| 1 | /* |
| 2 | * LibXDiff by Davide Libenzi ( File Differential Library ) |
| 3 | * Copyright (C) 2003 Davide Libenzi |
| 4 | * |
| 5 | * This library is free software; you can redistribute it and/or |
| 6 | * modify it under the terms of the GNU Lesser General Public |
| 7 | * License as published by the Free Software Foundation; either |
| 8 | * version 2.1 of the License, or (at your option) any later version. |
| 9 | * |
| 10 | * This library is distributed in the hope that it will be useful, |
| 11 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 13 | * Lesser General Public License for more details. |
| 14 | * |
| 15 | * You should have received a copy of the GNU Lesser General Public |
| 16 | * License along with this library; if not, see |
| 17 | * <http://www.gnu.org/licenses/>. |
| 18 | * |
| 19 | * Davide Libenzi <davidel@xmailserver.org> |
| 20 | * |
| 21 | */ |
| 22 | |
| 23 | #include "xinclude.h" |
| 24 | |
| 25 | |
| 26 | static int xdl_emit_record(xdfile_t *xdf, long ri, char const *pre, xdemitcb_t *ecb) |
| 27 | { |
| 28 | xrecord_t *rec = &xdf->recs[ri]; |
| 29 | |
| 30 | if (xdl_emit_diffrec((char const *)rec->ptr, (long)rec->size, pre, strlen(pre), ecb) < 0) |
| 31 | return -1; |
| 32 | |
| 33 | return 0; |
| 34 | } |
| 35 | |
| 36 | static long saturating_add(long a, long b) |
| 37 | { |
| 38 | return signed_add_overflows(a, b) ? LONG_MAX : a + b; |
| 39 | } |
| 40 | |
| 41 | /* |
| 42 | * Starting at the passed change atom, find the latest change atom to be included |
| 43 | * inside the differential hunk according to the specified configuration. |
| 44 | * Also advance xscr if the first changes must be discarded. |
| 45 | */ |
| 46 | xdchange_t *xdl_get_hunk(xdchange_t **xscr, xdemitconf_t const *xecfg) |
| 47 | { |
| 48 | xdchange_t *xch, *xchp, *lxch; |
| 49 | long max_common; |
| 50 | long max_ignorable; |
| 51 | long ignored = 0; /* number of ignored blank lines */ |
| 52 | |
| 53 | if (xecfg->ctxlen < 0) |
| 54 | BUG("negative context length: %ld", xecfg->ctxlen); |
| 55 | if (xecfg->interhunkctxlen < 0) |
| 56 | BUG("negative inter-hunk context length: %ld", xecfg->interhunkctxlen); |
| 57 | |
| 58 | max_common = saturating_add(saturating_add(xecfg->ctxlen, |
| 59 | xecfg->ctxlen), |
| 60 | xecfg->interhunkctxlen); |
| 61 | max_ignorable = xecfg->ctxlen; |
| 62 | |
| 63 | /* remove ignorable changes that are too far before other changes */ |
| 64 | for (xchp = *xscr; xchp && xchp->ignore; xchp = xchp->next) { |
| 65 | xch = xchp->next; |
| 66 | |
| 67 | if (xch == NULL || |
| 68 | xch->i1 - (xchp->i1 + xchp->chg1) >= max_ignorable) |
| 69 | *xscr = xch; |
| 70 | } |
| 71 | |
| 72 | if (!*xscr) |
| 73 | return NULL; |
