]>
git.ipfire.org Git - thirdparty/git.git/blob - diffcore-break.c
2 * Copyright (C) 2005 Junio C Hamano
8 #include "count-delta.h"
10 static int should_break(struct diff_filespec
*src
,
11 struct diff_filespec
*dst
,
15 /* dst is recorded as a modification of src. Are they so
16 * different that we are better off recording this as a pair
17 * of delete and create?
19 * There are two criteria used in this algorithm. For the
20 * purposes of helping later rename/copy, we take both delete
21 * and insert into account and estimate the amount of "edit".
22 * If the edit is very large, we break this pair so that
23 * rename/copy can pick the pieces up to match with other
26 * On the other hand, we would want to ignore inserts for the
27 * pure "complete rewrite" detection. As long as most of the
28 * existing contents were removed from the file, it is a
29 * complete rewrite, and if sizable chunk from the original
30 * still remains in the result, it is not a rewrite. It does
31 * not matter how much or how little new material is added to
34 * The score we leave for such a broken filepair uses the
35 * latter definition so that later clean-up stage can find the
36 * pieces that should not have been broken according to the
37 * latter definition after rename/copy runs, and merge the
38 * broken pair that have a score lower than given criteria
39 * back together. The break operation itself happens
40 * according to the former definition.
42 * The minimum_edit parameter tells us when to break (the
43 * amount of "edit" required for us to consider breaking the
44 * pair). We leave the amount of deletion in *merge_score_p
47 * The value we return is 1 if we want the pair to be broken,
51 unsigned long delta_size
, base_size
, src_copied
, literal_added
;
54 *merge_score_p
= 0; /* assume no deletion --- "do not break"
58 if (!S_ISREG(src
->mode
) || !S_ISREG(dst
->mode
))
59 return 0; /* leave symlink rename alone */
61 if (diff_populate_filespec(src
, 0) || diff_populate_filespec(dst
, 0))
62 return 0; /* error but caught downstream */
64 base_size
= ((src
->size
< dst
->size
) ? src
->size
: dst
->size
);
65 if (base_size
< MINIMUM_BREAK_SIZE
)
66 return 0; /* we do not break too small filepair */
68 delta
= diff_delta(src
->data
, src
->size
,
72 return 0; /* error but caught downstream */
74 /* Estimate the edit size by interpreting delta. */
75 if (count_delta(delta
, delta_size
,
76 &src_copied
, &literal_added
)) {
78 return 0; /* we cannot tell */
82 /* Compute merge-score, which is "how much is removed
83 * from the source material". The clean-up stage will
84 * merge the surviving pair together if the score is
85 * less than the minimum, after rename/copy runs.
87 if (src
->size
<= src_copied
)
88 ; /* all copied, nothing removed */
90 delta_size
= src
->size
- src_copied
;
91 *merge_score_p
= delta_size
* MAX_SCORE
/ src
->size
;
94 /* Extent of damage, which counts both inserts and
97 if (src
->size
+ literal_added
<= src_copied
)
98 delta_size
= 0; /* avoid wrapping around */
100 delta_size
= (src
->size
- src_copied
) + literal_added
;
102 /* We break if the edit exceeds the minimum.
103 * i.e. (break_score / MAX_SCORE < delta_size / base_size)
105 if (break_score
* base_size
< delta_size
* MAX_SCORE
)
111 void diffcore_break(int break_score
)
113 struct diff_queue_struct
*q
= &diff_queued_diff
;
114 struct diff_queue_struct outq
;
116 /* When the filepair has this much edit (insert and delete),
117 * it is first considered to be a rewrite and broken into a
118 * create and delete filepair. This is to help breaking a
119 * file that had too much new stuff added, possibly from
120 * moving contents from another file, so that rename/copy can
121 * match it with the other file.
123 * int break_score; we reuse incoming parameter for this.
126 /* After a pair is broken according to break_score and
127 * subjected to rename/copy, both of them may survive intact,
128 * due to lack of suitable rename/copy peer. Or, the caller
129 * may be calling us without using rename/copy. When that
130 * happens, we merge the broken pieces back into one
131 * modification together if the pair did not have more than
132 * this much delete. For this computation, we do not take
133 * insert into account at all. If you start from a 100-line
134 * file and delete 97 lines of it, it does not matter if you
135 * add 27 lines to it to make a new 30-line file or if you add
136 * 997 lines to it to make a 1000-line file. Either way what
137 * you did was a rewrite of 97%. On the other hand, if you
138 * delete 3 lines, keeping 97 lines intact, it does not matter
139 * if you add 3 lines to it to make a new 100-line file or if
140 * you add 903 lines to it to make a new 1000-line file.
