[OpenSPARC-T2-DV] / tools / perl-5.8.0 / lib / 5.8.0 / Encode / encode.h

#ifndef ENCODE_H
#define ENCODE_H

#ifndef U8
/* 
   A tad devious this:
   perl normally has a #define for U8 - if that isn't present then we
   typedef it - leaving it #ifndef so we can do data parts without
   getting extern references to the code parts
*/
typedef unsigned char U8;
#endif

typedef struct encpage_s encpage_t;

struct encpage_s
{
	/* fields ordered to pack nicely on 32-bit machines */
	const U8   *seq;       /* Packed output sequences we generate 
				  if we match */
	encpage_t  *next;      /* Page to go to if we match */
	U8         min;        /* Min value of octet to match this entry */
	U8         max;        /* Max value of octet to match this entry */
	U8         dlen;       /* destination length - 
				  size of entries in seq */
	U8         slen;       /* source length - 
				  number of source octets needed */
};

/*
  At any point in a translation there is a page pointer which points
  at an array of the above structures.

  Basic operation :
  get octet from source stream.
  if (octet >= min && octet < max) {
    if slen is 0 then we cannot represent this character.
    if we have less than slen octets (including this one) then 
      we have a partial character.
    otherwise
      copy dlen octets from seq + dlen*(octet-min) to output
      (dlen may be zero if we don't know yet.)
      load page pointer with next to continue.
      (is slen is one this is end of a character)
      get next octet.
  }
  else {
    increment the page pointer to look at next slot in the array
  }

  arrays SHALL be constructed so there is an entry which matches
  ..0xFF at the end, and either maps it or indicates no
  representation.

  if MSB of slen is set then mapping is an approximate "FALLBACK" entry.

*/


typedef struct encode_s encode_t;
struct encode_s
{
	encpage_t  *t_utf8;    /* Starting table for translation from 
				  the encoding to UTF-8 form */
	encpage_t  *f_utf8;    /* Starting table for translation 
				  from UTF-8 to the encoding */
	const U8   *rep;       /* Replacement character in this encoding 
				  e.g. "?" */
	int        replen;     /* Number of octets in rep */
	U8         min_el;     /* Minimum octets to represent a character */
	U8         max_el;     /* Maximum octets to represent a character */
	const char *name[2];   /* name(s) of this encoding */
};

#ifdef U8
/* See comment at top of file for deviousness */

extern int do_encode(encpage_t *enc, const U8 *src, STRLEN *slen,
                     U8 *dst, STRLEN dlen, STRLEN *dout, int approx);

extern void Encode_DefineEncoding(encode_t *enc);

#endif /* U8 */

#define ENCODE_NOSPACE  1
#define ENCODE_PARTIAL  2
#define ENCODE_NOREP    3
#define ENCODE_FALLBACK 4

#define FBCHAR_UTF8		"\xEF\xBF\xBD"

#define  ENCODE_DIE_ON_ERR     0x0001 /* croaks immediately */
#define  ENCODE_WARN_ON_ERR    0x0002 /* warn on error; may proceed */
#define  ENCODE_RETURN_ON_ERR  0x0004 /* immediately returns on NOREP */
#define  ENCODE_LEAVE_SRC      0x0008 /* $src updated unless set */
#define  ENCODE_PERLQQ         0x0100 /* perlqq fallback string */
#define  ENCODE_HTMLCREF       0x0200 /* HTML character ref. fb mode */
#define  ENCODE_XMLCREF        0x0400 /* XML  character ref. fb mode */

#define  ENCODE_FB_DEFAULT     0x0000
#define  ENCODE_FB_CROAK       0x0001
#define  ENCODE_FB_QUIET       ENCODE_RETURN_ON_ERR
#define  ENCODE_FB_WARN        (ENCODE_RETURN_ON_ERR|ENCODE_WARN_ON_ERR)
#define  ENCODE_FB_PERLQQ      ENCODE_PERLQQ
#define  ENCODE_FB_HTMLCREF    ENCODE_HTMLCREF
#define  ENCODE_FB_XMLCREF     ENCODE_XMLCREF