| 74 | |
| 75 | lxch = *xscr; |
| 76 | |
| 77 | for (xchp = *xscr, xch = xchp->next; xch; xchp = xch, xch = xch->next) { |
| 78 | long distance = xch->i1 - (xchp->i1 + xchp->chg1); |
| 79 | if (distance > max_common) |
| 80 | break; |
| 81 | |
| 82 | if (distance < max_ignorable && (!xch->ignore || lxch == xchp)) { |
| 83 | lxch = xch; |
| 84 | ignored = 0; |
| 85 | } else if (distance < max_ignorable && xch->ignore) { |
| 86 | ignored += xch->chg2; |
| 87 | } else if (lxch != xchp && |
| 88 | xch->i1 + ignored - (lxch->i1 + lxch->chg1) > max_common) { |
| 89 | break; |
| 90 | } else if (!xch->ignore) { |
| 91 | lxch = xch; |
| 92 | ignored = 0; |
| 93 | } else { |
| 94 | ignored += xch->chg2; |
| 95 | } |
| 96 | } |
| 97 | |
| 98 | return lxch; |
| 99 | } |
| 100 | |
| 101 | |
| 102 | static long def_ff(const char *rec, long len, char *buf, long sz) |
| 103 | { |
| 104 | if (len > 0 && |
| 105 | (isalpha((unsigned char)*rec) || /* identifier? */ |
| 106 | *rec == '_' || /* also identifier? */ |
| 107 | *rec == '$')) { /* identifiers from VMS and other esoterico */ |
| 108 | if (len > sz) |
| 109 | len = sz; |
| 110 | while (0 < len && isspace((unsigned char)rec[len - 1])) |
| 111 | len--; |
| 112 | memcpy(buf, rec, len); |
| 113 | return len; |
| 114 | } |
| 115 | return -1; |
| 116 | } |
| 117 | |
| 118 | static long match_func_rec(xdfile_t *xdf, xdemitconf_t const *xecfg, long ri, |
| 119 | char *buf, long sz) |
| 120 | { |
| 121 | xrecord_t *rec = &xdf->recs[ri]; |
| 122 | |
| 123 | if (!xecfg->find_func) |
| 124 | return def_ff((const char *)rec->ptr, (long)rec->size, buf, sz); |
| 125 | return xecfg->find_func((const char *)rec->ptr, (long)rec->size, buf, sz, xecfg->find_func_priv); |
| 126 | } |
| 127 | |
| 128 | static int is_func_rec(xdfile_t *xdf, xdemitconf_t const *xecfg, long ri) |
| 129 | { |
| 130 | char dummy[1]; |
| 131 | return match_func_rec(xdf, xecfg, ri, dummy, sizeof(dummy)) >= 0; |
| 132 | } |
| 133 | |
| 134 | struct func_line { |
| 135 | long len; |
| 136 | char buf[80]; |
| 137 | }; |
| 138 | |
| 139 | static long get_func_line(xdfenv_t *xe, xdemitconf_t const *xecfg, |
| 140 | struct func_line *func_line, long start, long limit) |
| 141 | { |
| 142 | long l, size, step = (start > limit) ? -1 : 1; |
| 143 | char *buf, dummy[1]; |
| 144 | |
| 145 | buf = func_line ? func_line->buf : dummy; |
| 146 | size = func_line ? sizeof(func_line->buf) : sizeof(dummy); |
| 147 | |
| 148 | for (l = start; l != limit && 0 <= l && l < (long)xe->xdf1.nrec; l += step) { |
| 149 | long len = match_func_rec(&xe->xdf1, xecfg, l, buf, size); |
| 150 | if (len >= 0) { |
| 151 | if (func_line) |
| 152 | func_line->len = len; |
| 153 | return l; |
