input.txt
A(0,1,2)
...
B(A,3)
...
C(B,4,5)
If the first parameter of a function is not equal 0 but corresponding to a function name, I want to replace it with all of the corresponding function's parameters (eg to replace the first parameter 'A' in function B above with all parameters of function A). That is to expect
output.txt
A(0,1,2)
...
B(0,1,2,3)
...
C(0,1,2,3,4,5)
How can we do this with awk/sed or python?
EDIT:
One idea I have is to store the function name as variables and its parameters as values in bash. In python, we may use dict, and consider function names as keys, and its parameters as values. The implementation is not that easy.
Awk
awk -F'[()]' '
$2 !~ /^0,/ {
split($2, a, /,/)
sub(/^[^,]+/, val[a[1]], $2)
}
{
val[$1] = $2
print $1 "(" $2 ")"
}
' input.txt > output.txt
Where sub(/^[^,]+/, val[a[1]], $2)
is used to match the first parameter in $2
and replace it with the value of val[a[1]]
which is defined by the execution of val[$1] = $2
on previous lines.
Here's a solution in Python:
import re
with open('input.txt') as f:
data = f.read()
data = [line.strip() for line in data.split('\n') if line]
sets, output = {}, open('output.txt', 'w')
for line in data:
if line == '...':
output.write(line + '\n')
continue
sets[line[0]] = line[2:-1]
output.write(line[0] + '(')
for char in line[2:-1]:
if re.match(r'[\d,]', char):
output.write(char)
else:
output.write(sets[char])
output.write(')\n')
output.close()
Let lines be the lines of the input file. The following code will work if all parameters are integers or a functionname
funcs = {}
for line in lines:
match = re.search( '(.*)\((.*)\)', line)
if not match:
raise RuntimeError('Line does not match expectation')
function_name = match.group(1)
parameters = map(str.strip, match.group(2).split(','))
parameter_list = []
for parameter in parameters:
try:
parameter_list.append(int(parameter))
except ValueError:
parameter_list.extend( funcs.get(parameter, []) )
funcs[function_name] = parameter_list
for func_name, paras in sorted(funcs.items()):
print '{function}({parameters})'.format(
function=func_name,
parameters=', '.join(map(str, paras))
)
There are probably a ton of ways to do this but I think this is a simple way to do what you want.
import re
import sys
def convertLine(line):
if re.match("^\\w{1}\(.*\)$", line) is None:
return line
retVal = re.sub( "A", "0,1,2",line[1:])
retVal = re.sub( "B", "0,1,2,3",retVal)
retVal = re.sub( "C", "0,1,2,3,4,5",retVal)
return line[0:1]+retVal
def main():
for line in sys.stdin.read().splitlines():
print convertLine(line)
if __name__ == "__main__":
main()
usage:
python ReplaceProg.py < input.txt
if your file is like this
A(0,1,2)
B(A,3)
C(B,4,5)
using python:
f = open('inpu_file.txt').readlines()
f[0] = f[0].strip()
for i,x in enumerate(f):
if i > 0:
f[i]=re.sub(f[i-1][0],",".join(re.findall('\d+',f[i-1])),x).strip()
print f
output:
['A(0,1,2)', 'B(0,1,2,3)', 'C(0,1,2,3,4,5)']
i don't understand that ... in every alternate line, if its there tell me i can edit the code for that.
Kinda long but more modular:
import re
def build_dict(fobj):
d = dict()
for line in fobj:
match = re.match('^(\w)\((.*)\)', line)
fname = match.group(1)
fargs = match.group(2)
d[fname] = replace(fargs, d)
fobj.seek(0) # Reset cursor to start of file
return d
def replace(s, d):
for each in d:
if each in s:
s = s.replace(each, d[each])
return s
def split_paren(s):
index = s.index('(')
return s[:index], s[index:]
def write_replace(fobj, d):
outname = fobj.name[:-4] + '.out'
outfile = open(outname, 'w')
for line in fobj:
first, second = split_paren(line)
second = replace(second, d)
outfile.write(first + second)
outfile.close()
if __name__ == '__main__':
with open('test.txt', 'r') as f:
d = build_dict(f)
write_replace(f, d)
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.