code/trunk/ucpinternal.h

/*************************************************
*           Unicode Property Table handler       *
*************************************************/

#ifndef _UCPINTERNAL_H
#define _UCPINTERNAL_H

/* Internal header file defining the layout of the bits in each pair of 32-bit
words that form a data item in the table. */

typedef struct cnode {
  pcre_uint32 f0;
  pcre_uint32 f1;
} cnode;

/* Things for the f0 field */

#define f0_scriptmask   0xff000000  /* Mask for script field */
#define f0_scriptshift          24  /* Shift for script value */
#define f0_rangeflag    0x00800000  /* Flag for a range item */
#define f0_charmask     0x001fffff  /* Mask for code point value */

/* Things for the f1 field */

#define f1_typemask     0xfc000000  /* Mask for char type field */
#define f1_typeshift            26  /* Shift for the type field */
#define f1_rangemask    0x0000ffff  /* Mask for a range offset */
#define f1_casemask     0x0000ffff  /* Mask for a case offset */
#define f1_caseneg      0xffff8000  /* Bits for negation */

/* The data consists of a vector of structures of type cnode. The two unsigned
32-bit integers are used as follows:

(f0) (1) The most significant byte holds the script number. The numbers are
         defined by the enum in ucp.h.

     (2) The 0x00800000 bit is set if this entry defines a range of characters.
         It is not set if this entry defines a single character

     (3) The 0x00600000 bits are spare.

     (4) The 0x001fffff bits contain the code point. No Unicode code point will
         ever be greater than 0x0010ffff, so this should be OK for ever.

(f1) (1) The 0xfc000000 bits contain the character type number. The numbers are
         defined by an enum in ucp.h.

     (2) The 0x03ff0000 bits are spare.

     (3) The 0x0000ffff bits contain EITHER the unsigned offset to the top of
         range if this entry defines a range, OR the *signed* offset to the
         character's "other case" partner if this entry defines a single
         character. There is no partner if the value is zero.

-------------------------------------------------------------------------------
| script (8) |.|.|.| codepoint (21) || type (6) |.|.| spare (8) | offset (16) |
-------------------------------------------------------------------------------
              | | |                              | |
              | | |-> spare                      | |-> spare
              | |                                |
              | |-> spare                        |-> spare
              |
              |-> range flag

The upper/lower casing information is set only for characters that come in
pairs. The non-one-to-one mappings in the Unicode data are ignored.

When searching the data, proceed as follows:

(1) Set up for a binary chop search.

(2) If the top is not greater than the bottom, the character is not in the
    table. Its type must therefore be "Cn" ("Undefined").

(3) Find the middle vector element.

(4) Extract the code point and compare. If equal, we are done.

(5) If the test character is smaller, set the top to the current point, and
    goto (2).

(6) If the current entry defines a range, compute the last character by adding
    the offset, and see if the test character is within the range. If it is,
    we are done.

(7) Otherwise, set the bottom to one element past the current point and goto
    (2).
*/

#endif /* _UCPINTERNAL_H */

/* End of ucpinternal.h */
1	nigel	75	/*************************************************
2	nigel	87	* Unicode Property Table handler *
3	nigel	75	*************************************************/
4
5	nigel	93	#ifndef _UCPINTERNAL_H
6			#define _UCPINTERNAL_H
7
8	nigel	87	/* Internal header file defining the layout of the bits in each pair of 32-bit
9			words that form a data item in the table. */
10	nigel	75
11			typedef struct cnode {
12	nigel	87	pcre_uint32 f0;
13			pcre_uint32 f1;
14	nigel	75	} cnode;
15
16			/* Things for the f0 field */
17
18	nigel	87	#define f0_scriptmask 0xff000000 /* Mask for script field */
19			#define f0_scriptshift 24 /* Shift for script value */
20	ph10	346	#define f0_rangeflag 0x00800000 /* Flag for a range item */
21	nigel	87	#define f0_charmask 0x001fffff /* Mask for code point value */
22	nigel	75
23	nigel	87	/* Things for the f1 field */
24	nigel	75
25	nigel	87	#define f1_typemask 0xfc000000 /* Mask for char type field */
26			#define f1_typeshift 26 /* Shift for the type field */
27			#define f1_rangemask 0x0000ffff /* Mask for a range offset */
28			#define f1_casemask 0x0000ffff /* Mask for a case offset */
29			#define f1_caseneg 0xffff8000 /* Bits for negation */
30	nigel	75
31	nigel	87	/* The data consists of a vector of structures of type cnode. The two unsigned
32			32-bit integers are used as follows:
33	nigel	75
34	nigel	87	(f0) (1) The most significant byte holds the script number. The numbers are
35			defined by the enum in ucp.h.
36	nigel	75
37	nigel	87	(2) The 0x00800000 bit is set if this entry defines a range of characters.
38			It is not set if this entry defines a single character
39	nigel	75
40	nigel	87	(3) The 0x00600000 bits are spare.
41	nigel	75
42	nigel	87	(4) The 0x001fffff bits contain the code point. No Unicode code point will
43			ever be greater than 0x0010ffff, so this should be OK for ever.
44	nigel	75
45	nigel	87	(f1) (1) The 0xfc000000 bits contain the character type number. The numbers are
46			defined by an enum in ucp.h.
47	nigel	75
48	nigel	87	(2) The 0x03ff0000 bits are spare.
49	nigel	75
50	nigel	87	(3) The 0x0000ffff bits contain EITHER the unsigned offset to the top of
51			range if this entry defines a range, OR the signed offset to the
52			character's "other case" partner if this entry defines a single
53			character. There is no partner if the value is zero.
54	nigel	75
55	nigel	87	-------------------------------------------------------------------------------
56			\| script (8) \|.\|.\|.\| codepoint (21) \|\| type (6) \|.\|.\| spare (8) \| offset (16) \|
57			-------------------------------------------------------------------------------
58			\| \| \| \| \|
59			\| \| \|-> spare \| \|-> spare
60			\| \| \|
61			\| \|-> spare \|-> spare
62			\|
63			\|-> range flag
64
65	nigel	75	The upper/lower casing information is set only for characters that come in
66	nigel	87	pairs. The non-one-to-one mappings in the Unicode data are ignored.
67	nigel	75
68	nigel	87	When searching the data, proceed as follows:
69	nigel	75
70	nigel	87	(1) Set up for a binary chop search.
71	nigel	75
72	nigel	87	(2) If the top is not greater than the bottom, the character is not in the
73			table. Its type must therefore be "Cn" ("Undefined").
74	nigel	75
75	nigel	87	(3) Find the middle vector element.
76	nigel	75
77	nigel	87	(4) Extract the code point and compare. If equal, we are done.
78	nigel	75
79	nigel	87	(5) If the test character is smaller, set the top to the current point, and
80			goto (2).
81	nigel	75
82	nigel	87	(6) If the current entry defines a range, compute the last character by adding
83			the offset, and see if the test character is within the range. If it is,
84			we are done.
85	nigel	75
86	nigel	87	(7) Otherwise, set the bottom to one element past the current point and goto
87			(2).
88	nigel	75	*/
89
90	nigel	93	#endif /* _UCPINTERNAL_H */
91
92	nigel	87	/* End of ucpinternal.h */