| 154 | } |
| 155 | } |
| 156 | return -1; |
| 157 | } |
| 158 | |
| 159 | static int is_empty_rec(xdfile_t *xdf, long ri) |
| 160 | { |
| 161 | xrecord_t *rec = &xdf->recs[ri]; |
| 162 | size_t i = 0; |
| 163 | |
| 164 | for (; i < rec->size && XDL_ISSPACE(rec->ptr[i]); i++); |
| 165 | |
| 166 | return i == rec->size; |
| 167 | } |
| 168 | |
| 169 | int xdl_emit_diff(xdfenv_t *xe, xdchange_t *xscr, xdemitcb_t *ecb, |
| 170 | xdemitconf_t const *xecfg) { |
| 171 | long s1, s2, e1, e2, lctx; |
| 172 | xdchange_t *xch, *xche; |
| 173 | long funclineprev = -1; |
| 174 | struct func_line func_line = { 0 }; |
| 175 | |
| 176 | for (xch = xscr; xch; xch = xche->next) { |
| 177 | xdchange_t *xchp = xch; |
| 178 | xche = xdl_get_hunk(&xch, xecfg); |
| 179 | if (!xch) |
| 180 | break; |
| 181 | |
| 182 | pre_context_calculation: |
| 183 | s1 = XDL_MAX(xch->i1 - xecfg->ctxlen, 0); |
| 184 | s2 = XDL_MAX(xch->i2 - xecfg->ctxlen, 0); |
| 185 | |
| 186 | if (xecfg->flags & XDL_EMIT_FUNCCONTEXT) { |
| 187 | long fs1, i1 = xch->i1; |
| 188 | |
| 189 | /* Appended chunk? */ |
| 190 | if (i1 >= (long)xe->xdf1.nrec) { |
| 191 | long i2 = xch->i2; |
| 192 | |
| 193 | /* |
| 194 | * We don't need additional context if |
| 195 | * a whole function was added. |
| 196 | */ |
| 197 | while (i2 < (long)xe->xdf2.nrec) { |
| 198 | if (is_func_rec(&xe->xdf2, xecfg, i2)) |
| 199 | goto post_context_calculation; |
| 200 | i2++; |
| 201 | } |
| 202 | |
| 203 | /* |
| 204 | * Otherwise get more context from the |
| 205 | * pre-image. |
| 206 | */ |
| 207 | i1 = (long)xe->xdf1.nrec - 1; |
| 208 | } |
| 209 | |
| 210 | fs1 = get_func_line(xe, xecfg, NULL, i1, -1); |
| 211 | while (fs1 > 0 && !is_empty_rec(&xe->xdf1, fs1 - 1) && |
| 212 | !is_func_rec(&xe->xdf1, xecfg, fs1 - 1)) |
| 213 | fs1--; |
| 214 | if (fs1 < 0) |
| 215 | fs1 = 0; |
| 216 | if (fs1 < s1) { |
| 217 | s2 = XDL_MAX(s2 - (s1 - fs1), 0); |
| 218 | s1 = fs1; |
| 219 | |
| 220 | /* |
| 221 | * Did we extend context upwards into an |
| 222 | * ignored change? |
| 223 | */ |
| 224 | while (xchp != xch && |
| 225 | xchp->i1 + xchp->chg1 <= s1 && |
| 226 | xchp->i2 + xchp->chg2 <= s2) |
| 227 | xchp = xchp->next; |
| 228 | |
| 229 | /* If so, show it after all. */ |
| 230 | if (xchp != xch) { |
| 231 | xch = xchp; |
| 232 | goto pre_context_calculation; |
| 233 | } |
| 234 | } |
| 235 | } |
| 236 | |
| 237 | post_context_calculation: |
| 238 | lctx = xecfg->ctxlen; |
| 239 | lctx = XDL_MIN(lctx, (long)xe->xdf1.nrec - (xche->i1 + xche->chg1)); |
| 240 | lctx = XDL_MIN(lctx, (long)xe->xdf2.nrec - (xche->i2 + xche->chg2)); |
| 241 | |
| 242 | e1 = xche->i1 + xche->chg1 + lctx; |
| 243 | e2 = xche->i2 + xche->chg2 + lctx; |
