Show More
| @@ -5,6 +5,7 import difflib | |||||
| 5 | import re |
|
5 | import re | |
| 6 | from optparse import OptionParser |
|
6 | from optparse import OptionParser | |
| 7 | from mercurial.bdiff import bdiff, blocks |
|
7 | from mercurial.bdiff import bdiff, blocks | |
|
|
8 | from mercurial.mdiff import bunidiff | |||
| 8 |
|
9 | |||
| 9 | VERSION="0.2" |
|
10 | VERSION="0.2" | |
| 10 | usage = "usage: %prog [options] file1 file2" |
|
11 | usage = "usage: %prog [options] file1 file2" | |
| @@ -23,127 +24,6 if not args: | |||||
| 23 | parser.print_help() |
|
24 | parser.print_help() | |
| 24 | sys.exit(1) |
|
25 | sys.exit(1) | |
| 25 |
|
26 | |||
| 26 | # somewhat self contained replacement for difflib.unified_diff |
|
|||
| 27 | # t1 and t2 are the text to be diffed |
|
|||
| 28 | # l1 and l2 are the text broken up into lines |
|
|||
| 29 | # header1 and header2 are the filenames for the diff output |
|
|||
| 30 | # context is the number of context lines |
|
|||
| 31 | # showfunc enables diff -p output |
|
|||
| 32 | # ignorews ignores all whitespace changes in the diff |
|
|||
| 33 | def bunidiff(t1, t2, l1, l2, header1, header2, context=3, showfunc=False, |
|
|||
| 34 | ignorews=False): |
|
|||
| 35 | def contextend(l, len): |
|
|||
| 36 | ret = l + context |
|
|||
| 37 | if ret > len: |
|
|||
| 38 | ret = len |
|
|||
| 39 | return ret |
|
|||
| 40 |
|
||||
| 41 | def contextstart(l): |
|
|||
| 42 | ret = l - context |
|
|||
| 43 | if ret < 0: |
|
|||
| 44 | return 0 |
|
|||
| 45 | return ret |
|
|||
| 46 |
|
||||
| 47 | def yieldhunk(hunk, header): |
|
|||
| 48 | if header: |
|
|||
| 49 | for x in header: |
|
|||
| 50 | yield x |
|
|||
| 51 | (astart, a2, bstart, b2, delta) = hunk |
|
|||
| 52 | aend = contextend(a2, len(l1)) |
|
|||
| 53 | alen = aend - astart |
|
|||
| 54 | blen = b2 - bstart + aend - a2 |
|
|||
| 55 |
|
||||
| 56 | func = "" |
|
|||
| 57 | if showfunc: |
|
|||
| 58 | # walk backwards from the start of the context |
|
|||
| 59 | # to find a line starting with an alphanumeric char. |
|
|||
| 60 | for x in xrange(astart, -1, -1): |
|
|||
| 61 | t = l1[x] |
|
|||
| 62 | if funcre.match(t): |
|
|||
| 63 | func = ' ' + t[:40] |
|
|||
| 64 | break |
|
|||
| 65 |
|
||||
| 66 | yield "@@ -%d,%d +%d,%d @@%s\n" % (astart + 1, alen, |
|
|||
| 67 | bstart + 1, blen, func) |
|
|||
| 68 | for x in delta: |
|
|||
| 69 | yield x |
|
|||
| 70 | for x in xrange(a2, aend): |
|
|||
| 71 | yield ' ' + l1[x] |
|
|||
| 72 |
|
||||
| 73 | header = [ "--- %s\t\n" % header1, "+++ %s\t\n" % header2 ] |
|
|||
| 74 |
|
||||
| 75 | if showfunc: |
|
|||
| 76 | funcre = re.compile('\w') |
|
|||
| 77 | if ignorews: |
|
|||
| 78 | wsre = re.compile('[ \t]') |
|
|||
| 79 |
|
||||
| 80 | # bdiff.blocks gives us the matching sequences in the files. The loop |
|
|||
| 81 | # below finds the spaces between those matching sequences and translates |
