比较python中的两个文件报告差异

Question

我有两个名为“hosts”的文件（在不同的目录中）

我想用python比较它们，看它们是否是IDENTICAL。如果它们不相同，我想在屏幕上打印差异。

到目前为止，我已尝试过这个

hosts0 = open(dst1 + "/hosts","r") 
hosts1 = open(dst2 + "/hosts","r")

lines1 = hosts0.readlines()

for i,lines2 in enumerate(hosts1):
    if lines2 != lines1[i]:
        print "line ", i, " in hosts1 is different \n"
        print lines2
    else:
        print "same"

但是当我跑这个时，我得到了

File "./audit.py", line 34, in <module>
  if lines2 != lines1[i]:
IndexError: list index out of range

这意味着其中一个主机的线路数量多于另一个主机。 有没有更好的方法来比较2个文件并报告差异？

Answer 1

import difflib

lines1 = '''
dog
cat
bird
buffalo
gophers
hound
horse
'''.strip().splitlines()

lines2 = '''
cat
dog
bird
buffalo
gopher
horse
mouse
'''.strip().splitlines()

# Changes:
# swapped positions of cat and dog
# changed gophers to gopher
# removed hound
# added mouse

for line in difflib.unified_diff(lines1, lines2, fromfile='file1', tofile='file2', lineterm=''):
    print line

输出以下内容：

--- file1
+++ file2
@@ -1,7 +1,7 @@
+cat
 dog
-cat
 bird
 buffalo
-gophers
-hound
+gopher
 horse
+mouse

这个差异为你提供了上下文 - 周围的行，以帮助清楚文件是如何不同的。你可以在这里看到两次“猫”，因为它从“狗”下方移除并添加到它上面。

您可以使用n = 0删除上下文。

for line in difflib.unified_diff(lines1, lines2, fromfile='file1', tofile='file2', lineterm='', n=0):
    print line

输出：

--- file1
+++ file2
@@ -0,0 +1 @@
+cat
@@ -2 +2,0 @@
-cat
@@ -5,2 +5 @@
-gophers
-hound
+gopher
@@ -7,0 +7 @@
+mouse

但现在它充满了“@@”行，告诉你文件中的位置已经改变了。让我们删除额外的行以使其更具可读性。

for line in difflib.unified_diff(lines1, lines2, fromfile='file1', tofile='file2', lineterm='', n=0):
    for prefix in ('---', '+++', '@@'):
        if line.startswith(prefix):
            break
    else:
        print line

给我们这个输出：

+cat
-cat
-gophers
-hound
+gopher
+mouse

现在你想要它做什么？ 如果忽略所有删除的行，则不会看到“hound”被删除。 如果你很高兴只显示文件的添加内容，那么你可以这样做：

diff = difflib.unified_diff(lines1, lines2, fromfile='file1', tofile='file2', lineterm='', n=0)
lines = list(diff)[2:]
added = [line[1:] for line in lines if line[0] == '+']
removed = [line[1:] for line in lines if line[0] == '-']

print 'additions:'
for line in added:
    print line
print
print 'additions, ignoring position'
for line in added:
    if line not in removed:
        print line

对外输出：

additions:
cat
gopher
mouse

additions, ignoring position:
gopher
mouse

您现在可以说，有两种方法可以“打印两个文件的差异”，因此如果您需要更多帮助，则需要非常具体。

Answer 2

difflib库对此非常有用，它来自标准库。我喜欢统一的差异格式。

http://docs.python.org/2/library/difflib.html#difflib.unified_diff

import difflib
import sys

with open('/tmp/hosts0', 'r') as hosts0:
    with open('/tmp/hosts1', 'r') as hosts1:
        diff = difflib.unified_diff(
            hosts0.readlines(),
            hosts1.readlines(),
            fromfile='hosts0',
            tofile='hosts1',
        )
        for line in diff:
            sys.stdout.write(line)

输出：

--- hosts0
+++ hosts1
@@ -1,5 +1,4 @@
 one
 two
-dogs
 three

这是一个忽略某些线条的狡猾版本。 可能有边缘情况不起作用，并且肯定有更好的方法来做到这一点，但也许它足以满足您的目的。

import difflib
import sys

with open('/tmp/hosts0', 'r') as hosts0:
    with open('/tmp/hosts1', 'r') as hosts1:
        diff = difflib.unified_diff(
            hosts0.readlines(),
            hosts1.readlines(),
            fromfile='hosts0',
            tofile='hosts1',
            n=0,
        )
        for line in diff:
            for prefix in ('---', '+++', '@@'):
                if line.startswith(prefix):
                    break
            else:
                sys.stdout.write(line[1:])

Answer 3

hosts0 = open("C:path\\a.txt","r")
hosts1 = open("C:path\\b.txt","r")

lines1 = hosts0.readlines()

for i,lines2 in enumerate(hosts1):
    if lines2 != lines1[i]:
        print "line ", i, " in hosts1 is different \n"
        print lines2
    else:
        print "same"

以上代码对我有用。你能指出你面临的错误吗？

Answer 4

您可以添加条件语句。如果你的数组超出了索引，那么打破并打印文件的其余部分。

Answer 5

import difflib
f=open('a.txt','r')  #open a file
f1=open('b.txt','r') #open another file to compare
str1=f.read()
str2=f1.read()
str1=str1.split()  #split the words in file by default through the spce
str2=str2.split()
d=difflib.Differ()     # compare and just print
diff=list(d.compare(str2,str1))
print '\n'.join(diff)