Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
Overview
| Comment: | Add the ability to index files. |
|---|---|
| Downloads: | Tarball | ZIP archive |
| Timelines: | family | ancestors | descendants | both | search-using-fts4 |
| Files: | files | file ages | folders |
| SHA1: |
83c67816701851c5765abf9ed4f74cee |
| User & Date: | drh 2014-12-17 18:28:45.248 |
Context
|
2014-12-17
| ||
| 18:51 | Much faster implementation of the ftsearch_content() function for file content. check-in: bbcce76f26 user: drh tags: search-using-fts4 | |
| 18:28 | Add the ability to index files. check-in: 83c6781670 user: drh tags: search-using-fts4 | |
| 16:27 | Remove the code that implemented legacy "fossil search" command. FTS4 will be used moving forward. check-in: 389c70a56a user: drh tags: search-using-fts4 | |
Changes
Changes to src/blob.c.
| ︙ | ︙ | |||
199 200 201 202 203 204 205 206 207 208 209 210 211 212 |
/*
** Reset a blob to be an empty container.
*/
void blob_reset(Blob *pBlob){
blob_is_init(pBlob);
pBlob->xRealloc(pBlob, 0);
}
/*
** Return true if the blob has been zeroed - in other words if it contains
** no malloced memory. This only works reliably if the blob has been
** initialized - it can return a false negative on an uninitialized blob.
*/
| > > > > > > > > | 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 |
/*
** Reset a blob to be an empty container.
*/
void blob_reset(Blob *pBlob){
blob_is_init(pBlob);
pBlob->xRealloc(pBlob, 0);
}
/*
** Return true if the result of blob_str() is a string that can be
** freed using fossil_free.
*/
int blob_is_malloced(Blob *pBlob){
return pBlob && pBlob->aData && pBlob->xRealloc==blobReallocMalloc;
}
/*
** Return true if the blob has been zeroed - in other words if it contains
** no malloced memory. This only works reliably if the blob has been
** initialized - it can return a false negative on an uninitialized blob.
*/
|
| ︙ | ︙ |
Changes to src/db.c.
| ︙ | ︙ | |||
836 837 838 839 840 841 842 843 844 845 846 847 848 849 |
db, "symbolic_name_to_rid", 2, SQLITE_UTF8, 0, db_sym2rid_function,
0, 0
);
if( g.fSqlTrace ) sqlite3_trace(db, db_sql_trace, 0);
re_add_sql_func(db);
foci_register(db);
ftsearch_add_sql_func(db);
sqlite3_exec(db, "PRAGMA foreign_keys=OFF;", 0, 0, 0);
return db;
}
/*
** Detaches the zLabel database.
| > | 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 |
db, "symbolic_name_to_rid", 2, SQLITE_UTF8, 0, db_sym2rid_function,
0, 0
);
if( g.fSqlTrace ) sqlite3_trace(db, db_sql_trace, 0);
re_add_sql_func(db);
foci_register(db);
ftsearch_add_sql_func(db);
add_content_sql_commands(db);
sqlite3_exec(db, "PRAGMA foreign_keys=OFF;", 0, 0, 0);
return db;
}
/*
** Detaches the zLabel database.
|
| ︙ | ︙ |
Changes to src/ftsearch.c.
| ︙ | ︙ | |||
58 59 60 61 62 63 64 |
int id;
if( zDocId==0 ){
if( zDocType[0]==0 || zDocType[1]==0 ) return 0;
zDocId = zDocType + 2;
}
id = atoi(zDocId);
switch( zDocType[0] ){
| | > > > > > > > > > > > > | < | > > > > | > > > > > > | 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 |
int id;
if( zDocId==0 ){
if( zDocType[0]==0 || zDocType[1]==0 ) return 0;
zDocId = zDocType + 2;
}
id = atoi(zDocId);
switch( zDocType[0] ){
case 'c': { /* A check-in comment. zDocId is the RID */
zRes = db_text(0,
"SELECT coalesce(ecomment,comment) || char(10) ||"
" 'user: ' || coalesce(euser,user) || char(10) ||"
" 'branch: ' || coalesce((SELECT value FROM tagxref"
" WHERE tagid=%d AND tagtype>0"
" AND rid=%d),'trunk')"
" FROM event"
" WHERE event.objid=%d"
" AND event.type GLOB 'c*'",
TAG_BRANCH, id, id);
break;
}
case 'f': { /* A file with zDocId as the filename.fnid */
zRes = db_text(0,
"SELECT content(mlink.fid)"
" FROM filename, mlink, event"
" WHERE filename.fnid=%d"
" AND mlink.fnid=filename.fnid"
" AND event.objid=mlink.mid"
" ORDER BY event.mtime DESC LIMIT 1",
id
);
break;
}
default: {
/* No-op */
}
}
return zRes;
