<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">Hi,</p>
<br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">an hour ago I read the article <a href="https://www.kdab.com/beware-of-qt-module-wide-includes/">https://www.kdab.com/beware-of-qt-module-wide-includes/</a> and saw that it also applies to the KMyMoney code base. So I made that simple change and compared the compile times on the 5.1 branch of the two versions (ran both twice).</p>
<br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">QtTest (testcases only) - the original version</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">real 0m16,376s</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">user 2m31,903s</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">sys 0m11,080s</p>
<br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">real 0m16,240s</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">user 2m34,316s</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">sys 0m11,436s</p>
<br /><br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">QTest (testcases only) - the improved version</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">real 0m13,602s</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">user 2m4,875s</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">sys 0m10,391s</p>
<br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">real 0m13,626s</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">user 2m5,270s</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">sys 0m10,305s</p>
<br /><br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">Removing that lower-case T gains around 30 seconds of compile time (user). Not too bad. The difference to the 'real' time savings of about 3 seconds is due to the fact that I run this with make -j13 here on 6 cores with hyperthreading.</p>
<br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">-- </p>
<br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">Regards</p>
<br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">Thomas Baumgart</p>
<br /><p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">https://www.signal.org/ Signal, the better WhatsApp</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">-------------------------------------------------------------</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">Faster, better, cheaper - choose two of the above.</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;"> - Old engineering proverb</p>
<p style="margin-top:0;margin-bottom:0;margin-left:0;margin-right:0;">-------------------------------------------------------------</p>
<br /></body>
</html>