summaryrefslogtreecommitdiff
path: root/xdiff
diff options
context:
space:
mode:
Diffstat (limited to 'xdiff')
-rw-r--r--xdiff/xdiff.h9
-rw-r--r--xdiff/xdiffi.c7
-rw-r--r--xdiff/xdiffi.h2
-rw-r--r--xdiff/xemit.c6
-rw-r--r--xdiff/xemit.h3
-rw-r--r--xdiff/xmerge.c237
-rw-r--r--xdiff/xpatience.c381
-rw-r--r--xdiff/xprepare.c18
-rw-r--r--xdiff/xutils.c6
9 files changed, 588 insertions, 81 deletions
diff --git a/xdiff/xdiff.h b/xdiff/xdiff.h
index 413082e1fd..4da052a3ff 100644
--- a/xdiff/xdiff.h
+++ b/xdiff/xdiff.h
@@ -32,6 +32,7 @@ extern "C" {
#define XDF_IGNORE_WHITESPACE (1 << 2)
#define XDF_IGNORE_WHITESPACE_CHANGE (1 << 3)
#define XDF_IGNORE_WHITESPACE_AT_EOL (1 << 4)
+#define XDF_PATIENCE_DIFF (1 << 5)
#define XDF_WHITESPACE_FLAGS (XDF_IGNORE_WHITESPACE | XDF_IGNORE_WHITESPACE_CHANGE | XDF_IGNORE_WHITESPACE_AT_EOL)
#define XDL_PATCH_NORMAL '-'
@@ -50,10 +51,16 @@ extern "C" {
#define XDL_BDOP_CPY 2
#define XDL_BDOP_INSB 3
+/* merge simplification levels */
#define XDL_MERGE_MINIMAL 0
#define XDL_MERGE_EAGER 1
#define XDL_MERGE_ZEALOUS 2
#define XDL_MERGE_ZEALOUS_ALNUM 3
+#define XDL_MERGE_LEVEL_MASK 0x0f
+
+/* merge output styles */
+#define XDL_MERGE_DIFF3 0x8000
+#define XDL_MERGE_STYLE_MASK 0x8000
typedef struct s_mmfile {
char *ptr;
@@ -78,9 +85,11 @@ typedef long (*find_func_t)(const char *line, long line_len, char *buffer, long
typedef struct s_xdemitconf {
long ctxlen;
+ long interhunkctxlen;
unsigned long flags;
find_func_t find_func;
void *find_func_priv;
+ void (*emit_func)();
} xdemitconf_t;
typedef struct s_bdiffparam {
diff --git a/xdiff/xdiffi.c b/xdiff/xdiffi.c
index 1bad8462fb..3e97462bdd 100644
--- a/xdiff/xdiffi.c
+++ b/xdiff/xdiffi.c
@@ -329,6 +329,9 @@ int xdl_do_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
xdalgoenv_t xenv;
diffdata_t dd1, dd2;
+ if (xpp->flags & XDF_PATIENCE_DIFF)
+ return xdl_do_patience_diff(mf1, mf2, xpp, xe);
+
if (xdl_prepare_env(mf1, mf2, xpp, xe) < 0) {
return -1;
@@ -538,6 +541,8 @@ int xdl_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
xdemitconf_t const *xecfg, xdemitcb_t *ecb) {
xdchange_t *xscr;
xdfenv_t xe;
+ emit_func_t ef = xecfg->emit_func ?
+ (emit_func_t)xecfg->emit_func : xdl_emit_diff;
if (xdl_do_diff(mf1, mf2, xpp, &xe) < 0) {
@@ -551,7 +556,7 @@ int xdl_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
return -1;
}
if (xscr) {
- if (xdl_emit_diff(&xe, xscr, ecb, xecfg) < 0) {
+ if (ef(&xe, xscr, ecb, xecfg) < 0) {
xdl_free_script(xscr);
xdl_free_env(&xe);
diff --git a/xdiff/xdiffi.h b/xdiff/xdiffi.h
index 3e099dc445..ad033a8e6a 100644
--- a/xdiff/xdiffi.h
+++ b/xdiff/xdiffi.h
@@ -55,5 +55,7 @@ int xdl_build_script(xdfenv_t *xe, xdchange_t **xscr);
void xdl_free_script(xdchange_t *xscr);
int xdl_emit_diff(xdfenv_t *xe, xdchange_t *xscr, xdemitcb_t *ecb,
xdemitconf_t const *xecfg);
+int xdl_do_patience_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
+ xdfenv_t *env);
#endif /* #if !defined(XDIFFI_H) */
diff --git a/xdiff/xemit.c b/xdiff/xemit.c
index d3d9c845c6..05bfa41f10 100644
--- a/xdiff/xemit.c
+++ b/xdiff/xemit.c
@@ -27,7 +27,6 @@
static long xdl_get_rec(xdfile_t *xdf, long ri, char const **rec);
static int xdl_emit_record(xdfile_t *xdf, long ri, char const *pre, xdemitcb_t *ecb);
-static xdchange_t *xdl_get_hunk(xdchange_t *xscr, xdemitconf_t const *xecfg);
@@ -58,11 +57,12 @@ static int xdl_emit_record(xdfile_t *xdf, long ri, char const *pre, xdemitcb_t *
* Starting at the passed change atom, find the latest change atom to be included
* inside the differential hunk according to the specified configuration.
