2017-03-11 9:41 GMT+01:00 Nicolas Cellier <
nicolas.cellier.aka.nice at gmail.com>:

> Yes, it's possible to attach scripts to a .mcz. (preamble, postscript for
> install, same for removal) - see the 'Scripts' on Monticello browser.
>
> If you merge a .mcz directly from the Monticello interface, then there is
> no provision that intermediate scripts will run. Only the lastest will.
>

Err, only the latest will run IF you first install the package OR if the
script did change.


> There might also be dependencies on changes made on other packages.
>
> If you upgrade your image from trunk mecanism ('update squeak') then the
> process is based on MonticelloConfigurationMap (mcm) and what happens is
> that:
> 1) we identify which mcm was last loaded into you image
> 2) we process each newer mcm (well known intermediate points) to sync the
> changes of several packages and exec intermediate scripts
> 3) when the last mcm is reached, we then load any newer .mcz found in
> trunk (based solely on version number).
>
> But this process does not allow for easy cherry picking.
>
> 2017-03-11 9:25 GMT+01:00 Ken Causey <klcausey at gmail.com>:
>
>> Thanks Nicolas.  I've continued looking into the missing
>> removeModalWindow issue and I suspect that it was removed in
>> Morphic-tfel.1310 based on the comment, but I can't find that version to
>> actually confirm.
>>
>> To continue to display my ignorance: Isn't there a mechanism for running
>> a 'script' on installation of a mcz or something like that? I'm thinking
>> there must have been such a script to remove or method the actionMap for
>> PasteUpMorph since the modal window support was deprecated/removed.  And
>> somehow when I load or merge a later version I don't end up running that.
>>
>> Ken
>>
>> On Sat, Mar 11, 2017 at 2:16 AM, Nicolas Cellier <
>> nicolas.cellier.aka.nice at gmail.com> wrote:
>>
>>> Hi Ken,
>>>
>>> 2017-03-11 5:39 GMT+01:00 KenCausey <klcausey at gmail.com>:
>>>
>>>> I can't seem to replicate that on Windows with a 32bit image at least.
>>>> However I found a much more serious issue that appears to have entered
>>>> with
>>>> jr.1315 which is when trying to quit, a debugger pops up with an MNU for
>>>> PasteUpMorph>>removeModalWindow.  Did you not see that?  This seems to
>>>> be
>>>> related to a change in the ShutDown behavior around World but the
>>>> ShutDown
>>>> list is not properly modified.
>>>>
>>>>
>>> No, I did not see that.
>>>
>>>
>>>> Would I have been more correct to submit my change against the latest
>>>> version in the squeak51 Morphic-mt.1296.mcz.
>>>>
>>>>
>>> The best is to submit a change based on head revision in trunk.
>>> Basing the change on any ancestor of head is OK, the newer the better to
>>> avoid possible conflicts.
>>> Basing the changes on a branch is not OK, because we either have to
>>> merge the changes of the whole branch, or cherry pick the diff.
>>>
>>>
>>>
>>>> I'm sorry it has been a long time since I have done any Squeaking and
>>>> to say
>>>> that I'm rusty would be an understatement.
>>>>
>>>>
>>>>
>>> No, no need to apologize, it's me who thought that I could push directly
>>> to trunk from web interface without double checking in an image because the
>>> changes were trivial enough. As if I never was caught before by excess of
>>> confidence, despite my age ;)
>>> Anyway, thanks for contributing and welcome back !
>>>
>>>
>>>
>>>
>>>>
>>>> --
>>>> View this message in context: http://forum.world.st/The-Trun
>>>> k-Morphic-KLC-1318-mcz-tp4938199p4938210.html
>>>> Sent from the Squeak - Dev mailing list archive at Nabble.com.
>>>>
>>>>
>>>
>>>
>>>
>>>
>>
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20170311/cc11bcfb/attachment-0001.html>