Thanks for replying, Michael!
Unfortunately the content probably isn't regular enough for parsing. I tend to
try a little light stream hacking for simple problems like these.
In most cases the bit of text I want
to read varies - I can't search explicitly for "Indoor". But I do know what
precedes and follows the bit of text I want.
Maybe this is just something that old
streams do better than Xstreams? For now I'm just combining old and
new:
substream rest readStream
skipThroughAll: '<tr><td><b>'; upToAll:
'</b>'
That's the beauty of
substreams for me: before if the bold tag was never closed, this would read all
the way to the end of the HTML page, losing all the information from here
onwards. Now the substream created with "ender: '</tr>'" limits the damage
to just this row.
Cheers,
Steve
On 8/7/10 3:17 PM, Steven Kelly wrote:
Hi,
I’m enjoying playing around with Xstreams.
In particular the substreams definitely makes parsing text much more
pleasantly OO. One thing I haven’t figured out yet is how to do things like
skipThroughAll:. E.g. if I want to pick the word “Indoor” out of this
HTML:
<tr><td><b>Indoor</b> </td>
There are good facilities for making a
substream that cutting off the end after “Indoor” (ending: '</b'). But is
there anything for cutting off after the beginning (and making sure the
beginning exists)? I suppose a nice equivalent would be “starting:
'<b>'”, and maybe also starting:ending:, since so often the separator
we’re trying to substream based on has both opening and closing forms – e.g.
XML, brackets, quotes.
It
sounds a lot like you're getting in to parsing here. You might consider using
Xtreams-PEG ?
Besides that, you can explore:
isIndoor := stream
exploring: [(stream read: 6) = 'Indoor'].
Of course in this case you must
be positionable, so if you're on a socket, make sure you first wrap it as a
positionable stream:
stream := socketstream
positioning.
Cheers,
Michael
Locked