*/
-static xdchange_t *xdl_get_hunk(xdchange_t *xscr, xdemitconf_t const *xecfg) {
+xdchange_t *xdl_get_hunk(xdchange_t *xscr, xdemitconf_t const *xecfg) {
xdchange_t *xch, *xchp;
+ long max_common = 2 * xecfg->ctxlen + xecfg->interhunkctxlen;
for (xchp = xscr, xch = xscr->next; xch; xchp = xch, xch = xch->next)
- if (xch->i1 - (xchp->i1 + xchp->chg1) > 2 * xecfg->ctxlen)
+ if (xch->i1 - (xchp->i1 + xchp->chg1) > max_common)
break;
return xchp;
diff --git a/xdiff/xemit.h b/xdiff/xemit.h
index 440a7390fa..c2e2e83027 100644
--- a/xdiff/xemit.h
+++ b/xdiff/xemit.h
@@ -24,7 +24,10 @@
#define XEMIT_H
+typedef int (*emit_func_t)(xdfenv_t *xe, xdchange_t *xscr, xdemitcb_t *ecb,
+ xdemitconf_t const *xecfg);
+xdchange_t *xdl_get_hunk(xdchange_t *xscr, xdemitconf_t const *xecfg);
int xdl_emit_diff(xdfenv_t *xe, xdchange_t *xscr, xdemitcb_t *ecb,
xdemitconf_t const *xecfg);
diff --git a/xdiff/xmerge.c b/xdiff/xmerge.c
index 82b3573e7a..d9737f04c2 100644
--- a/xdiff/xmerge.c
+++ b/xdiff/xmerge.c
@@ -30,17 +30,32 @@ typedef struct s_xdmerge {
* 2 = no conflict, take second.
*/
int mode;
+ /*
+ * These point at the respective postimages. E.g. <i1,chg1> is
+ * how side #1 wants to change the common ancestor; if there is no
+ * overlap, lines before i1 in the postimage of side #1 appear
+ * in the merge result as a region touched by neither side.
+ */
long i1, i2;
long chg1, chg2;
+ /*
+ * These point at the preimage; of course there is just one
+ * preimage, that is from the shared common ancestor.
+ */
+ long i0;
+ long chg0;
} xdmerge_t;
static int xdl_append_merge(xdmerge_t **merge, int mode,
- long i1, long chg1, long i2, long chg2)
+ long i0, long chg0,
+ long i1, long chg1,
+ long i2, long chg2)
{
xdmerge_t *m = *merge;
if (m && (i1 <= m->i1 + m->chg1 || i2 <= m->i2 + m->chg2)) {
if (mode != m->mode)
m->mode = 0;
+ m->chg0 = i0 + chg0 - m->i0;
m->chg1 = i1 + chg1 - m->i1;
m->chg2 = i2 + chg2 - m->i2;
} else {
@@ -49,6 +64,8 @@ static int xdl_append_merge(xdmerge_t **merge, int mode,
return -1;
m->next = NULL;
m->mode = mode;
+ m->i0 = i0;
+ m->chg0 = chg0;
m->i1 = i1;
m->chg1 = chg1;
m->i2 = i2;
@@ -91,11 +108,13 @@ static int xdl_merge_cmp_lines(xdfenv_t *xe1, int i1, xdfenv_t *xe2, int i2,
return 0;
}
-static int xdl_recs_copy(xdfenv_t *xe, int i, int count, int add_nl, char *dest)
+static int xdl_recs_copy_0(int use_orig, xdfenv_t *xe, int i, int count, int add_nl, char *dest)
{
- xrecord_t **recs = xe->xdf2.recs + i;
+ xrecord_t **recs;
int size = 0;
+ recs = (use_orig ? xe->xdf1.recs : xe->xdf2.recs) + i;
+
if (count < 1)
return 0;
@@ -113,65 +132,109 @@ static int xdl_recs_copy(xdfenv_t *xe, int i, int count, int add_nl, char *dest)
return size;
}
-static int xdl_fill_merge_buffer(xdfenv_t *xe1, const char *name1,
- xdfenv_t *xe2, const char *name2, xdmerge_t *m, char *dest)
+static int xdl_recs_copy(xdfenv_t *xe, int i, int count, int add_nl, char *dest)
+{
+ return xdl_recs_copy_0(0, xe, i, count, add_nl, dest);
+}
+
+static int xdl_orig_copy(xdfenv_t *xe, int i, int count, int add_nl, char *dest)
+{
+ return xdl_recs_copy_0(1, xe, i, count, add_nl, dest);
+}
+
+static int fill_conflict_hunk(xdfenv_t *xe1, const char *name1,
