summaryrefslogtreecommitdiff
path: root/contrib/diff-highlight/README
blob: 836b97a73075a51ef8bc66ba080085a7de4c8f32 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
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
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
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
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
diff-highlight
==============
 
Line oriented diffs are great for reviewing code, because for most
hunks, you want to see the old and the new segments of code next to each
other. Sometimes, though, when an old line and a new line are very
similar, it's hard to immediately see the difference.
 
You can use "--color-words" to highlight only the changed portions of
lines. However, this can often be hard to read for code, as it loses
the line structure, and you end up with oddly formatted bits.
 
Instead, this script post-processes the line-oriented diff, finds pairs
of lines, and highlights the differing segments.  It's currently very
simple and stupid about doing these tasks. In particular:
 
  1. It will only highlight hunks in which the number of removed and
     added lines is the same, and it will pair lines within the hunk by
     position (so the first removed line is compared to the first added
     line, and so forth). This is simple and tends to work well in
     practice. More complex changes don't highlight well, so we tend to
     exclude them due to the "same number of removed and added lines"
     restriction. Or even if we do try to highlight them, they end up
     not highlighting because of our "don't highlight if the whole line
     would be highlighted" rule.
 
  2. It will find the common prefix and suffix of two lines, and
     consider everything in the middle to be "different". It could
     instead do a real diff of the characters between the two lines and
     find common subsequences. However, the point of the highlight is to
     call attention to a certain area. Even if some small subset of the
     highlighted area actually didn't change, that's OK. In practice it
     ends up being more readable to just have a single blob on the line
     showing the interesting bit.
 
The goal of the script is therefore not to be exact about highlighting
changes, but to call attention to areas of interest without being
visually distracting.  Non-diff lines and existing diff coloration is
preserved; the intent is that the output should look exactly the same as
the input, except for the occasional highlight.
 
Use
---
 
You can try out the diff-highlight program with:
 
---------------------------------------------
git log -p --color | /path/to/diff-highlight
---------------------------------------------
 
If you want to use it all the time, drop it in your $PATH and put the
following in your git configuration:
 
---------------------------------------------
[pager]
	log = diff-highlight | less
	show = diff-highlight | less
	diff = diff-highlight | less
---------------------------------------------
 
 
Color Config
------------
 
You can configure the highlight colors and attributes using git's
config. The colors for "old" and "new" lines can be specified
independently. There are two "modes" of configuration:
 
  1. You can specify a "highlight" color and a matching "reset" color.
     This will retain any existing colors in the diff, and apply the
     "highlight" and "reset" colors before and after the highlighted
     portion.
 
  2. You can specify a "normal" color and a "highlight" color. In this
     case, existing colors are dropped from that line. The non-highlighted
     bits of the line get the "normal" color, and the highlights get the
     "highlight" color.
 
If no "new" colors are specified, they default to the "old" colors. If
no "old" colors are specified, the default is to reverse the foreground
and background for highlighted portions.
 
Examples:
 
---------------------------------------------
# Underline highlighted portions
[color "diff-highlight"]
oldHighlight = ul
oldReset = noul
---------------------------------------------
 
---------------------------------------------
# Varying background intensities
[color "diff-highlight"]
oldNormal = "black #f8cbcb"
oldHighlight = "black #ffaaaa"
newNormal = "black #cbeecb"
newHighlight = "black #aaffaa"
---------------------------------------------
 
 
Bugs
----
 
Because diff-highlight relies on heuristics to guess which parts of
changes are important, there are some cases where the highlighting is
more distracting than useful. Fortunately, these cases are rare in
practice, and when they do occur, the worst case is simply a little
extra highlighting. This section documents some cases known to be
sub-optimal, in case somebody feels like working on improving the
heuristics.
 
1. Two changes on the same line get highlighted in a blob. For example,
   highlighting:
 
----------------------------------------------
-foo(buf, size);
+foo(obj->buf, obj->size);
----------------------------------------------
 
   yields (where the inside of "+{}" would be highlighted):
 
----------------------------------------------
-foo(buf, size);
+foo(+{obj->buf, obj->}size);
----------------------------------------------
 
   whereas a more semantically meaningful output would be:
 
----------------------------------------------
-foo(buf, size);
+foo(+{obj->}buf, +{obj->}size);
----------------------------------------------
 
   Note that doing this right would probably involve a set of
   content-specific boundary patterns, similar to word-diff. Otherwise
   you get junk like:
 
-----------------------------------------------------
-this line has some -{i}nt-{ere}sti-{ng} text on it
+this line has some +{fa}nt+{a}sti+{c} text on it
-----------------------------------------------------
 
   which is less readable than the current output.
 
2. The multi-line matching assumes that lines in the pre- and post-image
   match by position. This is often the case, but can be fooled when a
   line is removed from the top and a new one added at the bottom (or
   vice versa). Unless the lines in the middle are also changed, diffs
   will show this as two hunks, and it will not get highlighted at all
   (which is good). But if the lines in the middle are changed, the
   highlighting can be misleading. Here's a pathological case:
 
-----------------------------------------------------
-one
-two
-three
-four
+two 2
+three 3
+four 4
+five 5
-----------------------------------------------------
 
   which gets highlighted as:
 
-----------------------------------------------------
-one
-t-{wo}
-three
-f-{our}
+two 2
+t+{hree 3}
+four 4
+f+{ive 5}
-----------------------------------------------------
 
   because it matches "two" to "three 3", and so forth. It would be
   nicer as:
 
-----------------------------------------------------
-one
-two
-three
-four
+two +{2}
+three +{3}
+four +{4}
+five 5
-----------------------------------------------------
 
   which would probably involve pre-matching the lines into pairs
   according to some heuristic.