Geoff Talvola <[EMAIL PROTECTED]> wrote:
> I've also run into situations where this would be useful, but I've just
> worked around them. Here's an idea. Ordinarily your writeContent() method
> will look something like:
>
> def writeContent():
> self.writeBeginning()
> self.writeMiddle()
> self.writeEnd()
>
> But suppose you want to re-arrange the order in which these are
> processed. You could do something like:
>
> def writeContent():
> middle = self.capture(self.writeMiddle)
> self.writeBeginning()
> self.write(middle)
> self.writeEnd()
>
> where self.capture is a helper method that causes its argument to be
> executed, but all self.write() and self.writeln() are captured and returned
> as a string instead of being sent out.
>
> I leave the implementation of self.capture() up to the reader :-) Also,
> I'm curious if people think this is a good approach.
It seems useful to me. I implemented it with a new method to response
(so you'd call self.response().capture(func)):
from DummyStreamOut import DummyStreamOut
## ...
def capture(self, func):
oldStrmOut = self._strmOut
self._strmOut = DummyStreamOut()
func()
self._strmOut.commit()
value = self._strmOut._buffer
self._strmOut = oldStrmOut
return value
And a new class:
## DummyStreamOut.py
import string
class DummyStreamOut:
def __init__(self):
self._data = []
self._committed = 0
def write(self, charstr):
self._data.append(charstr)
_needCommit = 0
def flush(self):
pass
def autoCommit(self, value):
pass
def commit(self):
self._committed = 1
self._buffer = string.join(self._data, '')
del self._data
def prepend(self, charstr):
self._data = [charstr] + self._data
def clear(self):
self._data = []
def size(self):
return len(self._buffer)
I didn't entirely understand the interface that response._strmOut was
supposed to have, but this was my best guess.
Ian
_______________________________________________
Webware-discuss mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/webware-discuss
