#2 Flush support
Add functionality to Yapter that allows the programmer
to start outputting already parsed pieces of a document
to the browser, if possible. Such a thing could look
like a simple [FLUSH] tag in a template document. When
Yapter comes across such a tag, it invokes the parser
to parse everything above that tag ans flushes it to
the browser, before continuing the further document
parsing.
This could enable browsers to start displaying data
before the whole page is parsed. Should be very useful
for large documents.
The tricky aspects of this tag are, of course, the
'strange' places where you might want to use them. (For
example: Within a block? Just after a block? And what
if a block is repeated a thousand times?) Before
implementing, therefore, this feature should be given
excessive thought.
Logged In: NO
Why would you do this by an [flush] tag? Normally the
programmer decides when he wants to flush the data, so why
not an flush function?
Logged In: YES
user_id=808324
In fact, we've already came to this conclusion ourselves.
And that flush() function is what is being worked on
currently. If you are interested in helping out testing,
please download the latest development snapshot from CVS:
cvs -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/yapter \
login
cvs -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/yapter \
co -P -r flush-branch src
It's still being developed on and currently only supports
flushing in outermost blocks, so flushing blocks within
blocks is impossible yet. Any feedback will be greatly
appreciated.
Logged In: NO
I think so too. A flush() function is more useful than a tag and
I know other systems which work this way too :)