+ xdfenv_t *xe2, const char *name2,
+ int size, int i, int style,
+ xdmerge_t *m, char *dest)
{
const int marker_size = 7;
int marker1_size = (name1 ? strlen(name1) + 1 : 0);
int marker2_size = (name2 ? strlen(name2) + 1 : 0);
- int conflict_marker_size = 3 * (marker_size + 1)
- + marker1_size + marker2_size;
- int size, i1, j;
-
- for (size = i1 = 0; m; m = m->next) {
- if (m->mode == 0) {
- size += xdl_recs_copy(xe1, i1, m->i1 - i1, 0,
- dest ? dest + size : NULL);
- if (dest) {
- for (j = 0; j < marker_size; j++)
- dest[size++] = '<';
- if (marker1_size) {
- dest[size] = ' ';
- memcpy(dest + size + 1, name1,
- marker1_size - 1);
- size += marker1_size;
- }
- dest[size++] = '\n';
- } else
- size += conflict_marker_size;
- size += xdl_recs_copy(xe1, m->i1, m->chg1, 1,
- dest ? dest + size : NULL);
- if (dest) {
- for (j = 0; j < marker_size; j++)
- dest[size++] = '=';
- dest[size++] = '\n';
- }
- size += xdl_recs_copy(xe2, m->i2, m->chg2, 1,
- dest ? dest + size : NULL);
- if (dest) {
- for (j = 0; j < marker_size; j++)
- dest[size++] = '>';
- if (marker2_size) {
- dest[size] = ' ';
- memcpy(dest + size + 1, name2,
- marker2_size - 1);
- size += marker2_size;
- }
- dest[size++] = '\n';
- }
- } else if (m->mode == 1)
- size += xdl_recs_copy(xe1, i1, m->i1 + m->chg1 - i1, 0,
- dest ? dest + size : NULL);
+ int j;
+
+ /* Before conflicting part */
+ size += xdl_recs_copy(xe1, i, m->i1 - i, 0,
+ dest ? dest + size : NULL);
+
+ if (!dest) {
+ size += marker_size + 1 + marker1_size;
+ } else {
+ for (j = 0; j < marker_size; j++)
+ dest[size++] = '<';
+ if (marker1_size) {
+ dest[size] = ' ';
+ memcpy(dest + size + 1, name1, marker1_size - 1);
+ size += marker1_size;
+ }
+ dest[size++] = '\n';
+ }
+
+ /* Postimage from side #1 */
+ size += xdl_recs_copy(xe1, m->i1, m->chg1, 1,
+ dest ? dest + size : NULL);
+
+ if (style == XDL_MERGE_DIFF3) {
+ /* Shared preimage */
+ if (!dest) {
+ size += marker_size + 1;
+ } else {
+ for (j = 0; j < marker_size; j++)
+ dest[size++] = '|';
+ dest[size++] = '\n';
+ }
+ size += xdl_orig_copy(xe1, m->i0, m->chg0, 1,
+ dest ? dest + size : NULL);
+ }
+
+ if (!dest) {
+ size += marker_size + 1;
+ } else {
+ for (j = 0; j < marker_size; j++)
+ dest[size++] = '=';
+ dest[size++] = '\n';
+ }
+
+ /* Postimage from side #2 */
+ size += xdl_recs_copy(xe2, m->i2, m->chg2, 1,
+ dest ? dest + size : NULL);
+ if (!dest) {
+ size += marker_size + 1 + marker2_size;
+ } else {
+ for (j = 0; j < marker_size; j++)
+ dest[size++] = '>';
+ if (marker2_size) {
+ dest[size] = ' ';
+ memcpy(dest + size + 1, name2, marker2_size - 1);
+ size += marker2_size;
+ }
+ dest[size++] = '\n';
+ }
+ return size;
+}
+
+static int xdl_fill_merge_buffer(xdfenv_t *xe1, const char *name1,
+ xdfenv_t *xe2, const char *name2,
+ xdmerge_t *m, char *dest, int style)
+{
+ int size, i;
+
+ for (size = i = 0; m; m = m->next) {
+ if (m->mode == 0)
+ size = fill_conflict_hunk(xe1, name1, xe2, name2,
+ size, i, style, m, dest);
+ else if (m->mode == 1)
+ size += xdl_recs_copy(xe1, i, m->i1 + m->chg1 - i, 0,
+ dest ? dest + size : NULL);
else if (m->mode == 2)
- size += xdl_recs_copy(xe2, m->i2 - m->i1 + i1,
- m->i1 + m->chg2 - i1, 0,
- dest ? dest + size : NULL);
