<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Sat, Dec 6, 2014 at 5:53 PM, Chris Muller <span dir="ltr"><<a href="mailto:asqueaker@gmail.com" target="_blank">asqueaker@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Oh yes that's the best source; the original. Maybe it can be rolled<br>
into as part of a new .changes / .sources for the 4.6 / 5.0 release.<br>
I liked the last thing Eliot said about it (a while back) that we<br>
could keep all history, just get rid of the "leaves". Does anyone<br>
remember the status of #condenseSources and #condenseChanges? Do they<br>
just work or is there some issue which was the reason we haven't done<br>
it in so long?<br></blockquote><div> </div><div>I believe they still work. I also have changes that work that implement getting rid of the leaves. I hear that there's better engineered code in Pharo too.</div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="HOEnZb"><div class="h5"><br>
On Sat, Dec 6, 2014 at 3:32 PM, Bert Freudenberg <<a href="mailto:bert@freudenbergs.de">bert@freudenbergs.de</a>> wrote:<br>
> Actually we just need to take a sources/changes file from before 11/20/2005 and compare all methods to the current ones ...<br>
><br>
> - Bert -<br>
><br>
>> On 06.12.2014, at 22:20, Chris Muller <<a href="mailto:asqueaker@gmail.com">asqueaker@gmail.com</a>> wrote:<br>
>><br>
>> Yes, I guess I didn't realize we had overwritten initials because I<br>
>> remember doing the FixUnderscores improvement myself to preserve the<br>
>> initials. Wait..<br>
>><br>
>> (CompiledMethod allInstances collect: [ : e | e timeStamp ] as: Bag)<br>
>> sortedCounts first: 10<br>
>><br>
>> Dang. So, yeah, it looks like quite a few method's original<br>
>> authorship information got trampled. Unfortunately, the 'mc history'<br>
>> function can only show what was actually _loaded_ into that DB. IOW,<br>
>> we have to have the .mcz's for the versions that were BEFORE 2/4/2006<br>
>> 20:41. The oldest one we have in trunk is<br>
>><br>
>> Name: Sound-md.6<br>
>> Author: md<br>
>> Time: 21 April 2006, 5:05:36 pm<br>
>> UUID: 498cd464-d148-11da-a5e8-000d933a223c<br>
>> Ancestors: Sound-md.5<br>
>><br>
>> which carries the same version of that method.<br>
>><br>
>> But at least we can identify the worst-offending methods (the list<br>
>> above), so that maybe they could be extracted from an old image or<br>
>> maybe Bob's history thing??<br>
>><br>
>><br>
>> On Sat, Dec 6, 2014 at 2:20 PM, David T. Lewis <<a href="mailto:lewis@mail.msen.com">lewis@mail.msen.com</a>> wrote:<br>
>>> On Fri, Dec 05, 2014 at 12:23:09PM +0100, Bert Freudenberg wrote:<br>
>>>> On 05.12.2014, at 04:11, tim Rowledge <<a href="mailto:tim@rowledge.org">tim@rowledge.org</a>> wrote:<br>
>>>>><br>
>>>>> (Stephane added explicitly since the code has his initials)<br>
>>>><br>
>>>> We can't rely on the initials unfortunately - this is from when we converted underscore assignments to colon-equals, which originally did not preserve initials.<br>
>>>><br>
>>>> It would actually be a very useful project for someone to fix this: write a snippet that goes through every method, compares it to its previous version (using Chris's MC history thingy for example), and if the only change is that ':=' replaced '_' then restore the previous time stamp.<br>
>>>><br>
>>>> - Bert -<br>
>>><br>
>>><br>
>>> This is a really good idea, so I am replying with a new subject line in<br>
>>> hopes of encouraging follow up.<br>
>>><br>
>>> Dave<br>
>>><br>
>>><br>
>><br>
><br>
><br>
><br>
><br>
><br>
><br>
<br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">best,<div>Eliot</div></div>
</div></div>