
++			string_len =  string_len < 512 ? string_len : 512;
++			name_len = 0;
++			/* convert from unicode to ASCII */
++			for (j = 0; j < string_len; j ++) {
++			  /* drop the NULL part, dont't use j+=2 to avoid
++			     endian problem. */
++			  if (string[j] != '\0'){
++			    name_buffer[name_len++] = string[j];
++			  }
++			}
++			name_buffer[name_len] = '\0';
++			return name_buffer;
++		    }
++	    }
+     }
+  
+     /* Not found */