在Python中从文件内部删除特定行

Question

I have a list of documents in a file. Basically the TDT2 corpus consisting on both mandarin and english files. I want to keep only the english documents and remove the mandarin ones. Manually doing so would take very long since the file is huge.

The structure looks something like this:

<ONTOPIC topicid=20001 level=YES docno=VOA19980630.1800.3165 fileid=19980630_1800_1900_VOA_ENG comments="NO">
<ONTOPIC topicid=20001 level=BRIEF docno=VOM19980220.0700.0559 fileid=19980220_0700_0800_VOA_MAN comments="NO">
<ONTOPIC topicid=20001 level=YES docno=VOM19980220.0700.1159 fileid=19980220_0700_0800_VOA_MAN comments="NO">

So I want to remove the files which have a 'MAN' in their fileid. How can I do this specific task in Python?

Answer 1

If lines are not written with \\n just remove it from the endswith clause. This will ignore any files which end with MAN comments="NO"> and output out the other files.

out = open('file2.txt','wb')    
for i in open('file.txt'):
    if i.endswith('MAN comments="NO">\n'):
        pass
    else:
        out.write(i)

out.close()

If you are sure 'MAN' will only be part of mandarian ones. Looks a bit cleaner.

out = open('file2.txt','wb')    
for i in open('file.txt'):
    if 'MAN' not in i:
        out.write(i)

out.close()

Answer 2

You can try this:

def start():
    sFile = "source.txt"
    dFile = "results.txt"
    with open(dFile, 'w') as dHandle:
        with open (sFile, "r") as fhandle:
            for fline in fhandle.readlines():
                if "MAN" not in fline:
                    dHandle.write(fline)

start()