使用Sed替换Python文件中的双括号为单括号的技术问询

Got it, let's break down how to solve this problem—replacing those unnecessary double parentheses (like print((...))) with single ones, even when the content spans multiple lines.

First: Single-Line Cases #

For straightforward single-line instances (like print((something)) or print(('hello'))), a simple Sed command will do the trick. We'll use extended regular expressions to target consecutive opening/closing parentheses without messing up legitimate nested calls (like foo(bar(baz))):

sed -E 's/(\()\(/\1/g; s/\)(\))/\1/g' your_python_script.py

How this works:

• s/(\()\(/\1/g: Captures the first opening parenthesis, then replaces two consecutive (( with just one (.
• s/(\))\)/\1/g: Captures the last closing parenthesis, replacing two consecutive )) with just one ).

Handling Multi-Line Strings #

If you have print statements split across lines (like the string concatenation example you mentioned), Sed's default line-by-line processing won't catch the closing )) that's on a different line. We need to tell Sed to read multiple lines into its "pattern space" first:

sed -E -N '
:multi_line_loop
N
# Apply the same parenthesis replacement as before
s/(\()\(/\1/g; s/\)(\))/\1/g
# If we still see a closing parenthesis after a newline, keep looping to read more lines
/\n.*\))/b multi_line_loop
# Print the first line of the pattern space and remove it to process the rest
P
D
' your_python_script.py

Going the Extra Mile: Remove All Levels of Extra Parentheses #

If you have cases with triple (or more) parentheses (like print(((deep_nested)))), you can add a loop to keep replacing until no more double parentheses exist:

sed -E ':cleanup_loop; s/(\()\(/\1/g; s/\)(\))/\1/g; t cleanup_loop' your_python_script.py

The t cleanup_loop command tells Sed to jump back to the cleanup_loop label only if a replacement was made—so it'll keep stripping extra parentheses until none are left.

Safety First: Backup Your File #

Always make a backup before modifying files in place! Use the -i flag with a backup suffix to let Sed handle this:

# Creates a backup file named your_python_script.py.bak, then modifies the original
sed -i.bak -E ':cleanup_loop; s/(\()\(/\1/g; s/\)(\))/\1/g; t cleanup_loop' your_python_script.py

Test It Out #

Let's say you have this code:

print((something))
print(('>> # some text and ' + 'some more text'))
print(('>> # line 1 ' + 
'line 2'))
foo(((bar)))

After running the final command, it'll become:

print(something)
print('>> # some text and ' + 'some more text')
print('>> # line 1 ' + 
'line 2')
foo(bar)

内容的提问来源于stack exchange，提问作者Sam