python difflib 具有统一上下文格式的字符差异

Question

我需要以 unix 统一 diff 风格显示每行的字符差异。有没有办法使用 difflib 来做到这一点？

我可以分别使用 difflib.unified_diff 和 difflib.Differ() (ndiff) 分别获得“统一差异”和“每行字符差异”，但如何组合它们？

这就是我正在寻找的:

#
# This is difflib.unified
#
>>> print ''.join(difflib.unified_diff('one\ntwo\nthree\n'.splitlines(1), 'ore\ntree\nemu\n'.splitlines(1), 'old', 'new'))
--- old
+++ new
@@ -1,3 +1,3 @@
-one
-two
-three
+ore
+tree
+emu

>>> 

#
# This is difflib.Differ
#
>>> print ''.join(difflib.ndiff('one\ntwo\nthree\n'.splitlines(1), 'ore\ntree\nemu\n'.splitlines(1))),
- one
?  ^
+ ore
?  ^
- two
- three
?  -
+ tree
+ emu
>>> 


#
# I want the merge of above two, something like this...
#
>>> print ''.join(unified_with_ndiff('one\ntwo\nthree\n'.splitlines(1), 'ore\ntree\nemu\n'.splitlines(1))),
--- old
+++ new
@@ -1,3 +1,3 @@
- one
?  ^
+ ore
?  ^
- two
- three
?  -
+ tree
+ emu

>>> 

Answer 1

深入研究 difflib 的源代码后，我自己找到了答案。

'''
# mydifflib.py
@author: Amit Barik
@summary: Overrides difflib.Differ to present the user with unified format (for Python 2.7).

Its basically merging of difflib.unified_diff() and difflib.Differ.compare()
'''

from difflib import SequenceMatcher
from difflib import Differ

class UnifiedDiffer(Differ):
    def unified_diff(self, a, b, fromfile='', tofile='', fromfiledate='',
                 tofiledate='', n=3, lineterm='\n'):
        r"""
        Compare two sequences of lines; generate the resulting delta, in unified
        format

        Each sequence must contain individual single-line strings ending with
        newlines. Such sequences can be obtained from the `readlines()` method
        of file-like objects.  The delta generated also consists of newline-
        terminated strings, ready to be printed as-is via the writeline()
        method of a file-like object.

        Example:

        >>> print ''.join(Differ().unified_diff('one\ntwo\nthree\n'.splitlines(1),
        ...                                'ore\ntree\nemu\n'.splitlines(1)),
        ...                                'old.txt', 'new.txt', 'old-date', 'new-date'),
        --- old.txt    old-date
        +++ new.txt    new-date
        @@ -1,5 +1,5 @@
          context1
        - one
        ?  ^
        + ore
        ?  ^
        - two
        - three
        ?  -
        + tree
        + emu
          context2
        """

        started = False
        for group in SequenceMatcher(None,a,b).get_grouped_opcodes(n):
            if not started:
                fromdate = '\t%s' % fromfiledate if fromfiledate else ''
                todate = '\t%s' % tofiledate if tofiledate else ''
                yield '--- %s%s%s' % (fromfile, fromdate, lineterm)
                yield '+++ %s%s%s' % (tofile, todate, lineterm)
                started = True
            i1, i2, j1, j2 = group[0][1], group[-1][2], group[0][3], group[-1][4]
            yield "@@ -%d,%d +%d,%d @@%s" % (i1+1, i2-i1, j1+1, j2-j1, lineterm)
            for tag, i1, i2, j1, j2 in group:
                if tag == 'replace':
                    for line in a[i1:i2]:
                        g = self._fancy_replace(a, i1, i2, b, j1, j2)
                elif tag == 'equal':
                    for line in a[i1:i2]:
                        g = self._dump(' ', a, i1, i2)
                    if n > 0:
                        for line in g:
                            yield line
                    continue
                elif tag == 'delete':
                    for line in a[i1:i2]:
                        g = self._dump('-', a, i1, i2)
                elif tag == 'insert':
                    for line in b[j1:j2]:
                        g = self._dump('+', b, j1, j2)
                else:
                    raise ValueError, 'unknown tag %r' % (tag,)

                for line in g:
                    yield line


def main():
    # Test
    a ='context1\none\ntwo\nthree\ncontext2\n'.splitlines(1)
    b = 'context1\nore\ntree\nemu\ncontext2\n'.splitlines(1)
    x = UnifiedDiffer().unified_diff(a, b, 'old.txt', 'new.txt', 'old-date', 'new-date', n=1)
    print ''.join(x)

if __name__ == '__main__':          
    main()