Using python I would like to know how to delete the first occurrence of a character between two strings in a file

For example if I had some file:

OPEN(
,a
,b
) CLOSE
OPEN(
c
) CLOSE
OPEN(
,d
,e
,f
) CLOSE

I would like to remove the first occurrence of the character ',' following each 'OPEN(' but only if it occurs before the next ') CLOSE' such that the resulting file would look like:

OPEN(
a
,b
) CLOSE
OPEN(
c
) CLOSE
OPEN(
d
,e
,f
) CLOSE

Any thoughts on how I should approach? I have tried using regex, but I don't know how to specify conditions. Could some combination of awk & sed be used?

attempted regex solution:

pattern = r'WITH \(([^)]+),([^)]+)\) AS'
replacement = r'WITH () AS'
sql_content_modified = re.sub(pattern, replacement, sql_content)

ended up solving with something like:

# Read the SQL file
    with open(f'{filename}', 'r') as file:
        content = file.read()
    content_modified = content.replace('(\n,', '(')
    content_modified = re.sub('--<([a-z]*.*[A-Z]*)>', '', content)  # removes the --<*> lines
    # Write the modified content back to the file
    with open(f'{filename}', 'w') as file:
        file.write(content_modified)
    remove_empty_lines_from_file(filename)

    # now do the same but replace "WITH (\n," with "WITH (\n" ...
    with open(f'{filename}', 'r') as file:
        content = file.read()
    content_modified = content.replace('(\n,', '(\n')
    with open(f'{filename}', 'w') as file:
        file.write(content_modified)

In GNU awk with your shown samples only, please try following awk code. Where I am setting RS to (^|\n)OPEN\\(\n,*[^\n]*\n[^)]*\\) *CLOSE as per requirement and then playing around with RT to get the required output.

Here is the Online Demo for used regex for understanding purposes, note awk we need to double escape but in regex site single escape works, so that's why regex looks little different here.

awk -v RS='(^|\n)OPEN\\(\n,*[^\n]*\n[^)]*\\) *CLOSE' '
RT{
  sub(/^\n/,"",RT)
  sub(/,/,"",RT)
  print RT
}
' Input_file

Not very efficient but it those work. (assuming a file named "test.txt" in the same directory as the .py file)

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

newLines = []
flag = False
for index, line in enumerate(lines):
    if(flag):
        if(lines[index+1] != ') CLOSE\n'):
            line = line.strip(',')
        flag = False
    if(line == "OPEN(\n"):
        flag = True
    newLines.append(line)
print(newLines)
with open("test.txt", 'w') as f:
    for line in newLines:
        f.write(line)

This might work for you (GNU sed):

sed '/OPEN/{:a;N;/CLOSE/!ba;s/,//}' file

Match the string OPEN then gather up further lines until one matching CLOSE. Then substitute the first , with nothing.

N.B. If there is no closing string, the remainder of the file will be read then printed once a read on the end-of-file is encounter i.e. the N command is enacted after the $ address.

Simple python script with comments for explanation.

delete_key.txt - input data file

delete_key_formatted.txt - output file with the formatted data

def format_data():
    char_remove = ","
    all_chars = []  # final list
    with open("delete_key.txt") as f:
        cleaned = f.readlines()
        is_open = False
        chars_in_between = []  # temp to store each iterate (open to close or open to open)
        for clean in cleaned:
            if "OPEN" in clean:
                if is_open:  # if we have reached open without close
                    all_chars.extend(chars_in_between)
                    chars_in_between = []
                is_open = True

            '''
            (Optional)We can add (if is_open:) condition to skip all the lines that come without open
            '''
            chars_in_between.append(clean)  # collecting all lines

            if "CLOSE" in clean and is_open:  # if we reached close after hitting open
                is_open = False
                chars_in_between[1] = chars_in_between[1][1:] if chars_in_between[1][0] == char_remove else chars_in_between[1]  # remove the first occurrence of the character ',' following each 'OPEN(' but only if it occurs before the next ') CLOSE'
                all_chars.extend(chars_in_between)  # appending to the current iterate to final list
                chars_in_between = []  # reset the temp current iterate
        if len(chars_in_between) > 0:  # if we reached end of file without close
            all_chars.extend(chars_in_between)

    with open("delete_key_formatted.txt", "w") as f:
        f.writelines(all_chars)


if __name__ == "__main__":
    format_data()

You might use regular expression with zero-length assertions following way, let file.txt content be

OPEN(
,a
,b
) CLOSE
OPEN(
c
) CLOSE
OPEN(
,d
,e
,f
) CLOSE

then

import re
with open('file.txt','r') as f:
    text = f.read()
with open('file.txt','w') as f:
    f.write(re.sub(r'(?<=\(\n),(?=.*CLOSE)', '', text, flags=re.DOTALL))

do change file.txt content to

OPEN(
a
,b
) CLOSE
OPEN(
c
) CLOSE
OPEN(
d
,e
,f
) CLOSE

Explanation: read content of whole file into text, then replace all ,-s which are after bracket-newline (this is positive lookbehind) and before zero-or-more any character followed by CLOSE (this is positive lookahead) with any character including newline (re.DOTALL), then write result to file with same name.
Consult re docs if you want to know more about zero-length assertions available.

(tested in Python 3.10.6)

To preface, I am misunderstanding the following condition.

> "... but only if it occurs before the next ') CLOSE' ..."

Nevertheless, you can use the following to match your expected result.

(?s)(?<=OPEN\(.),

Output

OPEN(
a
,b
) CLOSE
OPEN(
c
) CLOSE
OPEN(
d
,e
,f
) CLOSE