}
/* Return a human-readable description for the document described by
** the arguments.
**
** See ftsearch_content() for further information
*/
char *ftsearch_description(
const char *zDocType,
const char *zDocId,
int bLink /* Provide hyperlink in text if true */
){
char *zRes = 0; /* The result to be returned */
int id;
if( zDocId==0 ){
if( zDocType[0]==0 || zDocType[1]==0 ) return 0;
zDocId = zDocType + 2;
}
id = atoi(zDocId);
switch( zDocType[0] ){
case 'c': { /* A check-in comment. zDocId is the RID */
char *zUuid = db_text("","SELECT uuid FROM blob WHERE rid=%d", id);
zRes = mprintf("Check-in [%S]", zUuid);
fossil_free(zUuid);
break;
}
case 'f': { /* A file. zDocId is the FNID */
char *zName = db_text("","SELECT name FROM filename WHERE fnid=%d",id);
zRes = mprintf("File %s", zName);
fossil_free(zName);
break;
}
default: {
/* No-op */
}
}
return zRes;
}
|
| ︙ | ︙ | |||
123 124 125 126 127 128 129 |
char *zContent = 0;
char *zDesc = 0;
db_find_and_open_repository(0, 0);
verify_all_options();
if( g.argc!=3 ) usage("DOCUMENTCODE");
if( strlen(g.argv[2])>3 ){
zContent = ftsearch_content(g.argv[2],0);
| | | 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 |
char *zContent = 0;
char *zDesc = 0;
db_find_and_open_repository(0, 0);
verify_all_options();
if( g.argc!=3 ) usage("DOCUMENTCODE");
if( strlen(g.argv[2])>3 ){
zContent = ftsearch_content(g.argv[2],0);
zDesc = ftsearch_description(g.argv[2],0,0);
}
if( zDesc ){
fossil_print("Description: %s\n", zDesc);
fossil_free(zDesc);
}
if( zContent ){
fossil_print(
|
| ︙ | ︙ | |||
208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 |
db_end_transaction(0);
}
/*
** Completely rebuild the ftsearch indexes from scratch
*/
void ftsearch_rebuild_all(void){
db_begin_transaction();
ftsearch_disable_all();
/* The FTSSEARCHXREF table provides a mapping between the integer
** document-ids in FTS4 to the "document codes" that describe a
** referenced object
*/
db_multi_exec(
"CREATE TABLE %s.ftsearchxref(\n"
| > > > > > > > | 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 |
db_end_transaction(0);
}
/*
** Completely rebuild the ftsearch indexes from scratch
*/
void ftsearch_rebuild_all(void){
const char *zEnables;
db_begin_transaction();
ftsearch_disable_all();
zEnables = db_get("ftsearch-index-type", "cdeftw");
/* If none of the search categories are enabled, then do not
** bother constructing the search tables
*/
if( sqlite3_strglob("*[cdeftw]*", zEnables) ) return;
/* The FTSSEARCHXREF table provides a mapping between the integer
** document-ids in FTS4 to the "document codes" that describe a
** referenced object
*/
db_multi_exec(
"CREATE TABLE %s.ftsearchxref(\n"
|
| ︙ | ︙ | |||
240 241 242 243 244 245 246 |
/* This is the FTS4 table used for searching */
db_multi_exec(
"CREATE VIRTUAL TABLE %s.ftsearch"
" USING fts4(content='ftsearchbody',body);",
db_name("repository")
);
| > | | | | | | | | > > > > > > > > > > > > > > > > | 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 |
/* This is the FTS4 table used for searching */
db_multi_exec(
"CREATE VIRTUAL TABLE %s.ftsearch"
" USING fts4(content='ftsearchbody',body);",
db_name("repository")
);
if( strchr(zEnables, 'c')!=0 ){
/* Populate the FTSEARCHXREF table with references to all check-in
** comments currently in the event table
*/
db_multi_exec(
"INSERT INTO ftsearchxref(ftsid,mtime)"
" SELECT 'c-' || objid, mtime FROM event"
" WHERE type='ci';"
);
}
if( strchr(zEnables, 'f')!=0 ){
