<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On 9 July 2014 11:41, Matthias Bussonnier <span dir="ltr"><<a href="mailto:bussonniermatthias@gmail.com" target="_blank">bussonniermatthias@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
But you will piss off user that have users that have data frame that are just slightly too big to fit,<br>
and will be forced to explicitly tell not to truncate.</blockquote></div><br></div><div class="gmail_extra">I think truncation by default is necessary - data frames can easily be far too big to display wholesale. Of course choosing where you start truncating is difficult, but pandas exposes that as a configurable option (and the default displays far more than numpy before it starts truncating).<br>
<br>> you might have a .widget() that show a subpart of the table and fetches the new parts as you scroll<br><br></div><div class="gmail_extra">It would be great to see something like this, but it's also important to have a good static display. Apart from anything else, you can't rely on having a kernel running with that object in the namespace to render the notebook - dynamically loading data from the kernel won't work on nbviewer, for instance.<br>
<br></div><div class="gmail_extra">Thomas<br></div></div>