Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
Overview
Comment: | Add the Dline.nw field. Use it as a cache of the number of characters on a line with leading and trailing whitespace removed. |
---|---|
Downloads: | Tarball | ZIP archive | SQL archive |
Timelines: | family | ancestors | descendants | both | diff-improvement |
Files: | files | file ages | folders |
SHA3-256: |
f6112b93e9bbb7e37f24d83fb322c48f |
User & Date: | drh 2022-01-22 17:11:56 |
Context
2022-01-22
| ||
17:50 | Fix the computation of DLink.nw for the case of the -w (no whitespace) option for diff. ... (check-in: 44b6f128 user: drh tags: diff-improvement) | |
17:11 | Add the Dline.nw field. Use it as a cache of the number of characters on a line with leading and trailing whitespace removed. ... (check-in: f6112b93 user: drh tags: diff-improvement) | |
13:49 | Enhance the DLine.indent field so that it is always initialized to the number of space and control characters at the beginning of the line. ... (check-in: 2dad4158 user: drh tags: diff-improvement) | |
Changes
Changes to src/diff.c.
︙ | ︙ | |||
119 120 121 122 123 124 125 | ** of the line. If any line is longer than LENGTH_MASK characters, ** the file is considered binary. */ typedef struct DLine DLine; struct DLine { const char *z; /* The text of the line */ u64 h; /* Hash of the line */ | | > | 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 | ** of the line. If any line is longer than LENGTH_MASK characters, ** the file is considered binary. */ typedef struct DLine DLine; struct DLine { const char *z; /* The text of the line */ u64 h; /* Hash of the line */ unsigned short indent; /* Index of first non-space */ unsigned short n; /* number of bytes */ unsigned short nw; /* number of bytes without leading/trailing space */ unsigned int iNext; /* 1+(Index of next line with same the same hash) */ /* an array of DLine elements serves two purposes. The fields ** above are one per line of input text. But each entry is also ** a bucket in a hash table, as follows: */ unsigned int iHash; /* 1+(first entry in the hash chain) */ }; |
︙ | ︙ | |||
161 162 163 164 165 166 167 | int *aEdit; /* Array of copy/delete/insert triples */ int nEdit; /* Number of integers (3x num of triples) in aEdit[] */ int nEditAlloc; /* Space allocated for aEdit[] */ DLine *aFrom; /* File on left side of the diff */ int nFrom; /* Number of lines in aFrom[] */ DLine *aTo; /* File on right side of the diff */ int nTo; /* Number of lines in aTo[] */ | | > > > > > > > > > > > > > > > > > > > > > > | 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 | int *aEdit; /* Array of copy/delete/insert triples */ int nEdit; /* Number of integers (3x num of triples) in aEdit[] */ int nEditAlloc; /* Space allocated for aEdit[] */ DLine *aFrom; /* File on left side of the diff */ int nFrom; /* Number of lines in aFrom[] */ DLine *aTo; /* File on right side of the diff */ int nTo; /* Number of lines in aTo[] */ int (*xDiffer)(DLine*,DLine*); /* comparison function */ }; /* Fast isspace for use by diff */ static const char diffIsSpace[] = { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; #define diff_isspace(X) (diffIsSpace[(unsigned char)(X)]) /* ** Count the number of lines in the input string. Include the last line ** in the count even if it lacks the \n terminator. If an empty string ** is specified, the number of lines is zero. For the purposes of this ** function, a string is considered empty if it contains no characters ** -OR- it contains only NUL characters. |
