2 * Sherlock Gatherer: Data structures used in indices
4 * (c) 2001 Martin Mares <mj@ucw.cz>
9 #define MAX_WORD_LEN 64
14 WT_RESERVED, /* Reserved word type */
15 WT_TEXT, /* Ordinary text */
16 WT_EMPH, /* Emphasized text */
17 WT_SMALL, /* Small font */
18 WT_TITLE, /* Document title */
19 WT_SMALL_HEADING, /* Heading */
20 WT_BIG_HEADING, /* Larger heading */
21 WT_KEYWORD, /* Explicitly marked keyword */
22 WT_META, /* Various meta-information */
23 WT_ALT, /* Alternate texts for graphical elements */
24 WT_URL, /* Word extracted from document URL */
28 /* Descriptive names used for user output */
29 #define WORD_TYPE_USER_NAMES \
30 "reserved", "text", "emph", "small", "title", "hdr1", "hdr2", "keywd", \
31 "meta", "alt", "urlword", "type11", "type12", "type13", "type14", "type15"
33 /* Keywords for word type names */
34 #define WORD_TYPE_NAMES \
36 T(TEXT, 1 << WT_TEXT) \
37 T(EMPH, 1 << WT_EMPH) \
38 T(SMALL, 1 << WT_SMALL) \
39 T(TITLE, 1 << WT_TITLE) \
40 T(HDR, (1 << WT_SMALL_HEADING) | (1 << WT_BIG_HEADING)) \
41 T(HDR1, 1 << WT_SMALL_HEADING) \
42 T(HDR2, 1 << WT_BIG_HEADING) \
43 T(KEYWD, 1 << WT_KEYWORD) \
44 T(META, 1 << WT_META) \
46 T(URLWORD, 1 << WT_URL)
48 /* These types are not shown in document contexts */
49 #define WORD_TYPES_HIDDEN (1 << WT_URL)
54 ST_RESERVED, /* Reserved string type */
55 ST_URL, /* URL of the document */
56 ST_HOST, /* Host name */
57 ST_DOMAIN, /* Domain name */
58 ST_REF, /* URL reference */
59 ST_BACKREF, /* Back-reference (frame or redirect source) */
63 #define STRING_TYPE_USER_NAMES \
64 "URL", "host", "domain", "ref", "backref", "type5", "type6", "type7", \
65 "type8", "type9", "type10", "type11", "type12", "type13", "type14", "type15"
67 #define STRING_TYPE_NAMES \
69 T(HOST, 1 << ST_HOST) \
70 T(DOMAIN, 1 << ST_DOMAIN) \
72 T(BACKREF, 1 << ST_BACKREF)
74 #define STRING_TYPES_URL ((1 << ST_URL) | (1 << ST_REF) | (1 << ST_BACKREF))
75 /* These must be indexed in lowercase form */
76 #define STRING_TYPES_CASE_INSENSITIVE ((1 << ST_HOST) | (1 << ST_DOMAIN))
78 /* Index card attributes */
81 u32 card; /* Reference to card description (either oid or filepos) */
89 CARD_FLAG_EMPTY = 1, /* Empty document (redirect, robot file etc.) [scanner] */
90 CARD_FLAG_ACCENTED = 2, /* Document contains accented characters [scanner] */
91 CARD_FLAG_DUP = 4, /* Removed as a duplicate [merger] */
92 CARD_FLAG_MERGED = 8, /* Destination of a merge [merger] */
95 #define CARD_POS_SHIFT 5 /* Card positions are shifted this # of bytes to the right */
97 /* String fingerprints */
103 void fingerprint(byte *string, struct fingerprint *fp);
106 fp_hash(struct fingerprint *fp)
108 return (fp->hash[0] << 24) | (fp->hash[1] << 16) | (fp->hash[2] << 8) | fp->hash[3];
111 /* Reading of tagged text (Unicode values, tags mapped to 0x80000000 and higher) */
113 #define GET_TAGGED_CHAR(p,u) do { \
117 else if (u >= 0x80) \
126 else if (u >= 0xa0) \
128 ASSERT(*p >= 0x80 && *p <= 0xbf); \
129 u = 0x80010000 + ((u & 0x0f) << 6) + (*p++ & 0x3f); \