<div dir="ltr"><br><div class="gmail_extra"><br><br><div class="gmail_quote">2014-05-02 9:11 GMT-07:00 Eliot Miranda <span dir="ltr"><<a href="mailto:eliot.miranda@gmail.com" target="_blank">eliot.miranda@gmail.com</a>></span>:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi Colin, Hi All,<div><br></div><div> if in 4.5 one tries to evaluate a doit somewhere in a verdant transcript, one containing much content, then too much logging will occur to the changes file. Instead of the selected doit being logged, the entire contents of the transcript up to the end of the doit will be logged, and that can cause parse errors when recovering changes.</div>
<div><br></div><div>So how best to solve this? Here are some ideas and I wanted to ask people's opinion</div><div><br></div><div>- have CompilationCue remember the position of the source stream and instead of using "source contents" in streamToLog use "source position; upToEnd"</div>
<div><br></div><div>- have Parser/Scanner remember the characters parsed in a buffer and take them from there</div><div><br></div><div>- change ReadWriteStream to remember its initial position so that when one says</div>
<div>
(ReadWriteStream on: (String withAll: ((0 to: 255) collect: [:n| n asCharacter]))</div><div> from: $A asInteger + 1 to: $Z asInteger + 1) contents<br clear="all"><div>one gets 'ABCDEFGHIJKLMNOPQRSTUVWXYZ'</div>
<div>instead of</div><div><div> ' <span style="white-space:pre-wrap"> </span></div><div> </div><div> !"#$%&''()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ'</div>
</div><div><br></div><div>IMO the last option is the right thing to do but it feels like a big change.</div><span class="HOEnZb"><font color="#888888"><div></div></font></span></div></div></blockquote><div><br></div><div>
and indeed implementing the latter doesn't break anything in the standard test suite. I'm committing the fix plus some tests. </div></div>-- <br>best,<div>Eliot</div>
</div></div>