︙ | ︙ | |||
244 245 246 247 248 249 250 | a[i].z = z; k = nn; if( diffFlags & DIFF_STRIP_EOLCR ){ if( k>0 && z[k-1]=='\r' ){ k--; } } a[i].n = k; if( diffFlags & DIFF_IGNORE_EOLWS ){ | | < < > > | > | | 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 | a[i].z = z; k = nn; if( diffFlags & DIFF_STRIP_EOLCR ){ if( k>0 && z[k-1]=='\r' ){ k--; } } a[i].n = k; if( diffFlags & DIFF_IGNORE_EOLWS ){ while( k>0 && diff_isspace(z[k-1]) ){ k--; } } if( (diffFlags & DIFF_IGNORE_ALLWS)==DIFF_IGNORE_ALLWS ){ int numws = 0; for(s=0; s<k && z[s]<=' '; s++){} a[i].indent = s; for(h=0, x=s; x<k; x++){ char c = z[x]; if( diff_isspace(c) ){ ++numws; }else{ h = (h^c)*9000000000000000041LL; } } k -= numws; a[i].nw = k - s; }else{ int k2 = k & ~0x7; u64 m; for(h=x=s=0; x<k2; x += 8){ memcpy(&m, z+x, 8); h = (h^m)*9000000000000000041LL; } m = 0; memcpy(&m, z+x, k-k2); h ^= m; } |
︙ | ︙ | |||
287 288 289 290 291 292 293 | *pnLine = nLine; return a; } /* ** Return zero if two DLine elements are identical. */ | | | | | | | 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 | *pnLine = nLine; return a; } /* ** Return zero if two DLine elements are identical. */ static int compare_dline(DLine *pA, DLine *pB){ if( pA->h!=pB->h ) return 1; return memcmp(pA->z,pB->z, pA->h&LENGTH_MASK); } /* ** Return zero if two DLine elements are identical, ignoring ** all whitespace. The indent field of pA/pB already points ** to the first non-space character in the string. */ static int compare_dline_ignore_allws(DLine *pA, DLine *pB){ int a = pA->indent, b = pB->indent; if( pA->h==pB->h ){ while( a<pA->n || b<pB->n ){ if( a<pA->n && b<pB->n && pA->z[a++] != pB->z[b++] ) return 1; while( a<pA->n && diff_isspace(pA->z[a])) ++a; while( b<pB->n && diff_isspace(pB->z[b])) ++b; } return pA->n-a != pB->n-b; } return 1; } /* ** Return true if the regular expression *pRe matches any of the ** N dlines */ static int re_dline_match( ReCompiled *pRe, /* The regular expression to be matched */ DLine *aDLine, /* First of N DLines to compare against */ int N /* Number of DLines to check */ ){ while( N-- ){ if( re_match(pRe, (const unsigned char *)aDLine->z, LENGTH(aDLine)) ){ return 1; } aDLine++; |
︙ | ︙ | |||
616 617 618 619 620 621 622 | } /* ** Return true if the string starts with n spaces */ static int allSpaces(const char *z, int n){ int i; | | | 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 | } /* ** Return true if the string starts with n spaces */ static int allSpaces(const char *z, int n){ int i; for(i=0; i<n && diff_isspace(z[i]); i++){} return i==n; } /* ** Try to improve the human-readability of the LineChange p. ** ** (1) If the first change span shows a change of indentation, try to |
︙ | ︙ | |||