+ size += xdl_recs_copy(xe2, m->i2 - m->i1 + i,
+ m->i1 + m->chg2 - i, 0,
+ dest ? dest + size : NULL);
else
continue;
- i1 = m->i1 + m->chg1;
+ i = m->i1 + m->chg1;
}
- size += xdl_recs_copy(xe1, i1, xe1->xdf2.nrec - i1, 0,
- dest ? dest + size : NULL);
+ size += xdl_recs_copy(xe1, i, xe1->xdf2.nrec - i, 0,
+ dest ? dest + size : NULL);
return size;
}
@@ -323,9 +386,20 @@ static int xdl_simplify_non_conflicts(xdfenv_t *xe1, xdmerge_t *m,
*/
static int xdl_do_merge(xdfenv_t *xe1, xdchange_t *xscr1, const char *name1,
xdfenv_t *xe2, xdchange_t *xscr2, const char *name2,
- int level, xpparam_t const *xpp, mmbuffer_t *result) {
+ int flags, xpparam_t const *xpp, mmbuffer_t *result) {
xdmerge_t *changes, *c;
- int i1, i2, chg1, chg2;
+ int i0, i1, i2, chg0, chg1, chg2;
+ int level = flags & XDL_MERGE_LEVEL_MASK;
+ int style = flags & XDL_MERGE_STYLE_MASK;
+
+ if (style == XDL_MERGE_DIFF3) {
+ /*
+ * "diff3 -m" output does not make sense for anything
+ * more aggressive than XDL_MERGE_EAGER.
+ */
+ if (XDL_MERGE_EAGER < level)
+ level = XDL_MERGE_EAGER;
+ }
c = changes = NULL;
@@ -333,11 +407,14 @@ static int xdl_do_merge(xdfenv_t *xe1, xdchange_t *xscr1, const char *name1,
if (!changes)
changes = c;
if (xscr1->i1 + xscr1->chg1 < xscr2->i1) {
+ i0 = xscr1->i1;
i1 = xscr1->i2;
i2 = xscr2->i2 - xscr2->i1 + xscr1->i1;
+ chg0 = xscr1->chg1;
chg1 = xscr1->chg2;
chg2 = xscr1->chg1;
- if (xdl_append_merge(&c, 1, i1, chg1, i2, chg2)) {
+ if (xdl_append_merge(&c, 1,
+ i0, chg0, i1, chg1, i2, chg2)) {
xdl_cleanup_merge(changes);
return -1;
}
@@ -345,18 +422,21 @@ static int xdl_do_merge(xdfenv_t *xe1, xdchange_t *xscr1, const char *name1,
continue;
}
if (xscr2->i1 + xscr2->chg1 < xscr1->i1) {
+ i0 = xscr2->i1;
i1 = xscr1->i2 - xscr1->i1 + xscr2->i1;
i2 = xscr2->i2;
+ chg0 = xscr2->chg1;
chg1 = xscr2->chg1;
chg2 = xscr2->chg2;
- if (xdl_append_merge(&c, 2, i1, chg1, i2, chg2)) {
+ if (xdl_append_merge(&c, 2,
+ i0, chg0, i1, chg1, i2, chg2)) {
xdl_cleanup_merge(changes);
return -1;
}
xscr2 = xscr2->next;
continue;
}
- if (level < 1 || xscr1->i1 != xscr2->i1 ||
+ if (level == XDL_MERGE_MINIMAL || xscr1->i1 != xscr2->i1 ||
xscr1->chg1 != xscr2->chg1 ||
xscr1->chg2 != xscr2->chg2 ||
xdl_merge_cmp_lines(xe1, xscr1->i2,
@@ -366,19 +446,25 @@ static int xdl_do_merge(xdfenv_t *xe1, xdchange_t *xscr1, const char *name1,
int off = xscr1->i1 - xscr2->i1;
int ffo = off + xscr1->chg1 - xscr2->chg1;
+ i0 = xscr1->i1;
i1 = xscr1->i2;
i2 = xscr2->i2;
- if (off > 0)
+ if (off > 0) {
+ i0 -= off;
i1 -= off;
+ }
else
i2 += off;
+ chg0 = xscr1->i1 + xscr1->chg1 - i0;
chg1 = xscr1->i2 + xscr1->chg2 - i1;
chg2 = xscr2->i2 + xscr2->chg2 - i2;
- if (ffo > 0)
- chg2 += ffo;
- else
+ if (ffo < 0) {
+ chg0 -= ffo;
chg1 -= ffo;
- if (xdl_append_merge(&c, 0, i1, chg1, i2, chg2)) {
+ } else
+ chg2 += ffo;
+ if (xdl_append_merge(&c, 0,
+ i0, chg0, i1, chg1, i2, chg2)) {
xdl_cleanup_merge(changes);
return -1;
}
@@ -395,11 +481,14 @@ static int xdl_do_merge(xdfenv_t *xe1, xdchange_t *xscr1, const char *name1,
while (xscr1) {
if (!changes)
changes = c;
+ i0 = xscr1->i1;
i1 = xscr1->i2;
i2 = xscr1->i1 + xe2->xdf2.nrec - xe2->xdf1.nrec;
+ chg0 = xscr1->chg1;
chg1 = xscr1->chg2;
chg2 = xscr1->chg1;