/* Populate the FTSEARCHXREF table with references to all files
*/
db_multi_exec(
"INSERT INTO ftsearchxref(ftsid,mtime)"
" SELECT 'f-' || filename.fnid, max(event.mtime)"
" FROM filename, mlink, event"
" WHERE mlink.fnid=filename.fnid"
" AND event.objid=mlink.mid"
" AND %s"
" GROUP BY 1",
glob_expr("filename.name", db_get("search-file-glob","*"))
);
}
/* Index every document mentioned in the FTSEARCHXREF table */
db_multi_exec(
"INSERT INTO ftsearch(docid,body)"
" SELECT docid, ftsearch_content(ftsid) FROM ftsearchxref;"
);
db_end_transaction(0);
|
| ︙ | ︙ | |||
270 271 272 273 274 275 276 | ** The "search" command locates resources that contain the given web-search ** style PATTERN. This only works if the repository has be configured to ** enable searching. ** ** The "search-config" is used to setup the search feature of the repository. ** Subcommands are: ** | | > > > > > > > > > > > > | < | < < < < > > > | 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 |
** The "search" command locates resources that contain the given web-search
** style PATTERN. This only works if the repository has be configured to
** enable searching.
**
** The "search-config" is used to setup the search feature of the repository.
** Subcommands are:
**
** fossil search-config doclist
**
** List all the documents currently indexed
**
** fossil search-config rebuild
**
** Completely rebuild the search index.
**
** fossil search-config reset
**
** Disable search and remove the search indexes from the repository.
**
** fossil search-config setting NAME ?VALUE?
**
** Set or query a search setting. NAMES are:
** file-glob Comma-separated list of GLOBs for file search
** ticket-expr SQL expression to render TICKET content
** ticketchng-expr SQL expression to render TICKETCHNG content
** index-type Zero or more characters from [cdeftw]
**
** The index-type determines what resources are indexed and available for
** searching. If the index-type is an empty string, the search is
** complete disabled. These are the valid index-types:
** c: check-in comments
** d: check-in difference marks
** e: event text
** f: file text (subject to the file-glob)
** t: ticket text (requires ticket-expr and ticketchng-expr)
** w: wiki pages
**
** It is necessary to run "fossil search-config rebuild" after making
** setting changes in order to reconstruct the search index
**
** fossil search-config status
**
** Report on the status of the search configuration.
*/
void ftsearch_cmd(void){
static const char *azSettings[] = {
"file-glob", "index-type", "ticket-expr", "ticketchng-expr"
};
const char *zSubCmd;
int nSubCmd;
db_find_and_open_repository(0, 0);
verify_all_options();
if( g.argc<3 ) usage("search PATTERN");
zSubCmd = g.argv[2];
nSubCmd = (int)strlen(zSubCmd);
|
| ︙ | ︙ | |||
326 327 328 329 330 331 332 |
#endif
if( !db_table_exists("repository","ftsearch") ){
fossil_fatal("search is disabled - see \"fossil help search\""
" for more information");
}
db_prepare(&q, "SELECT "
" snippet(ftsearch,%Q,%Q,'...'),"
| | > | | > | | > > > > > > > > > > | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | > > > > > > > > > | > > > > > > > > > | 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 |
#endif
if( !db_table_exists("repository","ftsearch") ){
fossil_fatal("search is disabled - see \"fossil help search\""
" for more information");
}
db_prepare(&q, "SELECT "
" snippet(ftsearch,%Q,%Q,'...'),"