687 688 689 690 691 692 693 | ** Given two lines of text, pFrom and pTo, compute a set of changes ** between those two lines, for enhanced display purposes. ** ** The result is written into the LineChange object given by the ** third parameter. */ static void oneLineChange( | | | | 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 | ** Given two lines of text, pFrom and pTo, compute a set of changes ** between those two lines, for enhanced display purposes. ** ** The result is written into the LineChange object given by the ** third parameter. */ static void oneLineChange( DLine *pLeft, /* Left line of the change */ DLine *pRight, /* Right line of the change */ LineChange *p /* OUTPUT: Write the results here */ ){ int nLeft; /* Length of left line in bytes */ int nRight; /* Length of right line in bytes */ int nShort; /* Shortest of left and right */ int nPrefix; /* Length of common prefix */ int nSuffix; /* Length of common suffix */ |
︙ | ︙ | |||
742 743 744 745 746 747 748 | int iBestVal = -1; int i; int nLong = nLeft<nRight ? nRight : nLeft; int nGap = nLong - nShort; for(i=nShort-nSuffix; i<=nPrefix; i++){ int iVal = 0; char c = zLeft[i]; | | | | 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 | int iBestVal = -1; int i; int nLong = nLeft<nRight ? nRight : nLeft; int nGap = nLong - nShort; for(i=nShort-nSuffix; i<=nPrefix; i++){ int iVal = 0; char c = zLeft[i]; if( diff_isspace(c) ){ iVal += 5; }else if( !fossil_isalnum(c) ){ iVal += 2; } c = zLeft[i+nGap-1]; if( diff_isspace(c) ){ iVal += 5; }else if( !fossil_isalnum(c) ){ iVal += 2; } if( iVal>iBestVal ){ iBestVal = iVal; iBest = i; |
︙ | ︙ | |||
883 884 885 886 887 888 889 | ** ** To subclass, create an instance of the DiffBuilder object and fill ** in appropriate method implementations. */ typedef struct DiffBuilder DiffBuilder; struct DiffBuilder { void (*xSkip)(DiffBuilder*, unsigned int, int); | | | | | | | 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 | ** ** To subclass, create an instance of the DiffBuilder object and fill ** in appropriate method implementations. */ typedef struct DiffBuilder DiffBuilder; struct DiffBuilder { void (*xSkip)(DiffBuilder*, unsigned int, int); void (*xCommon)(DiffBuilder*,DLine*); void (*xInsert)(DiffBuilder*,DLine*); void (*xDelete)(DiffBuilder*,DLine*); void (*xReplace)(DiffBuilder*,DLine*, DLine*); void (*xEdit)(DiffBuilder*,DLine*,DLine*); void (*xEnd)(DiffBuilder*); unsigned int lnLeft; /* Lines seen on the left (delete) side */ unsigned int lnRight; /* Lines seen on the right (insert) side */ unsigned int nPending; /* Number of pending lines */ int eState; /* State of the output */ int width; /* Display width */ Blob *pOut; /* Output blob */ |
︙ | ︙ | |||
913 914 915 916 917 918 919 | static void dfdebugSkip(DiffBuilder *p, unsigned int n, int isFinal){ blob_appendf(p->pOut, "SKIP %d (%d..%d left and %d..%d right)%s\n", n, p->lnLeft+1, p->lnLeft+n, p->lnRight+1, p->lnRight+n, isFinal ? " FINAL" : ""); p->lnLeft += n; p->lnRight += n; } | | | | | | | 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 | static void dfdebugSkip(DiffBuilder *p, unsigned int n, int isFinal){ blob_appendf(p->pOut, "SKIP %d (%d..%d left and %d..%d right)%s\n", n, p->lnLeft+1, p->lnLeft+n, p->lnRight+1, p->lnRight+n, isFinal ? " FINAL" : ""); p->lnLeft += n; p->lnRight += n; } static void dfdebugCommon(DiffBuilder *p, DLine *pLine){ p->lnLeft++; p->lnRight++; blob_appendf(p->pOut, "COMMON %8u %8u %.