|
|||
| 82 | # them into diff output. |
|
|||
| 83 | # |
|
|||
| 84 | diff = blocks(t1, t2) |
|
|||
| 85 | hunk = None |
|
|||
| 86 | for i in xrange(len(diff)): |
|
|||
| 87 | # The first match is special. |
|
|||
| 88 | # we've either found a match starting at line 0 or a match later |
|
|||
| 89 | # in the file. If it starts later, old and new below will both be |
|
|||
| 90 | # empty and we'll continue to the next match. |
|
|||
| 91 | if i > 0: |
|
|||
| 92 | s = diff[i-1] |
|
|||
| 93 | else: |
|
|||
| 94 | s = [0, 0, 0, 0] |
|
|||
| 95 | delta = [] |
|
|||
| 96 | s1 = diff[i] |
|
|||
| 97 | a1 = s[1] |
|
|||
| 98 | a2 = s1[0] |
|
|||
| 99 | b1 = s[3] |
|
|||
| 100 | b2 = s1[2] |
|
|||
| 101 | old = l1[a1:a2] |
|
|||
| 102 | new = l2[b1:b2] |
|
|||
| 103 |
|
||||
| 104 | # bdiff sometimes gives huge matches past eof, this check eats them, |
|
|||
| 105 | # and deals with the special first match case described above |
|
|||
| 106 | if not old and not new: |
|
|||
| 107 | continue |
|
|||
| 108 |
|
||||
| 109 | if ignorews: |
|
|||
| 110 | wsold = wsre.sub('', "".join(old)) |
|
|||
| 111 | wsnew = wsre.sub('', "".join(new)) |
|
|||
| 112 | if wsold == wsnew: |
|
|||
| 113 | continue |
|
|||
| 114 |
|
||||
| 115 | astart = contextstart(a1) |
|
|||
| 116 | bstart = contextstart(b1) |
|
|||
| 117 | prev = None |
|
|||
| 118 | if hunk: |
|
|||
| 119 | # join with the previous hunk if it falls inside the context |
|
|||
| 120 | if astart < hunk[1] + context + 1: |
|
|||
| 121 | prev = hunk |
|
|||
| 122 | astart = hunk[1] |
|
|||
| 123 | bstart = hunk[3] |
|
|||
| 124 | else: |
|
|||
| 125 | for x in yieldhunk(hunk, header): |
|
|||
| 126 | yield x |
|
|||
| 127 | # we only want to yield the header if the files differ, and |
|
|||
| 128 | # we only want to yield it once. |
|
|||
| 129 | header = None |
|
|||
| 130 | if prev: |
|
|||
| 131 | # we've joined the previous hunk, record the new ending points. |
|
|||
| 132 | hunk[1] = a2 |
|
|||
| 133 | hunk[3] = b2 |
|
|||
| 134 | delta = hunk[4] |
|
|||
| 135 | else: |
|
|||
| 136 | # create a new hunk |
|
|||
| 137 | hunk = [ astart, a2, bstart, b2, delta ] |
|
|||
| 138 |
|
||||
| 139 | delta[len(delta):] = [ ' ' + x for x in l1[astart:a1] ] |
|
|||
| 140 | delta[len(delta):] = [ '-' + x for x in old ] |
|
|||
| 141 | delta[len(delta):] = [ '+' + x for x in new ] |
|
|||
| 142 |
|
||||
| 143 | if hunk: |
|
|||
| 144 | for x in yieldhunk(hunk, header): |
|
|||
| 145 | yield x |
|
|||
| 146 |
|
||||
| 147 | # simple utility function to put all the |
|
27 | # simple utility function to put all the | |
| 148 | # files from a directory tree into a dict |
|
28 | # files from a directory tree into a dict | |
| 149 | def buildlist(names, top): |
|
29 | def buildlist(names, top): | |
General Comments 0
You need to be logged in to leave comments.
Login now