if one wants to keep the start and end lines/keywords while extracting the lines between 2 strings.
Please find below the code snippet that I used to extract sql statements from a shell script
def process_lines(in_filename, out_filename, start_kw, end_kw):
try:
inp = open(in_filename, 'r', encoding='utf-8', errors='ignore')
out = open(out_filename, 'w+', encoding='utf-8', errors='ignore')
except FileNotFoundError as err:
print(f"File {in_filename} not found", err)
raise
except OSError as err:
print(f"OS error occurred trying to open {in_filename}", err)
raise
except Exception as err:
print(f"Unexpected error opening {in_filename} is", repr(err))
raise
else:
with inp, out:
copy = False
for line in inp:
# first IF block to handle if the start and end on same line
if line.lstrip().lower().startswith(start_kw) and line.rstrip().endswith(end_kw):
copy = True
if copy: # keep the starts with keyword
out.write(line)
copy = False
continue
elif line.lstrip().lower().startswith(start_kw):
copy = True
if copy: # keep the starts with keyword
out.write(line)
continue
elif line.rstrip().endswith(end_kw):
if copy: # keep the ends with keyword
out.write(line)
copy = False
continue
elif copy:
# write
out.write(line)
if __name__ == '__main__':
infile = "/Users/ag29266/Downloads/pims/BTEQ_PI_FIN_PI_MBR_RISK1.sh"
outfile = f"{infile}.sql"
statement_start_list = ['database', 'create', 'insert', 'delete', 'update', 'merge', 'delete']
statement_end = ";"
process_lines(infile, outfile, tuple(statement_start_list), statement_end)
