unicode: Support characters beyond the first unicode plane

We used 16-bit variables to store the 'character code' everywhere but
this won't let us represent anything beyond U+FFFF.

This patch changes those variables to a custom type that can be 32 or 16
bits depending on the build, and adjusts numerous internal APIs and
datastructures to match.  This includes:

 * utf8decode() and friends
 * font manipulation, caching, rendering, and generation
 * on-screen keyboard
 * FAT filesystem (parsing and generating utf16 LFNs)
 * WIN32 simulator platform code

Note that this patch doesn't _enable_ >16bit unicode support; a followup
patch will turn that on for appropriate targets.

Appears to work on:

  * hosted linux, native, linux simulator in both 16/32-bit modes.

Needs testing on:

  * windows and macos simulator (16bit+32bit)

Change-Id: Iba111b27d2433019b6bff937cf1ebd2c4353a0e8

firmware/export/bidi.h

@@ -21,6 +21,6 @@
 #ifndef BIDI_H
 #define BIDI_H
 
-extern unsigned short *bidi_l2v(const unsigned char *str, int orientation);
+ucschar_t *bidi_l2v(const unsigned char *str, int orientation);
 
 #endif /* BIDI_H */

firmware/export/config.h

@@ -1461,4 +1461,11 @@ Lyre prototype 1 */
 #error "HAVE_LCD_SLEEP_SETTING requires HAVE_LCD_SLEEP"
 #endif
 
+// XXX Figure out a better place to put this?
+#ifdef UNICODE32
+#define ucschar_t unsigned int
+#else
+#define ucschar_t unsigned short
+#endif
+
 #endif /* __CONFIG_H__ */

firmware/export/cpu.h

@@ -18,6 +18,9 @@
  * KIND, either express or implied.
  *
  ****************************************************************************/
+#ifndef __CPU_H
+#define __CPU_H
+
 #include "config.h"
 
 #if CONFIG_CPU == MCF5249
@@ -80,3 +83,5 @@
 #if CONFIG_CPU == STM32H743
 #include "cpu-stm32h743.h"
 #endif
+
+#endif /* __CPU_H */

firmware/export/font.h

@@ -86,7 +86,7 @@ struct font {
     int          maxwidth;        /* max width in pixels*/
     unsigned int height;          /* height in pixels*/
     int          ascent;          /* ascent (baseline) height*/
-    int          firstchar;       /* first character in bitmap*/
+    unsigned int firstchar;       /* first character in bitmap*/
     int          size;            /* font size in glyphs*/
     int          depth;           /* depth of the font, 0=1bit and 1=4bit */
     const unsigned char *bits;    /* 8-bit column bitmap data*/
@@ -95,24 +95,24 @@ struct font {
     const unsigned char *width;   /* character widths or NULL if fixed*/
     int          defaultchar;     /* default char (not glyph index)*/
     int32_t      bits_size;       /* # bytes of glyph bits*/
-    
+
     /* file, buffer and cache management */
     int          fd;              /* fd for the font file. >= 0 if cached */
     int          fd_width;        /* fd for the font file. >= 0 if cached */
-    int          fd_offset;       /* fd for the font file. >= 0 if cached */    
+    int          fd_offset;       /* fd for the font file. >= 0 if cached */
     int          handle;          /* core_allocator handle */
-    unsigned char *buffer_start;    /* buffer to store the font in */       
-    unsigned char *buffer_position; /* position in the buffer */    
+    unsigned char *buffer_start;    /* buffer to store the font in */
+    unsigned char *buffer_position; /* position in the buffer */
     unsigned char *buffer_end;      /* end of the buffer */
     size_t         buffer_size;     /* size of the buffer in bytes */
     bool         disabled;        /* font disabled (use blank as fallback if not in cache) */
-#ifndef __PCTOOL__    
+#ifndef __PCTOOL__
     struct font_cache cache;
     uint32_t file_width_offset;    /* offset to file width data    */
     uint32_t file_offset_offset;   /* offset to file offset data   */
     int long_offset;
-#endif    
-    
+#endif
+
 };
 
 /* font routines*/
@@ -134,7 +134,7 @@ void font_enable_all(void);
 struct font* font_get(int font);
 int font_getstringnsize(const unsigned char *str, size_t maxbytes, int *w, int *h, int fontnumber);
 int font_getstringsize(const unsigned char *str, int *w, int *h, int fontnumber);
-int font_get_width(struct font* ft, unsigned short ch);
-const unsigned char * font_get_bits(struct font* ft, unsigned short ch);
+int font_get_width(struct font* ft, ucschar_t ch);
+const unsigned char * font_get_bits(struct font* ft, ucschar_t ch);
 
 #endif

firmware/export/hangul.h

@@ -21,5 +21,4 @@
 
 extern const char jamo_table[51][3];
 
-unsigned short hangul_join(unsigned short lead, unsigned short vowel,
-                                unsigned short tail);
+ucschar_t hangul_join(ucschar_t lead, ucschar_t vowel, ucschar_t tail);