*s\n", p->lnLeft, p->lnRight, (int)pLine->n, pLine->z); } static void dfdebugInsert(DiffBuilder *p, DLine *pLine){ p->lnRight++; blob_appendf(p->pOut, "INSERT %8d %.*s\n", p->lnRight, (int)pLine->n, pLine->z); } static void dfdebugDelete(DiffBuilder *p, DLine *pLine){ p->lnLeft++; blob_appendf(p->pOut, "DELETE %8u %.*s\n", p->lnLeft, (int)pLine->n, pLine->z); } static void dfdebugReplace(DiffBuilder *p, DLine *pX, DLine *pY){ p->lnLeft++; p->lnRight++; blob_appendf(p->pOut, "REPLACE %8u %.*s\n", p->lnLeft, (int)pX->n, pX->z); blob_appendf(p->pOut, " %8u %.*s\n", p->lnRight, (int)pY->n, pY->z); } static void dfdebugEdit(DiffBuilder *p, DLine *pX, DLine *pY){ int i, j; int x; LineChange chng; p->lnLeft++; p->lnRight++; blob_appendf(p->pOut, "EDIT %8u %.*s\n", p->lnLeft, (int)pX->n, pX->z); |
︙ | ︙ | |||
1029 1030 1031 1032 1033 1034 1035 | ** will be an empty string. (1) might be empty for the first triple if ** the line begins with an edit. After all triples, there might be one ** additional string which is a common suffix. */ static void dftclSkip(DiffBuilder *p, unsigned int n, int isFinal){ blob_appendf(p->pOut, "SKIP %u\n", n); } | | | | | | | 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 | ** will be an empty string. (1) might be empty for the first triple if ** the line begins with an edit. After all triples, there might be one ** additional string which is a common suffix. */ static void dftclSkip(DiffBuilder *p, unsigned int n, int isFinal){ blob_appendf(p->pOut, "SKIP %u\n", n); } static void dftclCommon(DiffBuilder *p, DLine *pLine){ blob_appendf(p->pOut, "COM "); blob_append_tcl_literal(p->pOut, pLine->z, pLine->n); blob_append_char(p->pOut, '\n'); } static void dftclInsert(DiffBuilder *p, DLine *pLine){ blob_append(p->pOut, "INS ", -1); blob_append_tcl_literal(p->pOut, pLine->z, pLine->n); blob_append_char(p->pOut, '\n'); } static void dftclDelete(DiffBuilder *p, DLine *pLine){ blob_append(p->pOut, "DEL ", -1); blob_append_tcl_literal(p->pOut, pLine->z, pLine->n); blob_append_char(p->pOut, '\n'); } static void dftclReplace(DiffBuilder *p, DLine *pX, DLine *pY){ blob_append(p->pOut, "EDIT \"\" ", -1); blob_append_tcl_literal(p->pOut, pX->z, pX->n); blob_append_char(p->pOut, ' '); blob_append_tcl_literal(p->pOut, pY->z, pY->n); blob_append_char(p->pOut, '\n'); } static void dftclEdit(DiffBuilder *p, DLine *pX, DLine *pY){ int i, x; LineChange chng; blob_append(p->pOut, "EDIT", 4); oneLineChange(pX, pY, &chng); for(i=x=0; i<chng.n; i++){ blob_append_char(p->pOut, ' '); blob_append_tcl_literal(p->pOut, pX->z + x, chng.a[i].iStart1 - x); |
︙ | ︙ | |||
1113 1114 1115 1116 1117 1118 1119 | ** represent common-text, left-text, and right-text. The last string ** in SUBARRAY is the common-suffix. Any string can be empty if it does ** not apply. */ static void dfjsonSkip(DiffBuilder *p, unsigned int n, int isFinal){ blob_appendf(p->pOut, "1,%u,\n", n); } | | | | | | | 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 | ** represent common-text, left-text, and right-text. The last string ** in SUBARRAY is the common-suffix. Any string can be empty if it does ** not apply. */ static void dfjsonSkip(DiffBuilder *p, unsigned int n, int isFinal){ blob_appendf(p->pOut, "1,%u,\n", n); } static void dfjsonCommon(DiffBuilder *p, DLine *pLine){ blob_append(p->pOut, "2,",2); blob_append_json_literal(p->pOut, pLine->z, (int)pLine->n); blob_append(p->pOut, ",\n",2); } static void dfjsonInsert(DiffBuilder *p, DLine *pLine){ blob_append(p->pOut, "3,",2); blob_append_json_literal(p->pOut, pLine->z, (int)pLine->n); blob_append(p->pOut, ",\n",2); } static void dfjsonDelete(DiffBuilder *p, DLine *pLine){ blob_append(p->pOut, "4,",2); blob_append_json_literal(p->pOut, pLine->z, (int)pLine->n); blob_append(p->pOut, ",\n",2); } static void dfjsonReplace(DiffBuilder *p, DLine *pX, DLine *pY){ blob_append(p->pOut, "5,[\"\",",-1); blob_append_json_literal(p->pOut, pX->z, (int)pX->n); blob_append(p->pOut, ",",1); blob_append_json_literal(p->pOut, pY->z, (int)pY->n); blob_append(p->pOut, ",\"\"],\n",-1); } static void dfjsonEdit(DiffBuilder *p, DLine *pX, DLine *pY){ int i, x; LineChange chng; blob_append(p->pOut, "5,[", 3); oneLineChange(pX, pY, &chng); for(i=x=0; i<chng.n; i++){ if(i>0){ blob_append_char(p->pOut, ','); |
︙ | ︙ | |||
1268 1269 1270 1271 1272 1273 1274 | blob_append(p->pOut, "<tr>", 4); } blob_append(p->pOut, "<td class=\"diffln difflne\">" "︙</td><td></td><td></td></tr>\n", -1); p->lnLeft += n; p->lnRight += n; } | | | | | | 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 | blob_append(p->pOut, "<tr>", 4); } blob_append(p->pOut, "<td class=\"diffln difflne\">" "︙</td><td></td><td></td></tr>\n", -1); p->lnLeft += n; p->lnRight += n; } static void dfunifiedCommon(DiffBuilder *p, DLine *pLine){ dfunifiedStartRow(p); dfunifiedFinishDelete(p); dfunifiedFinishInsert(p); p->lnLeft++; p->lnRight++; blob_appendf(p->pOut,"%d\n", p->lnLeft); blob_appendf(&p->aCol[0],"%d\n",p->lnRight); blob_append_char(&p->aCol[1], '\n'); htmlize_to_blob(&p->aCol[2], pLine->z, (int)pLine->n); blob_append_char(&p->aCol[2], '\n'); } static void dfunifiedInsert(DiffBuilder *p, DLine *pLine){ dfunifiedStartRow(p); p->lnRight++; blob_appendf(&p->aCol[3],"%d\n", p->lnRight); blob_append(&p->aCol[4], "<ins>", 5); htmlize_to_blob(&p->aCol[4], pLine->z, (int)pLine->n); blob_append(&p->aCol[4], "</ins>\n", 7); p->nPending++; } static void dfunifiedDelete(DiffBuilder *p, DLine *pLine){ dfunifiedStartRow(p); dfunifiedFinishInsert(p); if( p->eState==0 ){ dfunifiedFinishInsert(p); blob_append(p->pOut, "<del>", 5); blob_append(&p->aCol[2], "<del>", 5); p->eState = 1; } p->lnLeft++; blob_appendf(p->pOut,"%d\n", p->lnLeft); blob_append_char(&p->aCol[0],'\n'); blob_append(&p->aCol[1],"-\n",2); blob_append(&p->aCol[2], "<del>", 5); htmlize_to_blob(&p->aCol[2], pLine->z, (int)pLine->n); blob_append(&p->aCol[2], "</del>\n", 7); } static void dfunifiedReplace(DiffBuilder *p, DLine *pX, DLine *pY){ dfunifiedStartRow(p); if( p->eState==0 ){ dfunifiedFinishInsert(p); blob_append(p->pOut, "<del>", 5); blob_append(&p->aCol[2], "<del>", 5); p->eState = 1; } |
︙ | ︙ | |||
1329 1330 1331 1332 1333 1334 1335 | blob_appendf(&p->aCol[3],"%d\n", p->lnRight); htmlize_to_blob(&p->aCol[4], pY->z, pY->n); blob_append_char(&p->aCol[4], '\n'); p->nPending++; } | | | 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 | blob_appendf(&p->aCol[3],"%d\n", p->lnRight); htmlize_to_blob(&p->aCol[4], pY->z, pY->n); blob_append_char(&p->aCol[4], '\n'); p->nPending++; } static void dfunifiedEdit(DiffBuilder *p, DLine *pX, DLine *pY){ int i; int x; LineChange chng; oneLineChange(pX, pY, &chng); dfunifiedStartRow(p); if( p->eState==0 ){ dfunifiedFinishInsert(p); |