- if (xdl_append_merge(&c, 1, i1, chg1, i2, chg2)) {
+ if (xdl_append_merge(&c, 1,
+ i0, chg0, i1, chg1, i2, chg2)) {
xdl_cleanup_merge(changes);
return -1;
}
@@ -408,11 +497,14 @@ static int xdl_do_merge(xdfenv_t *xe1, xdchange_t *xscr1, const char *name1,
while (xscr2) {
if (!changes)
changes = c;
+ i0 = xscr2->i1;
i1 = xscr2->i1 + xe1->xdf2.nrec - xe1->xdf1.nrec;
i2 = xscr2->i2;
+ chg0 = xscr2->chg1;
chg1 = xscr2->chg1;
chg2 = xscr2->chg2;
- if (xdl_append_merge(&c, 2, i1, chg1, i2, chg2)) {
+ if (xdl_append_merge(&c, 2,
+ i0, chg0, i1, chg1, i2, chg2)) {
xdl_cleanup_merge(changes);
return -1;
}
@@ -421,16 +513,17 @@ static int xdl_do_merge(xdfenv_t *xe1, xdchange_t *xscr1, const char *name1,
if (!changes)
changes = c;
/* refine conflicts */
- if (level > 1 &&
+ if (XDL_MERGE_ZEALOUS <= level &&
(xdl_refine_conflicts(xe1, xe2, changes, xpp) < 0 ||
- xdl_simplify_non_conflicts(xe1, changes, level > 2) < 0)) {
+ xdl_simplify_non_conflicts(xe1, changes,
+ XDL_MERGE_ZEALOUS < level) < 0)) {
xdl_cleanup_merge(changes);
return -1;
}
/* output */
if (result) {
int size = xdl_fill_merge_buffer(xe1, name1, xe2, name2,
- changes, NULL);
+ changes, NULL, style);
result->ptr = xdl_malloc(size);
if (!result->ptr) {
xdl_cleanup_merge(changes);
@@ -438,14 +531,14 @@ static int xdl_do_merge(xdfenv_t *xe1, xdchange_t *xscr1, const char *name1,
}
result->size = size;
xdl_fill_merge_buffer(xe1, name1, xe2, name2, changes,
- result->ptr);
+ result->ptr, style);
}
return xdl_cleanup_merge(changes);
}
int xdl_merge(mmfile_t *orig, mmfile_t *mf1, const char *name1,
mmfile_t *mf2, const char *name2,
- xpparam_t const *xpp, int level, mmbuffer_t *result) {
+ xpparam_t const *xpp, int flags, mmbuffer_t *result) {
xdchange_t *xscr1, *xscr2;
xdfenv_t xe1, xe2;
int status;
@@ -482,7 +575,7 @@ int xdl_merge(mmfile_t *orig, mmfile_t *mf1, const char *name1,
} else {
status = xdl_do_merge(&xe1, xscr1, name1,
&xe2, xscr2, name2,
- level, xpp, result);
+ flags, xpp, result);
}
xdl_free_script(xscr1);
xdl_free_script(xscr2);
diff --git a/xdiff/xpatience.c b/xdiff/xpatience.c
new file mode 100644
index 0000000000..e42c16a807
--- /dev/null
+++ b/xdiff/xpatience.c
@@ -0,0 +1,381 @@
+/*
+ * LibXDiff by Davide Libenzi ( File Differential Library )
+ * Copyright (C) 2003-2009 Davide Libenzi, Johannes E. Schindelin
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Davide Libenzi <davidel@xmailserver.org>
+ *
+ */
+#include "xinclude.h"
+#include "xtypes.h"
+#include "xdiff.h"
+
+/*
+ * The basic idea of patience diff is to find lines that are unique in
+ * both files. These are intuitively the ones that we want to see as
+ * common lines.
+ *
+ * The maximal ordered sequence of such line pairs (where ordered means
+ * that the order in the sequence agrees with the order of the lines in
+ * both files) naturally defines an initial set of common lines.
+ *
+ * Now, the algorithm tries to extend the set of common lines by growing
+ * the line ranges where the files have identical lines.
+ *
+ * Between those common lines, the patience diff algorithm is applied
+ * recursively, until no unique line pairs can be found; these line ranges
+ * are handled by the well-known Myers algorithm.
