Committed to stHub/Magritte3 & MetaRepoForPharo40:
- New version 3.3 (semantic minor version bump for added functionality). For
pharo4.x, add Magritte-GT package and add it to the default group.
- Declare 3.3 as stable
- Declare 3.3 baseline as development
-----
Cheers,
Sean
--
View this message in context: http://forum.world.st/Version-3-3-tp4817750.html
Sent from the Magritte, Pier and Related Tools mailing list archive at Nabble.com.
Should or could the MADescriptionMorph hierarchy use/inherit-fro MAVisitor?
"Magritte Meta-Described Web Application Development" says "Custom
Rendering... Sometimes other layouts are more convenient, for example the
widgets should be laid out from left to right with the labels on top...
Magritte allows one to define one’s own builder by subclassing a Visitor and
overriding some of the methods used to place the user interface elements".
With the current implementation, it seems one would have to hack the
MADescriptionMorph classes themselves for that kind of change, but if a
visitor was used, they could just supply a custom visitor class, no?
-----
Cheers,
Sean
--
View this message in context: http://forum.world.st/Visitors-Underused-tp4815294.html
Sent from the Magritte, Pier and Related Tools mailing list archive at Nabble.com.
Anyone have it working? I tried to load in Pharo 4.0, but stopped on missing
JsJSONParser class
-----
Cheers,
Sean
--
View this message in context: http://forum.world.st/Magritte-Magic-tp4813780.html
Sent from the Magritte, Pier and Related Tools mailing list archive at Nabble.com.
I want to create a UI from Magritte descriptions where changes are accepted
into the model object immediately per field, instead of going through the
momento. Any idea the best strategy to go about this? Thanks.
-----
Cheers,
Sean
--
View this message in context: http://forum.world.st/UI-vs-Form-tp4815286.html
Sent from the Magritte, Pier and Related Tools mailing list archive at Nabble.com.
My other Morphic changes have not caused any obvious push back, so I'd like
to go ahead with my suggestion from a while back...
Motivation: When I first used Magritte, I was terribly confused that
#asMorph does not actually return a morph, and requires "object asMorph
morph" to get the actual morph
(http://forum.world.st/Customizing-Magritte-Morphic-Forms-tt4251518.html#a42…)
Solution: Rename classes messages to reveal intention
1) #asMagritteMorph
a. change #asMagritteMorph to something like maMorphicContainer (returns a
presenter/builder, not a Morph)
b. if we want #asMagritteMorph, have it return the actual morph (i.e.
"self maMorphicContainer morph")
2) MAContainerMorph and friends -> MAMorphicContainer - these are not
Morphs. I was actually astonished when I saw they did not inherit from
Morph.
If there is no opposition, I'll make these changes shortly.
BTW: what is the policy re non-backward compatible changes? How should they
be documented? In Pharo, methods are deprecated for a version, but I didn't
want to do that because I assumed Magritte should run on other Smalltalks.
Although I see Squeak has #deprecated:, so maybe that would be safe...
-----
Cheers,
Sean
--
View this message in context: http://forum.world.st/Proposal-to-rename-some-Morphic-classes-tp4815311.html
Sent from the Magritte, Pier and Related Tools mailing list archive at Nabble.com.
Can I be added to the SmalltalkHub Magritte team as `DaleHenrichs`? I've
got a GemStone bugfix that fixes three tests that are getting errors in
Magritte3.2 ...
Thanks in advance,
Dale
Hello,
In magritte3, in relation description class, I can see a #reference:
method. However I don't read explanation about its use, and so far never
need to use it.
What is it and how is it used?
Thanks
Hilaire
--
Dr. Geo - http://drgeo.eu
iStoa - http://istoa.drgeo.eu
There was a, now outdated, Magritte-GLORP integration, to generate
GLORP descriptors based on Magritte metadata.
But as I say it is outdated, and it probably won't work with the
latest Magritte version (3.x) that uses descriptors on the instance
side.
Regards!
Esteban A. Maringolo
2015-01-17 12:34 GMT-03:00 David Carlos Manuelda <stormbyte(a)gmail.com>:
> I've read in soem tutorial (don't remember which right now) that magritte
> have support for data persistency.
>
> If I did not understood it bad, it would mean that it can autogenerate SQL
> queries from descriptions so they don't need to be coded manually into the
> database class methods, but just could not find any info and/or examples
> regarding it.
>
> Can someone point me in the right direction, or tall me if I understood it
> bad?
>
> _______________________________________________
> seaside mailing list
> seaside(a)lists.squeakfoundation.org
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
> Hi Markus,
>
> Yes, we removed this in bootstrap, because it was getting too full. But on the other hand: we do want to show in the tutorial, so we can properly demo the options of QCMagritte …
>
> We added it back in the tutorial, but it still needs bootstrap styling… maybe we can take a look later this week. We changed the parent of the bootstrap overview from simpleOverview to Overview and fixed a bug in the renderButtonsOn: in QCBToManyComponent.
>
> Cheers,
> Diego
Markus wrote:
>Am I not rtft correctly again? The entry "Overview" says:
>
>"For adding an extra column you should go to the report options ... and
>click on choose columns."
Hmm, I updated the code but not the text. The report options were not
looking good in the bootstrap style.
>Where do I find the report options?
Let me take a look.
Stephan
