Mercurial > hg
annotate mercurial/pure/bdiff.py @ 29095:3f33831a9202
develwarn: move config gating inside the develwarn function
The config gating is almost always the same and contributor tend to forget it.
We move the logic inside the function. Call site will be updated in later
changeset. We might make the sub config mandatory in the future (once all old
call sites are gone).
author | Pierre-Yves David <pierre-yves.david@ens-lyon.org> |
---|---|
date | Sun, 08 May 2016 10:43:41 +0200 |
parents | 9ab45fbe045e |
children | a8933d992a71 |
rev | line source |
---|---|
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
1 # bdiff.py - Python implementation of bdiff.c |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
2 # |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
3 # Copyright 2009 Matt Mackall <mpm@selenic.com> and others |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
4 # |
8225
46293a0c7e9f
updated license to be explicit about GPL version 2
Martin Geisler <mg@lazybytes.net>
parents:
7944
diff
changeset
|
5 # This software may be used and distributed according to the terms of the |
10263 | 6 # GNU General Public License version 2 or any later version. |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
7 |
27335
c4e3ff497f89
bdiff: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
15530
diff
changeset
|
8 from __future__ import absolute_import |
c4e3ff497f89
bdiff: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
15530
diff
changeset
|
9 |
28389
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
10 import array |
27335
c4e3ff497f89
bdiff: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
15530
diff
changeset
|
11 import difflib |
c4e3ff497f89
bdiff: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
15530
diff
changeset
|
12 import re |
c4e3ff497f89
bdiff: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
15530
diff
changeset
|
13 import struct |
7944
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
14 |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
15 def splitnewlines(text): |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
16 '''like str.splitlines, but only split on newlines.''' |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
17 lines = [l + '\n' for l in text.split('\n')] |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
18 if lines: |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
19 if lines[-1] == '\n': |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
20 lines.pop() |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
21 else: |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
22 lines[-1] = lines[-1][:-1] |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
23 return lines |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
24 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
25 def _normalizeblocks(a, b, blocks): |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
26 prev = None |
14066
14fac6c0536a
pure bdiff: don't use a generator
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
10282
diff
changeset
|
27 r = [] |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
28 for curr in blocks: |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
29 if prev is None: |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
30 prev = curr |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
31 continue |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
32 shift = 0 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
33 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
34 a1, b1, l1 = prev |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
35 a1end = a1 + l1 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
36 b1end = b1 + l1 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
37 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
38 a2, b2, l2 = curr |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
39 a2end = a2 + l2 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
40 b2end = b2 + l2 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
41 if a1end == a2: |
10282
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
10263
diff
changeset
|
42 while (a1end + shift < a2end and |
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
10263
diff
changeset
|
43 a[a1end + shift] == b[b1end + shift]): |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
44 shift += 1 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
45 elif b1end == b2: |
10282
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
10263
diff
changeset
|
46 while (b1end + shift < b2end and |
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
10263
diff
changeset
|
47 a[a1end + shift] == b[b1end + shift]): |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
48 shift += 1 |
14066
14fac6c0536a
pure bdiff: don't use a generator
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
10282
diff
changeset
|
49 r.append((a1, b1, l1 + shift)) |
10282
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
10263
diff
changeset
|
50 prev = a2 + shift, b2 + shift, l2 - shift |
14066
14fac6c0536a
pure bdiff: don't use a generator
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
10282
diff
changeset
|
51 r.append(prev) |
14fac6c0536a
pure bdiff: don't use a generator
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
10282
diff
changeset
|
52 return r |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
53 |
28389
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
54 def _tostring(c): |
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
55 if type(c) is array.array: |
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
56 # this copy overhead isn't ideal |
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
57 return c.tostring() |
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
58 return str(c) |
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
59 |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
60 def bdiff(a, b): |
28389
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
61 a = _tostring(a).splitlines(True) |
9ab45fbe045e
bdiff: (pure) support array.array arrays (issue5130)
timeless <timeless@mozdev.org>
parents:
27335
diff
changeset
|
62 b = _tostring(b).splitlines(True) |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
63 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
64 if not a: |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
65 s = "".join(b) |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
66 return s and (struct.pack(">lll", 0, 0, len(s)) + s) |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
67 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
68 bin = [] |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
69 p = [0] |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
70 for i in a: p.append(p[-1] + len(i)) |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
71 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
72 d = difflib.SequenceMatcher(None, a, b).get_matching_blocks() |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
73 d = _normalizeblocks(a, b, d) |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
74 la = 0 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
75 lb = 0 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
76 for am, bm, size in d: |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
77 s = "".join(b[lb:bm]) |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
78 if am > la or s: |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
79 bin.append(struct.pack(">lll", p[la], p[am], len(s)) + s) |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
80 la = am + size |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
81 lb = bm + size |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
82 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
83 return "".join(bin) |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
84 |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
85 def blocks(a, b): |
7944
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
86 an = splitnewlines(a) |
e9b48afd0e78
pure/bdiff: fix circular import
Matt Mackall <mpm@selenic.com>
parents:
7703
diff
changeset
|
87 bn = splitnewlines(b) |
7703
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
88 d = difflib.SequenceMatcher(None, an, bn).get_matching_blocks() |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
89 d = _normalizeblocks(an, bn, d) |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
90 return [(i, i + n, j, j + n) for (i, j, n) in d] |
9044d3567f6d
pure Python implementation of bdiff.c
Martin Geisler <mg@daimi.au.dk>
parents:
diff
changeset
|
91 |
15530
eeac5e179243
mdiff: replace wscleanup() regexps with C loops
Patrick Mezard <pmezard@gmail.com>
parents:
14066
diff
changeset
|
92 def fixws(text, allws): |
eeac5e179243
mdiff: replace wscleanup() regexps with C loops
Patrick Mezard <pmezard@gmail.com>
parents:
14066
diff
changeset
|
93 if allws: |
eeac5e179243
mdiff: replace wscleanup() regexps with C loops
Patrick Mezard <pmezard@gmail.com>
parents:
14066
diff
changeset
|
94 text = re.sub('[ \t\r]+', '', text) |
eeac5e179243
mdiff: replace wscleanup() regexps with C loops
Patrick Mezard <pmezard@gmail.com>
parents:
14066
diff
changeset
|
95 else: |
eeac5e179243
mdiff: replace wscleanup() regexps with C loops
Patrick Mezard <pmezard@gmail.com>
parents:
14066
diff
changeset
|
96 text = re.sub('[ \t\r]+', ' ', text) |
eeac5e179243
mdiff: replace wscleanup() regexps with C loops
Patrick Mezard <pmezard@gmail.com>
parents:
14066
diff
changeset
|
97 text = text.replace(' \n', '\n') |
eeac5e179243
mdiff: replace wscleanup() regexps with C loops
Patrick Mezard <pmezard@gmail.com>
parents:
14066
diff
changeset
|
98 return text |