+ */
+
+#define NON_UNIQUE ULONG_MAX
+
+/*
+ * This is a hash mapping from line hash to line numbers in the first and
+ * second file.
+ */
+struct hashmap {
+ int nr, alloc;
+ struct entry {
+ unsigned long hash;
+ /*
+ * 0 = unused entry, 1 = first line, 2 = second, etc.
+ * line2 is NON_UNIQUE if the line is not unique
+ * in either the first or the second file.
+ */
+ unsigned long line1, line2;
+ /*
+ * "next" & "previous" are used for the longest common
+ * sequence;
+ * initially, "next" reflects only the order in file1.
+ */
+ struct entry *next, *previous;
+ } *entries, *first, *last;
+ /* were common records found? */
+ unsigned long has_matches;
+ mmfile_t *file1, *file2;
+ xdfenv_t *env;
+ xpparam_t const *xpp;
+};
+
+/* The argument "pass" is 1 for the first file, 2 for the second. */
+static void insert_record(int line, struct hashmap *map, int pass)
+{
+ xrecord_t **records = pass == 1 ?
+ map->env->xdf1.recs : map->env->xdf2.recs;
+ xrecord_t *record = records[line - 1], *other;
+ /*
+ * After xdl_prepare_env() (or more precisely, due to
+ * xdl_classify_record()), the "ha" member of the records (AKA lines)
+ * is _not_ the hash anymore, but a linearized version of it. In
+ * other words, the "ha" member is guaranteed to start with 0 and
+ * the second record's ha can only be 0 or 1, etc.
+ *
+ * So we multiply ha by 2 in the hope that the hashing was
+ * "unique enough".
+ */
+ int index = (int)((record->ha << 1) % map->alloc);
+
+ while (map->entries[index].line1) {
+ other = map->env->xdf1.recs[map->entries[index].line1 - 1];
+ if (map->entries[index].hash != record->ha ||
+ !xdl_recmatch(record->ptr, record->size,
+ other->ptr, other->size,
+ map->xpp->flags)) {
+ if (++index >= map->alloc)
+ index = 0;
+ continue;
+ }
+ if (pass == 2)
+ map->has_matches = 1;
+ if (pass == 1 || map->entries[index].line2)
+ map->entries[index].line2 = NON_UNIQUE;
+ else
+ map->entries[index].line2 = line;
+ return;
+ }
+ if (pass == 2)
+ return;
+ map->entries[index].line1 = line;
+ map->entries[index].hash = record->ha;
+ if (!map->first)
+ map->first = map->entries + index;
+ if (map->last) {
+ map->last->next = map->entries + index;
+ map->entries[index].previous = map->last;
+ }
+ map->last = map->entries + index;
+ map->nr++;
+}
+
+/*
+ * This function has to be called for each recursion into the inter-hunk
+ * parts, as previously non-unique lines can become unique when being
+ * restricted to a smaller part of the files.
+ *
+ * It is assumed that env has been prepared using xdl_prepare().
+ */
+static int fill_hashmap(mmfile_t *file1, mmfile_t *file2,
+ xpparam_t const *xpp, xdfenv_t *env,
+ struct hashmap *result,
+ int line1, int count1, int line2, int count2)
+{
+ result->file1 = file1;
+ result->file2 = file2;
+ result->xpp = xpp;
+ result->env = env;
+
+ /* We know exactly how large we want the hash map */
+ result->alloc = count1 * 2;
+ result->entries = (struct entry *)
+ xdl_malloc(result->alloc * sizeof(struct entry));
+ if (!result->entries)
+ return -1;
+ memset(result->entries, 0, result->alloc * sizeof(struct entry));
+
+ /* First, fill with entries from the first file */
+ while (count1--)
+ insert_record(line1++, result, 1);
+
+ /* Then search for matches in the second file */
+ while (count2--)
+ insert_record(line2++, result, 2);
+
+ return 0;
+}
+
+/*
+ * Find the longest sequence with a smaller last element (meaning a smaller
+ * line2, as we construct the sequence with entries ordered by line1).
+ */
+static int binary_search(struct entry **sequence, int longest,
+ struct entry *entry)
+{
+ int left = -1, right = longest;
+
+ while (left + 1 < right) {
+ int middle = (left + right) / 2;
+ /* by construction, no two entries can be equal */
+ if (sequence[middle]->line2 > entry->line2)
+ right = middle;
+ else
+ left = middle;
+ }
+ /* return the index in "sequence", _not_ the sequence length */
+ return left;
+}
+
+/*
+ * The idea is to start with the list of common unique lines sorted by
+ * the order in file1. For each of these pairs, the longest (partial)
+ * sequence whose last element's line2 is smaller is determined.
+ *
+ * For efficiency, the sequences are kept in a list containing exactly one
+ * item per sequence length: the sequence with the smallest last
+ * element (in terms of line2).