︙ | ︙ | |||
1498 1499 1500 1501 1502 1503 1504 | "<td class=\"diffln difflnl difflne\">︙</td>" "<td></td><td></td>" "<td class=\"diffln difflnr difflne\">︙</td>" "<td/td></tr>\n", -1); p->lnLeft += n; p->lnRight += n; } | | | | | | | 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 | "<td class=\"diffln difflnl difflne\">︙</td>" "<td></td><td></td>" "<td class=\"diffln difflnr difflne\">︙</td>" "<td/td></tr>\n", -1); p->lnLeft += n; p->lnRight += n; } static void dfsplitCommon(DiffBuilder *p, DLine *pLine){ dfsplitStartRow(p); dfsplitChangeState(p, 0); p->lnLeft++; p->lnRight++; blob_appendf(p->pOut,"%d\n", p->lnLeft); htmlize_to_blob(&p->aCol[0], pLine->z, (int)pLine->n); blob_append_char(&p->aCol[0], '\n'); blob_append_char(&p->aCol[1], '\n'); blob_appendf(&p->aCol[2],"%d\n",p->lnRight); htmlize_to_blob(&p->aCol[3], pLine->z, (int)pLine->n); blob_append_char(&p->aCol[3], '\n'); } static void dfsplitInsert(DiffBuilder *p, DLine *pLine){ dfsplitStartRow(p); dfsplitChangeState(p, 2); p->lnRight++; blob_append_char(p->pOut, '\n'); blob_append_char(&p->aCol[0], '\n'); blob_append(&p->aCol[1], ">\n", -1); blob_appendf(&p->aCol[2],"%d\n", p->lnRight); blob_append(&p->aCol[3], "<ins>", 5); htmlize_to_blob(&p->aCol[3], pLine->z, (int)pLine->n); blob_append(&p->aCol[3], "</ins>\n", 7); } static void dfsplitDelete(DiffBuilder *p, DLine *pLine){ dfsplitStartRow(p); dfsplitChangeState(p, 1); p->lnLeft++; blob_appendf(p->pOut,"%d\n", p->lnLeft); blob_append(&p->aCol[0], "<del>", 5); htmlize_to_blob(&p->aCol[0], pLine->z, (int)pLine->n); blob_append(&p->aCol[0], "</del>\n", 7); blob_append(&p->aCol[1], "<\n", -1); blob_append_char(&p->aCol[2],'\n'); blob_append_char(&p->aCol[3],'\n'); } static void dfsplitReplace(DiffBuilder *p, DLine *pX, DLine *pY){ dfsplitStartRow(p); dfsplitChangeState(p, 3); p->lnLeft++; p->lnRight++; blob_appendf(p->pOut,"%d\n", p->lnLeft); htmlize_to_blob(&p->aCol[0], pX->z, pX->n); blob_append_char(&p->aCol[0], '\n'); blob_append(&p->aCol[1], "|\n", 2); blob_appendf(&p->aCol[2],"%d\n", p->lnRight); htmlize_to_blob(&p->aCol[3], pY->z, pY->n); blob_append_char(&p->aCol[3], '\n'); } static void dfsplitEdit(DiffBuilder *p, DLine *pX, DLine *pY){ int i; int x; LineChange chng; oneLineChange(pX, pY, &chng); dfsplitStartRow(p); dfsplitChangeState(p, 3); p->lnLeft++; |
︙ | ︙ | |||
1653 1654 1655 1656 1657 1658 1659 | /* ** Append at least iMin characters (not bytes) and at most iMax characters ** from pX onto the into of p. ** ** This comment contains multibyte unicode characters (ü, Æ, ð) in order ** to test the ability of the diff code to handle such characters. */ | | | | | | | 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 | /* ** Append at least iMin characters (not bytes) and at most iMax characters ** from pX onto the into of p. ** ** This comment contains multibyte unicode characters (ü, Æ, ð) in order ** to test the ability of the diff code to handle such characters. */ static void sbs_append_chars(Blob *p, int iMin, int iMax, DLine *pX){ int i; const char *z = pX->z; for(i=0; i<iMax && i<pX->n; i++){ char c = z[i]; blob_append_char(p, c); if( (c&0xc0)==0x80 ){ iMin++; iMax++; } } while( i<iMin ){ blob_append_char(p, ' '); i++; } } static void dfsbsCommon(DiffBuilder *p, DLine *pLine){ p->lnLeft++; p->lnRight++; blob_appendf(p->pOut,"%6u ", p->lnLeft); sbs_append_chars(p->pOut, p->width, p->width, pLine); blob_appendf(p->pOut," %6u ", p->lnRight); sbs_append_chars(p->pOut, 0, p->width, pLine); blob_append_char(p->pOut, '\n'); } static void dfsbsInsert(DiffBuilder *p, DLine *pLine){ p->lnRight++; blob_appendf(p->pOut,"%6s %*s > %6u ", "", p->width, "", p->lnRight); sbs_append_chars(p->pOut, 0, p->width, pLine); blob_append_char(p->pOut, '\n'); } static void dfsbsDelete(DiffBuilder *p, DLine *pLine){ p->lnLeft++; blob_appendf(p->pOut,"%6u ", p->lnLeft); sbs_append_chars(p->pOut, p->width, p->width, pLine); blob_append(p->pOut," <\n", 3); } static void dfsbsEdit(DiffBuilder *p, DLine *pX, DLine *pY){ p->lnLeft++; p->lnRight++; blob_appendf(p->pOut,"%6u ", p->lnLeft); sbs_append_chars(p->pOut, p->width, p->width, pX); blob_appendf(p->pOut, " | %6u ", p->lnRight); sbs_append_chars(p->pOut, 0, p->width, pY); blob_append_char(p->pOut, '\n'); |
︙ | ︙ | |||
1730 1731 1732 1733 1734 1735 1736 | ** (1) Remove leading and trailing whitespace. ** (2) Truncate both strings to at most 250 characters ** (3) If the two strings have a common prefix, measure that prefix ** (4) Find the length of the longest common subsequence that is ** at least 150% longer than the common prefix. ** (5) Longer common subsequences yield lower scores. */ | | | > > > > > > > > > | > > > > > > | | | < | 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 | ** (1) Remove leading and trailing whitespace. ** (2) Truncate both strings to at most 250 characters ** (3) If the two strings have a common prefix, measure that prefix ** (4) Find the length of the longest common subsequence that is ** at least 150% longer than the common prefix. ** (5) Longer common subsequences yield lower scores. */ static int match_dline(DLine *pA, DLine *pB){ const char *zA; /* Left string */ const char *zB; /* right string */ int nA; /* Bytes in zA[] */ int nB; /* Bytes in zB[] */ int nMin; int nPrefix; int avg; /* Average length of A and B */ int i, j, k; /* Loop counters */ int best = 0; /* Longest match found so far */ int score; /* Final score. 0..100 */ unsigned char c; /* Character being examined */ unsigned char aFirst[256]; /* aFirst[X] = index in zB[] of first char X */ unsigned char aNext[252]; /* aNext[i] = index in zB[] of next zB[i] char */ zA = pA->z; if( pA->nw==0 && pA->n ){ for(i=0; i<pA->n && diff_isspace(zA[i]); i++){} pA->indent = i; for(j=pA->n-1; j>i && diff_isspace(zA[j]); j--){} pA->nw = j - i + 1; } zA += pA->indent; nA = pA->nw; zB = pB->z; if( pB->nw==0 && pB->n ){ for(i=0; i<pB->n && diff_isspace(zB[i]); i++){} pB->indent = i; for(j=pB->n-1; j>i && diff_isspace(zB[j]); j--){} pB->nw = j - i + 1; } zB += pB->indent; nB = pB->nw; if( nA>250 ) nA = 250; if( nB>250 ) nB = 250; avg = (nA+nB)/2; if( avg==0 ) return 0; nMin = nA; if( nB<nMin ) nMin = nB; if( nMin==0 ) return 68; |
︙ | ︙ | |||
