[thesite] Tip Harvester question (was: [***] Formatting tips )

[Dean Mah]

When I was working on the Perl version I grabbed tips where the tag
appeared as the first non-whitespace characters.  I would also skip
over the part of a message that followed the "--- Original Message
---" text that Outlook tends to put in for quoted messages.  I would
also handle tips within tips, i.e., if you had something like:

<tip>
<tip>
</tip>
</tip>

it would pull it out as two separate tips, which would happen from time
to time.

Dean


Seth Bienek writes:

> The tip parser that I wrote looks for a line break followed by  <tip> and
> includes everything between the closing bracket and the first bracket of
> </tip>.  If it doesn't find the closing tag, it throws the whole thing out.
> 
> The drawback to this approach is that if someone replies and their email
> software doesn't support "quoting", the tip would go in twice.
> 
> I have put some thought into this, but have not been able to come up with a
> real solution other than a review process, or doing exact comparisons of
> every tip (which seems like it would inhibit performance, and I'm not sure
> if you can even compare two whole text fields..)
> 
> I'm open to suggestions..