+ */
+static struct entry *find_longest_common_sequence(struct hashmap *map)
+{
+ struct entry **sequence = xdl_malloc(map->nr * sizeof(struct entry *));
+ int longest = 0, i;
+ struct entry *entry;
+
+ for (entry = map->first; entry; entry = entry->next) {
+ if (!entry->line2 || entry->line2 == NON_UNIQUE)
+ continue;
+ i = binary_search(sequence, longest, entry);
+ entry->previous = i < 0 ? NULL : sequence[i];
+ sequence[++i] = entry;
+ if (i == longest)
+ longest++;
+ }
+
+ /* No common unique lines were found */
+ if (!longest) {
+ xdl_free(sequence);
+ return NULL;
+ }
+
+ /* Iterate starting at the last element, adjusting the "next" members */
+ entry = sequence[longest - 1];
+ entry->next = NULL;
+ while (entry->previous) {
+ entry->previous->next = entry;
+ entry = entry->previous;
+ }
+ xdl_free(sequence);
+ return entry;
+}
+
+static int match(struct hashmap *map, int line1, int line2)
+{
+ xrecord_t *record1 = map->env->xdf1.recs[line1 - 1];
+ xrecord_t *record2 = map->env->xdf2.recs[line2 - 1];
+ return xdl_recmatch(record1->ptr, record1->size,
+ record2->ptr, record2->size, map->xpp->flags);
+}
+
+static int patience_diff(mmfile_t *file1, mmfile_t *file2,
+ xpparam_t const *xpp, xdfenv_t *env,
+ int line1, int count1, int line2, int count2);
+
+static int walk_common_sequence(struct hashmap *map, struct entry *first,
+ int line1, int count1, int line2, int count2)
+{
+ int end1 = line1 + count1, end2 = line2 + count2;
+ int next1, next2;
+
+ for (;;) {
+ /* Try to grow the line ranges of common lines */
+ if (first) {
+ next1 = first->line1;
+ next2 = first->line2;
+ while (next1 > line1 && next2 > line2 &&
+ match(map, next1 - 1, next2 - 1)) {
+ next1--;
+ next2--;
+ }
+ } else {
+ next1 = end1;
+ next2 = end2;
+ }
+ while (line1 < next1 && line2 < next2 &&
+ match(map, line1, line2)) {
+ line1++;
+ line2++;
+ }
+
+ /* Recurse */
+ if (next1 > line1 || next2 > line2) {
+ struct hashmap submap;
+
+ memset(&submap, 0, sizeof(submap));
+ if (patience_diff(map->file1, map->file2,
+ map->xpp, map->env,
+ line1, next1 - line1,
+ line2, next2 - line2))
+ return -1;
+ }
+
+ if (!first)
+ return 0;
+
+ while (first->next &&
+ first->next->line1 == first->line1 + 1 &&
+ first->next->line2 == first->line2 + 1)
+ first = first->next;
+
+ line1 = first->line1 + 1;
+ line2 = first->line2 + 1;
+
+ first = first->next;
+ }
+}
+
+static int fall_back_to_classic_diff(struct hashmap *map,
+ int line1, int count1, int line2, int count2)
+{
+ /*
+ * This probably does not work outside Git, since
+ * we have a very simple mmfile structure.
+ *
+ * Note: ideally, we would reuse the prepared environment, but
+ * the libxdiff interface does not (yet) allow for diffing only
+ * ranges of lines instead of the whole files.
+ */
+ mmfile_t subfile1, subfile2;
+ xpparam_t xpp;
+ xdfenv_t env;
+
+ subfile1.ptr = (char *)map->env->xdf1.recs[line1 - 1]->ptr;
+ subfile1.size = map->env->xdf1.recs[line1 + count1 - 2]->ptr +
+ map->env->xdf1.recs[line1 + count1 - 2]->size - subfile1.ptr;
+ subfile2.ptr = (char *)map->env->xdf2.recs[line2 - 1]->ptr;
+ subfile2.size = map->env->xdf2.recs[line2 + count2 - 2]->ptr +
+ map->env->xdf2.recs[line2 + count2 - 2]->size - subfile2.ptr;
+ xpp.flags = map->xpp->flags & ~XDF_PATIENCE_DIFF;
+ if (xdl_do_diff(&subfile1, &subfile2, &xpp, &env) < 0)
+ return -1;
+
+ memcpy(map->env->xdf1.rchg + line1 - 1, env.xdf1.rchg, count1);
+ memcpy(map->env->xdf2.rchg + line2 - 1, env.xdf2.rchg, count2);
+
+ xdl_free_env(&env);
+
+ return 0;
+}
+
+/*
+ * Recursively find the longest common sequence of unique lines,
+ * and if none was found, ask xdl_do_diff() to do the job.
+ *
+ * This function assumes that env was prepared with xdl_prepare_env().