1845 1846 1847 1848 1849 1850 1851 | ** Algorithm: Wagner's minimum edit-distance algorithm, modified by ** adding a cost to each match based on how well the two rows match ** each other. Insertion and deletion costs are 50. Match costs ** are between 0 and 100 where 0 is a perfect match 100 is a complete ** mismatch. */ static unsigned char *diffBlockAlignment( | | | | 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 | ** Algorithm: Wagner's minimum edit-distance algorithm, modified by ** adding a cost to each match based on how well the two rows match ** each other. Insertion and deletion costs are 50. Match costs ** are between 0 and 100 where 0 is a perfect match 100 is a complete ** mismatch. */ static unsigned char *diffBlockAlignment( DLine *aLeft, int nLeft, /* Text on the left */ DLine *aRight, int nRight, /* Text on the right */ DiffConfig *pCfg, /* Configuration options */ int *pNResult /* OUTPUT: Bytes of result */ ){ int i, j, k; /* Loop counters */ int *a; /* One row of the Wagner matrix */ int *pToFree; /* Space that needs to be freed */ unsigned char *aM; /* Wagner result matrix */ |
︙ | ︙ | |||
1876 1877 1878 1879 1880 1881 1882 | /* For large alignments, use a divide and conquer algorithm that is ** O(NlogN). The result is not as precise, but this whole thing is an ** approximation anyhow, and the faster response time is an acceptable ** trade-off for reduced precision. */ if( nLeft*nRight>DIFF_ALIGN_MX && (pCfg->diffFlags & DIFF_SLOW_SBS)==0 ){ | | | | 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 | /* For large alignments, use a divide and conquer algorithm that is ** O(NlogN). The result is not as precise, but this whole thing is an ** approximation anyhow, and the faster response time is an acceptable ** trade-off for reduced precision. */ if( nLeft*nRight>DIFF_ALIGN_MX && (pCfg->diffFlags & DIFF_SLOW_SBS)==0 ){ DLine *aSmall; /* The smaller of aLeft and aRight */ DLine *aBig; /* The larger of aLeft and aRight */ int nSmall, nBig; /* Size of aSmall and aBig. nSmall<=nBig */ int iDivSmall, iDivBig; /* Divider point for aSmall and aBig */ int iDivLeft, iDivRight; /* Divider point for aLeft and aRight */ unsigned char *a1, *a2; /* Results of the alignments on two halves */ int n1, n2; /* Number of entries in a1 and a2 */ int score, bestScore; /* Score and best score seen so far */ if( nLeft>nRight ){ |
︙ | ︙ | |||
2021 2022 2023 2024 2025 2026 2027 | ** Format a diff using a DiffBuilder object */ static void formatDiff( DContext *p, /* The computed diff */ DiffConfig *pCfg, /* Configuration options */ DiffBuilder *pBuilder /* The formatter object */ ){ | | | | 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 | ** Format a diff using a DiffBuilder object */ static void formatDiff( DContext *p, /* The computed diff */ DiffConfig *pCfg, /* Configuration options */ DiffBuilder *pBuilder /* The formatter object */ ){ DLine *A; /* Left side of the diff */ DLine *B; /* Right side of the diff */ unsigned int a = 0; /* Index of next line in A[] */ unsigned int b = 0; /* Index of next line in B[] */ const int *R; /* Array of COPY/DELETE/INSERT triples */ unsigned int r; /* Index into R[] */ unsigned int nr; /* Number of COPY/DELETE/INSERT triples to process */ unsigned int mxr; /* Maximum value for r */ unsigned int na, nb; /* Number of lines shown from A and B */ |
︙ | ︙ |