Hello python-list,
I have a question about the match objects that are returned from the
match() method of compiled regular expression objects from the 're'
module. To parse Postscript T1 fonts that were disassembled into
plaintext, I came up with the following code:
import re
rmoveto = re.compile('^\s*(-?\d+)\s+(-?\d+)\s+rmoveto$')
rlineto = re.compile('^\s*(-?\d+)\s+(-?\d+)\s+rlineto$')
# ... other expressions with up to six paren groups
f = open(filename, 'r')
for line in f.readlines():
m = rmoveto.match(line)
if m:
x = x + int(m.group(1))
y = y + int(m.group(2))
glyph.append(('move', (x, y)))
continue
m = rlineto.match(line)
if m:
x = x + int(m.group(1))
y = y + int(m.group(2))
glyph.append(('line', (x, y)))
continue
# ... and so forth for the other expressions
Now here is my question: is there a simple way to join the following
two python code lines:
m = rmoveto.match(line)
if m:
into one single line like in the following:
if rmoveto.match(line):
x = x + int(rmoveto.group(1))
y = y + int(rmoveto.group(2))
glyph.append(('move', (x, y)))
elif rlineto.match(line):
# ...
The above syntax does not work because the compiled regular expression
object rmoveto doesn't provide a method called group(), as it comes
from module 're'. The obsolete package 'regex' did provide this, if I
read the docs correctly.
As a workaround, I also tried to use a nested function like this:
def match(expr):
m = expr.match(line)
return m
if match(rmoveto):
x = x + int(m.group(1))
# ...
This approach failed because the match function has its own local m,
so it didn't update the outer m. I couldn't use 'global m' either
because the whole thing, including the outer m, happens to be inside a
function, too.
How do you people handle this?
Thanks for your time,
Johann C. Rocholl
--
http://mail.python.org/mailman/listinfo/python-list
