<div class="markdown_content"><ul>
<li><strong>status</strong>: open --> wont-fix</li>
<li><strong>Group</strong>: --> TODO-3.5</li>
<li><strong>Comment</strong>:</li>
</ul>
<p>This would be nice but is too low a priority to remain on the list for now.</p>
<hr />
<p><strong> <a class="alink strikethrough" href="http://sourceforge.net/p/cwb/feature-requests/4">[feature-requests:#4]</a> Matching against word lists in a more flexible way</strong></p>
<p><strong>Status:</strong> wont-fix<br />
<strong>Group:</strong> TODO-3.5<br />
<strong>Created:</strong> Fri Jan 26, 2007 02:57 PM UTC by Manuel Kountz<br />
<strong>Last Updated:</strong> Fri Jan 26, 2007 02:58 PM UTC<br />
<strong>Owner:</strong> nobody</p>
<p>Assume someone wants to query for concatenative tokens like "vor+zu+gehen", where prefixes and verb stems both come from a list, while the "zu" is a fixed part of the query. It seems that CQP does not support anything like searching for something which is formed from an element of a word list and some fixed part, or which is formed by combining two word lists (other than through the API).</p>
<p>One solution would be to treat the word list as a regular expression in any case, transforming it into a transducer disjoining all elements in the list, and eventually combining this transducer with another one which implements a RE from a query. (AFAIK this is the way SFST treats lexicon files.) This should also speed up word list matching, which is said to be comparatively slow.</p>
<hr />
<p>Sent from sourceforge.net because cwb@sslmit.unibo.it is subscribed to <a href="https://sourceforge.net/p/cwb/feature-requests">https://sourceforge.net/p/cwb/feature-requests/</a></p>
<p>To unsubscribe from further messages, a project admin can change settings at <a href="https://sourceforge.net/p/cwb/admin/feature-requests/options.">https://sourceforge.net/p/cwb/admin/feature-requests/options.</a> Or, if this is a mailing list, you can unsubscribe from the mailing list.</p></div>