141 * Either way you did a lot of additions and not a rewrite.
142 * This merge happens to catch the latter case. A merge_score
143 * of 80% would be a good default value (a broken pair that
144 * has score lower than merge_score will be merged back
150 /* See comment on DEFAULT_BREAK_SCORE and
151 * DEFAULT_MERGE_SCORE in diffcore.h
153 merge_score
= (break_score
>> 16) & 0xFFFF;
154 break_score
= (break_score
& 0xFFFF);
157 break_score
= DEFAULT_BREAK_SCORE
;
159 merge_score
= DEFAULT_MERGE_SCORE
;
161 outq
.nr
= outq
.alloc
= 0;
164 for (i
= 0; i
< q
->nr
; i
++) {
165 struct diff_filepair
*p
= q
->queue
[i
];
168 /* We deal only with in-place edit of non directory.
169 * We do not break anything else.
171 if (DIFF_FILE_VALID(p
->one
) && DIFF_FILE_VALID(p
->two
) &&
172 !S_ISDIR(p
->one
->mode
) && !S_ISDIR(p
->two
->mode
) &&
173 !strcmp(p
->one
->path
, p
->two
->path
)) {
174 if (should_break(p
->one
, p
->two
,
175 break_score
, &score
)) {
176 /* Split this into delete and create */
177 struct diff_filespec
*null_one
, *null_two
;
178 struct diff_filepair
*dp
;
180 /* Set score to 0 for the pair that
181 * needs to be merged back together
182 * should they survive rename/copy.
183 * Also we do not want to break very
186 if (score
< merge_score
)
189 /* deletion of one */
190 null_one
= alloc_filespec(p
->one
->path
);
191 dp
= diff_queue(&outq
, p
->one
, null_one
);
195 /* creation of two */
196 null_two
= alloc_filespec(p
->two
->path
);
197 dp
= diff_queue(&outq
, null_two
, p
->two
);
201 free(p
); /* not diff_free_filepair(), we are
202 * reusing one and two here.
215 static void merge_broken(struct diff_filepair
*p
,
216 struct diff_filepair
*pp
,
217 struct diff_queue_struct
*outq
)
219 /* p and pp are broken pairs we want to merge */
220 struct diff_filepair
*c
= p
, *d
= pp
, *dp
;
221 if (DIFF_FILE_VALID(p
->one
)) {
222 /* this must be a delete half */
226 if (!DIFF_FILE_VALID(d
->one
))
227 die("internal error in merge #1");
228 if (DIFF_FILE_VALID(d
->two
))
229 die("internal error in merge #2");
230 if (DIFF_FILE_VALID(c
->one
))
231 die("internal error in merge #3");
232 if (!DIFF_FILE_VALID(c
->two
))
233 die("internal error in merge #4");
235 dp
= diff_queue(outq
, d
->one
, c
->two
);
236 dp
->score
= p
->score
;
237 diff_free_filespec_data(d
->two
);
238 diff_free_filespec_data(c
->one
);
243 void diffcore_merge_broken(void)
245 struct diff_queue_struct
*q
= &diff_queued_diff
;
246 struct diff_queue_struct outq
;
249 outq
.nr
= outq
.alloc
= 0;
252 for (i
= 0; i
< q
->nr
; i
++) {
253 struct diff_filepair
*p
= q
->queue
[i
];
255 /* we already merged this with its peer */
257 else if (p
->broken_pair
&&
258 !strcmp(p
->one
->path
, p
->two
->path
)) {
259 /* If the peer also survived rename/copy, then
260 * we merge them back together.
262 for (j
= i
+ 1; j
< q
->nr
; j
++) {
263 struct diff_filepair
*pp
= q
->queue
[j
];
264 if (pp
->broken_pair
&&
265 !strcmp(pp
->one
->path
, pp
->two
->path
) &&
266 !strcmp(p
->one
->path
, pp
->two
->path
)) {
267 /* Peer survived. Merge them */
268 merge_broken(p
, pp
, &outq
);
274 /* The peer did not survive, so we keep