Python equivalent of Perl's $/

[attn.steven.kuo at gmail.com]

On Aug 19, 11:13 am, John K Masters <johnmast... at oxtedonline.net>
wrote:
> I am currently working my way through Jeffrey Friedl's book Mastering
> Regular Expressions. Great book apart from the fact it uses Perl for the
> examples.
>
> One particular expression that interests me is '$/ = ".\n"' which,
> rather than splitting a file into lines, splits on a period-newline
> boundary. Combined with Perl's 'while (<>)' construct this seems a great
> way to process the files I am interested in.
>
> Without wishing to start a flame war, is there a way to do this in Python?
>



import StringIO

text = """\
To mimic Perl's input record separator in
Python, you can use a generator.
And a substring test.
Perhaps something like the following
is what you wanted.
"""

mockfile = StringIO.StringIO(text)

def genrecords(mockfile, sep=".\n"):
    buffer = ""
    while True:
        while sep in buffer:
            idx = buffer.find(sep) + len(sep)
            yield buffer[:idx]
            buffer = buffer[idx:]
        rl = mockfile.readline()
        if rl == "":
            break
        else:
            buffer = '%s%s' % (buffer, rl)
    yield buffer
    raise StopIteration

for record in genrecords(mockfile):
    print "READ:", record


--
Hope this helps,
Steven