https://sourceforge.net/p/gnuplot/bugs/2844/2025-12-18T00:44:08.640000ZRecent changes to 2844: 6.0.3: The `dt (n1,n2)` syntax makes it hard to use programmatically defined dashtype2025-12-18T00:44:08.640000Z2025-12-18T00:44:08.640000ZEthan Merritthttps://sourceforge.net/u/sfeam/https://sourceforge.netfbfda0fdb8f9e461b87784ae286c2f45e52e8144

<div class="markdown_content"><ul> <li><strong>status</strong>: open --&gt; pending-fixed</li> <li><strong>Group</strong>: --&gt; </li> <li><strong>Priority</strong>: --&gt; </li> </ul></div>

2025-12-14T04:19:04.149000Z2025-12-14T04:19:04.149000ZIlya Zakharevichhttps://sourceforge.net/u/ilya-z/https://sourceforge.netf8a3a215f0aac7ad357d3de9c979838d459de3ce

<div class="markdown_content"><p>I want to use a calculated dashtype, but while <code>dt k</code> works, <code>dt (k+1)</code> does not. (With the message</p> <div class="codehilite"><pre><span></span><code> ^ expecting comma </code></pre></div> <p>)</p> <p>And AFAIK, in other situation complicated expressions in <code>gnuplot</code>⸣s <code>plot</code> commands “should be” (?) put in parentheses. In some cases one can manage with abominations like <code>k&gt;15?1:k==6?2:0</code> — but sometimes the difficulties are not surmountable.</p> <p>Is there a need to prohibit lists of length 1 (as in <code>dt (k+1)</code> — meaning the same as without the parens)?</p> <p>(Knowing fine details of <code>Perl</code> tokenizer) I can see that <code>dt +(k+1)</code> works too. So as a stopgap, one could just document such syntax…</p></div>

2025-12-14T04:19:04.149000Z2025-12-14T04:19:04.149000ZIlya Zakharevichhttps://sourceforge.net/u/ilya-z/https://sourceforge.net55dfb628933e90438af8bea2d4298b924211938a

<div class="markdown_content"><p>Ticket 2844 has been modified: 6.0.3: The <code>dt (n1,n2)</code> syntax makes it hard to use programmatically defined dashtype<br/> Edited By: Ethan Merritt (sfeam)<br/> Status updated: 'open' =&gt; 'pending-fixed'</p></div>