#endif /* ENCODE_H */
Commit	Line	Data
86530b38 AT	1	#ifndef ENCODE_H
	2	#define ENCODE_H
	3
	4	#ifndef U8
	5	/*
	6	A tad devious this:
	7	perl normally has a #define for U8 - if that isn't present then we
	8	typedef it - leaving it #ifndef so we can do data parts without
	9	getting extern references to the code parts
	10	*/
	11	typedef unsigned char U8;
	12	#endif
	13
	14	typedef struct encpage_s encpage_t;
	15
	16	struct encpage_s
	17	{
	18	/* fields ordered to pack nicely on 32-bit machines */
	19	const U8 seq; / Packed output sequences we generate
	20	if we match */
	21	encpage_t next; / Page to go to if we match */
	22	U8 min; /* Min value of octet to match this entry */
	23	U8 max; /* Max value of octet to match this entry */
	24	U8 dlen; /* destination length -
	25	size of entries in seq */
	26	U8 slen; /* source length -
	27	number of source octets needed */
	28	};
	29
	30	/*
	31	At any point in a translation there is a page pointer which points
	32	at an array of the above structures.
	33
	34	Basic operation :
	35	get octet from source stream.
	36	if (octet >= min && octet < max) {
	37	if slen is 0 then we cannot represent this character.
	38	if we have less than slen octets (including this one) then
	39	we have a partial character.
	40	otherwise
	41	copy dlen octets from seq + dlen*(octet-min) to output
	42	(dlen may be zero if we don't know yet.)
	43	load page pointer with next to continue.
	44	(is slen is one this is end of a character)
	45	get next octet.
	46	}
	47	else {
	48	increment the page pointer to look at next slot in the array
	49	}
	50
	51	arrays SHALL be constructed so there is an entry which matches
	52	..0xFF at the end, and either maps it or indicates no
	53	representation.
	54
	55	if MSB of slen is set then mapping is an approximate "FALLBACK" entry.
	56
	57	*/
	58
	59
	60	typedef struct encode_s encode_t;
	61	struct encode_s
	62	{
	63	encpage_t t_utf8; / Starting table for translation from
	64	the encoding to UTF-8 form */
65	encpage_t f_utf8; / Starting table for translation
66	from UTF-8 to the encoding */
67	const U8 rep; / Replacement character in this encoding
68	e.g. "?" */
69	int replen; /* Number of octets in rep */
70	U8 min_el; /* Minimum octets to represent a character */
71	U8 max_el; /* Maximum octets to represent a character */
72	const char name[2]; / name(s) of this encoding */
73	};
74
75	#ifdef U8
76	/* See comment at top of file for deviousness */
77
78	extern int do_encode(encpage_t enc, const U8 src, STRLEN *slen,
79	U8 dst, STRLEN dlen, STRLEN dout, int approx);
80
81	extern void Encode_DefineEncoding(encode_t *enc);
82
83	#endif /* U8 */
84
85	#define ENCODE_NOSPACE 1
86	#define ENCODE_PARTIAL 2
87	#define ENCODE_NOREP 3
88	#define ENCODE_FALLBACK 4
89
90	#define FBCHAR_UTF8 "\xEF\xBF\xBD"
91
92	#define ENCODE_DIE_ON_ERR 0x0001 /* croaks immediately */
93	#define ENCODE_WARN_ON_ERR 0x0002 /* warn on error; may proceed */
94	#define ENCODE_RETURN_ON_ERR 0x0004 /* immediately returns on NOREP */
95	#define ENCODE_LEAVE_SRC 0x0008 /* $src updated unless set */
96	#define ENCODE_PERLQQ 0x0100 /* perlqq fallback string */
97	#define ENCODE_HTMLCREF 0x0200 /* HTML character ref. fb mode */
98	#define ENCODE_XMLCREF 0x0400 /* XML character ref. fb mode */
99
100	#define ENCODE_FB_DEFAULT 0x0000
101	#define ENCODE_FB_CROAK 0x0001
102	#define ENCODE_FB_QUIET ENCODE_RETURN_ON_ERR
103	#define ENCODE_FB_WARN (ENCODE_RETURN_ON_ERR\|ENCODE_WARN_ON_ERR)
104	#define ENCODE_FB_PERLQQ ENCODE_PERLQQ
105	#define ENCODE_FB_HTMLCREF ENCODE_HTMLCREF
106	#define ENCODE_FB_XMLCREF ENCODE_XMLCREF
107
108	#endif /* ENCODE_H */