+ */
+static int patience_diff(mmfile_t *file1, mmfile_t *file2,
+ xpparam_t const *xpp, xdfenv_t *env,
+ int line1, int count1, int line2, int count2)
+{
+ struct hashmap map;
+ struct entry *first;
+ int result = 0;
+
+ /* trivial case: one side is empty */
+ if (!count1) {
+ while(count2--)
+ env->xdf2.rchg[line2++ - 1] = 1;
+ return 0;
+ } else if (!count2) {
+ while(count1--)
+ env->xdf1.rchg[line1++ - 1] = 1;
+ return 0;
+ }
+
+ memset(&map, 0, sizeof(map));
+ if (fill_hashmap(file1, file2, xpp, env, &map,
+ line1, count1, line2, count2))
+ return -1;
+
+ /* are there any matching lines at all? */
+ if (!map.has_matches) {
+ while(count1--)
+ env->xdf1.rchg[line1++ - 1] = 1;
+ while(count2--)
+ env->xdf2.rchg[line2++ - 1] = 1;
+ xdl_free(map.entries);
+ return 0;
+ }
+
+ first = find_longest_common_sequence(&map);
+ if (first)
+ result = walk_common_sequence(&map, first,
+ line1, count1, line2, count2);
+ else
+ result = fall_back_to_classic_diff(&map,
+ line1, count1, line2, count2);
+
+ xdl_free(map.entries);
+ return result;
+}
+
+int xdl_do_patience_diff(mmfile_t *file1, mmfile_t *file2,
+ xpparam_t const *xpp, xdfenv_t *env)
+{
+ if (xdl_prepare_env(file1, file2, xpp, env) < 0)
+ return -1;
+
+ /* environment is cleaned up in xdl_diff() */
+ return patience_diff(file1, file2, xpp, env,
+ 1, env->xdf1.nrec, 1, env->xdf2.nrec);
+}
diff --git a/xdiff/xprepare.c b/xdiff/xprepare.c
index e87ab57c65..1689085235 100644
--- a/xdiff/xprepare.c
+++ b/xdiff/xprepare.c
@@ -23,10 +23,9 @@
#include "xinclude.h"
-
#define XDL_KPDIS_RUN 4
#define XDL_MAX_EQLIMIT 1024
-
+#define XDL_SIMSCAN_WINDOW 100
typedef struct s_xdlclass {
@@ -291,7 +290,8 @@ int xdl_prepare_env(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
xdl_free_classifier(&cf);
- if (xdl_optimize_ctxs(&xe->xdf1, &xe->xdf2) < 0) {
+ if (!(xpp->flags & XDF_PATIENCE_DIFF) &&
+ xdl_optimize_ctxs(&xe->xdf1, &xe->xdf2) < 0) {
xdl_free_ctx(&xe->xdf2);
xdl_free_ctx(&xe->xdf1);
@@ -313,6 +313,18 @@ static int xdl_clean_mmatch(char const *dis, long i, long s, long e) {
long r, rdis0, rpdis0, rdis1, rpdis1;
/*
+ * Limits the window the is examined during the similar-lines
+ * scan. The loops below stops when dis[i - r] == 1 (line that
+ * has no match), but there are corner cases where the loop
+ * proceed all the way to the extremities by causing huge
+ * performance penalties in case of big files.
+ */
+ if (i - s > XDL_SIMSCAN_WINDOW)
+ s = i - XDL_SIMSCAN_WINDOW;
+ if (e - i > XDL_SIMSCAN_WINDOW)
+ e = i + XDL_SIMSCAN_WINDOW;
+
+ /*
* Scans the lines before 'i' to find a run of lines that either
* have no match (dis[j] == 0) or have multiple matches (dis[j] > 1).
* Note that we always call this function with dis[i] > 1, so the
diff --git a/xdiff/xutils.c b/xdiff/xutils.c
index d7974d1a3e..04ad468702 100644
--- a/xdiff/xutils.c
+++ b/xdiff/xutils.c
@@ -245,12 +245,14 @@ static unsigned long xdl_hash_record_with_whitespace(char const **data,
while (ptr + 1 < top && isspace(ptr[1])
&& ptr[1] != '\n')
ptr++;
- if (flags & XDF_IGNORE_WHITESPACE_CHANGE
+ if (flags & XDF_IGNORE_WHITESPACE)
+ ; /* already handled */
+ else if (flags & XDF_IGNORE_WHITESPACE_CHANGE
&& ptr[1] != '\n') {
ha += (ha << 5);
ha ^= (unsigned long) ' ';
}
- if (flags & XDF_IGNORE_WHITESPACE_AT_EOL
+ else if (flags & XDF_IGNORE_WHITESPACE_AT_EOL
&& ptr[1] != '\n') {
while (ptr2 != ptr + 1) {
ha += (ha << 5);
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-26 04:55:50 +0000