<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<br>
<blockquote type="cite"
cite="mid:CAC_ZHzY56h6dFOgo48CtgZ49BgF4Lc6Jb8yDW-LP66VTpSj8Kg@mail.gmail.com">
<div dir="ltr">
<div class="gmail_quote"># HG changeset patch<br>
# User Thomas De Schampheleire <<a
href="mailto:thomas.de_schampheleire@nokia.com"
target="_blank" moz-do-not-send="true">thomas.de_schampheleire@nokia.com</a>><br>
# Date 1546030612 -3600<br>
# Fri Dec 28 21:56:52 2018 +0100<br>
# Node ID 43b3bbf90a48adfcfb4837655cffee3a939bf48f<br>
# Parent 445d6875c2eefdd8d10676e761cfe1e82581f78a<br>
setup.py: allow Paste 3.0.x<br>
</div>
</div>
</blockquote>
<p><br>
</p>
<p>My main concern for whitelisting new major versions of
dependencies is whether they are sufficiently backwards
compatible. We should try to mitigate that with some amount of
testing and review, and preferably a brief mentioning of it in the
commit message.</p>
<p>What can be said about this upgrade?</p>
<p><br>
</p>
<p>Note that we also *could* upgrade and support:</p>
<p>celery==4.2.1<br>
Markdown==3.0.1<br>
Pygments==2.3.1<br>
pytest==4.0.2<br>
pytest-localserver==0.5.0<br>
Routes==2.4.1<br>
Sphinx==1.8.3</p>
<p>- but I know that some of these break things and are non-trivial
...</p>
<p>/Mads</p>
<br>
</body>
</html>