| 244 | |
| 245 | if (xecfg->flags & XDL_EMIT_FUNCCONTEXT) { |
| 246 | long fe1 = get_func_line(xe, xecfg, NULL, |
| 247 | xche->i1 + xche->chg1, |
| 248 | (long)xe->xdf1.nrec); |
| 249 | while (fe1 > 0 && is_empty_rec(&xe->xdf1, fe1 - 1)) |
| 250 | fe1--; |
| 251 | if (fe1 < 0) |
| 252 | fe1 = (long)xe->xdf1.nrec; |
| 253 | if (fe1 > e1) { |
| 254 | e2 = XDL_MIN(e2 + (fe1 - e1), (long)xe->xdf2.nrec); |
| 255 | e1 = fe1; |
| 256 | } |
| 257 | |
| 258 | /* |
| 259 | * Overlap with next change? Then include it |
| 260 | * in the current hunk and start over to find |
| 261 | * its new end. |
| 262 | */ |
| 263 | if (xche->next) { |
| 264 | long l = XDL_MIN(xche->next->i1, |
| 265 | (long)xe->xdf1.nrec - 1); |
| 266 | if (l - xecfg->ctxlen <= e1 || |
| 267 | get_func_line(xe, xecfg, NULL, l, e1) < 0) { |
| 268 | xche = xche->next; |
| 269 | goto post_context_calculation; |
| 270 | } |
| 271 | } |
| 272 | } |
| 273 | |
| 274 | /* |
| 275 | * Emit current hunk header. |
| 276 | */ |
| 277 | |
| 278 | if (xecfg->flags & XDL_EMIT_FUNCNAMES) { |
| 279 | get_func_line(xe, xecfg, &func_line, |
| 280 | s1 - 1, funclineprev); |
| 281 | funclineprev = s1 - 1; |
| 282 | } |
| 283 | if (!(xecfg->flags & XDL_EMIT_NO_HUNK_HDR) && |
| 284 | xdl_emit_hunk_hdr(s1 + 1, e1 - s1, s2 + 1, e2 - s2, |
| 285 | func_line.buf, func_line.len, ecb) < 0) |
| 286 | return -1; |
| 287 | |
| 288 | /* |
| 289 | * Emit pre-context. |
| 290 | */ |
| 291 | for (; s2 < xch->i2; s2++) |
| 292 | if (xdl_emit_record(&xe->xdf2, s2, " ", ecb) < 0) |
| 293 | return -1; |
| 294 | |
| 295 | for (s1 = xch->i1, s2 = xch->i2;; xch = xch->next) { |
| 296 | /* |
| 297 | * Merge previous with current change atom. |
| 298 | */ |
| 299 | for (; s1 < xch->i1 && s2 < xch->i2; s1++, s2++) |
| 300 | if (xdl_emit_record(&xe->xdf2, s2, " ", ecb) < 0) |
| 301 | return -1; |
| 302 | |
| 303 | /* |
| 304 | * Removes lines from the first file. |
| 305 | */ |
| 306 | for (s1 = xch->i1; s1 < xch->i1 + xch->chg1; s1++) |
| 307 | if (xdl_emit_record(&xe->xdf1, s1, "-", ecb) < 0) |
| 308 | return -1; |
| 309 | |
| 310 | /* |
| 311 | * Adds lines from the second file. |
| 312 | */ |
| 313 | for (s2 = xch->i2; s2 < xch->i2 + xch->chg2; s2++) |
| 314 | if (xdl_emit_record(&xe->xdf2, s2, "+", ecb) < 0) |
| 315 | return -1; |
| 316 | |
| 317 | if (xch == xche) |
| 318 | break; |
| 319 | s1 = xch->i1 + xch->chg1; |
| 320 | s2 = xch->i2 + xch->chg2; |
| 321 | } |
| 322 | |
| 323 | /* |
| 324 | * Emit post-context. |
| 325 | */ |
| 326 | for (s2 = xche->i2 + xche->chg2; s2 < e2; s2++) |
| 327 | if (xdl_emit_record(&xe->xdf2, s2, " ", ecb) < 0) |
| 328 | return -1; |
| 329 | } |
| 330 | |
| 331 | return 0; |
| 332 | } |