如何替换文件中的单词？

Question

In test.txt:

rt : objective
tr350rt : objective
rtrt : objective
@username : objective
@user_1236 : objective
@254test!! : objective
@test : objective
#15 : objective

My codes:

import re
file3 = 'C://Users/Desktop/test.txt'
rfile3 = open(file3).read()
for altext in rfile3.split("\n"):
    saltext = altext.split("\t")
    for saltword in saltext:
        ssaltword = saltword.split(" ")
        if re.search(r'^rt$', ssaltword[0]):
        print ssaltword[0], ssaltword[2]
        testreplace = open(file3, 'w').write(rfile3.replace(ssaltword[0], ""))
        if re.search(r'^@\w', ssaltword[0]):
            print ssaltword[0], ssaltword[2]
        testreplace = open(file3, 'w').write(rfile3.replace(ssaltword[0], ""))

I got:

 : objective
tr350 : objective
 : objective
@username : objective
@user_1236 : objective
@254test!! : objective
 : objective
#15 : objective

I am trying to replace only "rt" and all @ with space

But from my codes all "rt" were replaced and only one @ was replaced.

I would like to get:

 : objective
tr350rt : objective
rtrt : objective
 : objective
 : objective
 : objective
 : objective
#15 : objective

Any suggestion?

Answer 1

I think regex is overkill here:

with open("test.txt") as in_fp, open("test2.txt", "w") as out_fp:
    for line in in_fp:
        ls = line.split()
        if ls and (ls[0].startswith("@") or ls[0] == "rt"):
            line = line.replace(ls[0], "", 1)
        out_fp.write(line)

produces

localhost-2:coding $ cat test2.txt 
 : objective
tr350rt : objective
rtrt : objective
 : objective
 : objective
 : objective
 : objective
#15 : objective

Note that I've also changed it not to overwrite the original.

Edit: if you really want to overwrite the original in-place, then I'd read the whole thing into memory first:

with open("test.txt") as fp:
    lines = fp.readlines()

with open("test.txt", "w") as out_fp:
    for line in lines:
        ls = line.split()
        if ls and (ls[0].startswith("@") or ls[0] == "rt"):
            line = line.replace(ls[0], "", 1)
        out_fp.write(line)

Answer 2

import re
with open("test.txt") as infile:
    text = infile.read()
    newtext = re.sub(r"(?m)^(?:rt\b|@\w+)(?=\s*:)", " ", text)

Explanation:

(?m)      # Turn on multiline mode
^         # Match start of line
(?:       # Either match...
 rt\b     # rt (as a complete word
|         # or
 @\w+     # @ followed by an alphanumeric "word"
)         # End of alternation
(?=\s*:)  # Assert that a colon follows (after optional whitespace)

Answer 3

Try this,

import os

mydict = {"@":'',"rt":''}

filepath = 'C://Users/Desktop/test.txt'
s = open(filepath).read()
for k, v in mydict.iteritems():
    s = s.replace(k, v)
f = open(filepath, 'w')
f.write(s)
f.close()

Answer 4

Not even necessary to use regex here:

with open("test.txt") as file:
    lines = file.readlines()
    for line in lines:
        if (line.startswith("@") and ":" in line) or line.startswith("rt :"):
            line = " :" + line.split(":", 1)[1]