" ftsearchxref.ftsid,"
" date(ftsearchxref.mtime)"
" FROM ftsearch, ftsearchxref"
" WHERE ftsearch.body MATCH %Q"
" AND ftsearchxref.docid=ftsearch.docid"
" ORDER BY ftsearchxref.mtime DESC LIMIT 50;",
zMark1, zMark2, zSubCmd);
while( db_step(&q)==SQLITE_ROW ){
const char *zSnippet = db_column_text(&q,0);
char *zDesc = ftsearch_description(db_column_text(&q,1),0,0);
const char *zDate = db_column_text(&q,2);
if( i++ > 0 ){
fossil_print("----------------------------------------------------\n");
}
fossil_print("%s (%s)\n%s\n", zDesc, zDate, zSnippet);
fossil_free(zDesc);
}
db_finalize(&q);
}else if( strncmp(zSubCmd, "doclist", nSubCmd)==0 ){
if( db_table_exists("repository","ftsearch") ){
Stmt q;
db_prepare(&q, "SELECT ftsid, date(mtime) FROM ftsearchxref"
" ORDER BY mtime DESC");
while( db_step(&q)==SQLITE_ROW ){
const char *zDate = db_column_text(&q,1);
const char *zFtsid = db_column_text(&q,0);
char *zDesc = ftsearch_description(zFtsid,0,0);
fossil_print("%s (%s)\n", zDesc, zDate);
fossil_free(zDesc);
}
db_finalize(&q);
}
}else if( strncmp(zSubCmd, "rebuild", nSubCmd)==0 ){
ftsearch_rebuild_all();
}else if( strncmp(zSubCmd, "reset", nSubCmd)==0 ){
ftsearch_disable_all();
}else if( strncmp(zSubCmd, "settings", nSubCmd)==0 ){
const char *zName = g.argv[3];
const char *zValue = g.argc>=5 ? g.argv[4] : 0;
char *zFullname;
int i;
if( g.argc<4 ) usage("setting NAME ?VALUE?");
for(i=0; i<count(azSettings); i++){
if( strcmp(zName, azSettings[i])==0 ) break;
}
if( i>=count(azSettings) ){
Blob x;
blob_init(&x,0,0);
for(i=0; i<count(azSettings); i++) blob_appendf(&x," %s", azSettings[i]);
fossil_fatal("unknown setting \"%s\" - should be one of:%s",
zName, blob_str(&x));
}
zFullname = mprintf("search-%s", zName);
if( zValue==0 ){
zValue = db_get(zFullname, 0);
}else{
db_set(zFullname, zValue, 0);
}
if( zValue==0 ){
fossil_print("%s is not defined\n", zName);
}else{
fossil_print("%s: %s\n", zName, zValue);
}
}else if( strncmp(zSubCmd, "status", nSubCmd)==0 ){
int i;
fossil_print("search settings:\n");
for(i=0; i<count(azSettings); i++){
char *zFullname = mprintf("search-%s", azSettings[i]);
char *zValue = db_get(zFullname, 0);
if( zValue==0 ){
fossil_print(" %s is undefined\n", azSettings[i]);
}else{
fossil_print(" %s: %s\n", azSettings[i], zValue);
}
fossil_free(zFullname);
}
if( db_table_exists("repository","ftsearchxref") ){
int n = db_int(0, "SELECT count(*) FROM ftsearchxref");
fossil_print("search is enabled with %d documents indexed\n", n);
}else{
fossil_print("search is disabled\n");
}
}
db_end_transaction(0);
}
|
Changes to src/sqlcmd.c.
| ︙ | ︙ | |||
38 39 40 41 42 43 44 |
int argc,
sqlite3_value **argv
){
int rid;
Blob cx;
const char *zName;
assert( argc==1 );
| > > > | | | | | > | < | 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 |
int argc,
sqlite3_value **argv
){
int rid;
Blob cx;
const char *zName;
assert( argc==1 );
if( sqlite3_value_type(argv[0])==SQLITE_INTEGER ){
rid = sqlite3_value_int(argv[0]);
}else{
zName = (const char*)sqlite3_value_text(argv[0]);
if( zName==0 ) return;
g.db = sqlite3_context_db_handle(context);
g.repositoryOpen = 1;
rid = name_to_rid(zName);
}
if( rid==0 ) return;
if( content_get(rid, &cx) ){
sqlite3_result_blob(context, blob_buffer(&cx), blob_size(&cx),
blob_is_malloced(&cx) ? fossil_free : SQLITE_TRANSIENT);
}
}
/*
** Implementation of the "compress(X)" SQL function. The input X is
** compressed using zLib and the output is returned.
*/
|
| ︙ | ︙ |