preprocessing

cjk

Hello,
Can anyone give me some advice about preprocessing.
I'm trying to extract the title from an XML doc where all the element names are the same but each have a unique attribue
e.g
<entrydata columnnumber="0" name="Title">
<text>This is the title</text></entrydata>
<entrydata columnnumber="1" name="url">
<text>http://this is the url</text></entry

Any advice?

thanks
cjk

Migrateduser

If your documents are always structured as shown in your example, you could do something like this (note, this is not a completely fleshed out preprocessor, but contains the key elements and logic):
1. Define some key tokens:

tokens
{
TITLE "qTITLE\q>"
START_TEXT "<text>";
END_TEXT "</text>";
}
2. Create some actions for storing stuff in buffers, etc.
action SET_TITLE
{
settaga("Title");
}
action TITLE
{
collect_view();
appenda();
}

action COLLECT
{
collect_view();
}
action DONE
{
collect_view();
exit();
}



3. Create some states to chain through to grab the title

states
{

state FIND_TITLE
TITLE NOTHING FIND_TITLE
START_TEXT NOTHING COLLECT_TITLE
;


state COLLECT_TITLE
OTHER COLLECT COLLECT_TITLE
END_TEXT SET_TITLE COLLECT_REST
;

state COLLECT_REST
OTHER COLLECT COLLECT_REST
EOS DONE COLLECT_REST
}