Alexandre,
As I am about to embark on a major set of validation runs, I thought it would be a good idea to use red/yellow/green icons to signify the validation status of each configuration (and each version ... since validation issues are related to particular versions) I will also add some support for looking at the validation issues ... But the main reason for this message is that to do this (easily for me) I am monkeying with the way that the window is being built .... the widgetSpecs don't change but for now at least I am not going to use the ToolBuilder, because it does not seem possible to easily add icons to list items, whereas I can already see my way clearly to getting icons displayed without using tool builder ... I am also interested in having MetacelloBrowser run in squeak and I am suspicious that the "polymorph overrides" are Pharo specific ... of course I don't know that for certain... I think having the immediate feedback on every configuration is a good thing and with all of the validation work I will be doing, I think a) putting it into the tool and then b) using it myself to identify and fix validation issues is a very good thing ... I am warning you just in case you end up heading into similar territory with the work you have planned ... Dale |
Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it!
MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. Cheers, Alexandre On 26 Feb 2011, at 16:20, Dale Henrichs wrote: > Alexandre, > > As I am about to embark on a major set of validation runs, I thought it would be a good idea to use red/yellow/green icons to signify the validation status of each configuration (and each version ... since validation issues are related to particular versions) I will also add some support for looking at the validation issues ... > > But the main reason for this message is that to do this (easily for me) I am monkeying with the way that the window is being built .... the widgetSpecs don't change but for now at least I am not going to use the ToolBuilder, because it does not seem possible to easily add icons to list items, whereas I can already see my way clearly to getting icons displayed without using tool builder ... > > I am also interested in having MetacelloBrowser run in squeak and I am suspicious that the "polymorph overrides" are Pharo specific ... of course I don't know that for certain... > > I think having the immediate feedback on every configuration is a good thing and with all of the validation work I will be doing, I think a) putting it into the tool and then b) using it myself to identify and fix validation issues is a very good thing ... > > I am warning you just in case you end up heading into similar territory with the work you have planned ... > > Dale -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
On Feb 26, 2011, at 2:03 PM, Alexandre Bergel wrote: > Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... > > MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... > > Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... Dale |
>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it!
> > Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... 100% Agree >> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. > > It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. >> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. > > Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( > > I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... Maybe Doru can say a bit more. I will commit the OB-like commands in a couple of hours. Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >> >> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... > > 100% Agree > >>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >> >> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... > > Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... > >>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >> >> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >> >> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... > > > Maybe Doru can say a bit more. > > I will commit the OB-like commands in a couple of hours. I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( I've got some more changes myself so I'll get them committed pronto... Dale |
On 27 Feb 2011, at 02:41, Dale Henrichs wrote: > > On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >> >>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>> >>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>> >>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >> >> >> Maybe Doru can say a bit more. >> >> I will commit the OB-like commands in a couple of hours. > > I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( The only dependency the Morphic rendering of Glamour has is to Polymorph - because it uses the widgets and the theme. If Squeak would be able to load Polymorph, then it should run, but I am not aware of anyone investing in this direction. Glamour does indeed have other dependencies to Magritte and Mondrian, but those are optional. If you load the Core and Morphic groups, you only get the Morphic/Polymorph dependency. Cheers, Doru > I've got some more changes myself so I'll get them committed pronto... > > Dale -- www.tudorgirba.com "If you interrupt the barber while he is cutting your hair, you will end up with a messy haircut." |
On Feb 27, 2011, at 10:06 AM, Tudor Girba wrote: > > On 27 Feb 2011, at 02:41, Dale Henrichs wrote: > >> > > The only dependency the Morphic rendering of Glamour has is to Polymorph - because it uses the widgets and the theme. If Squeak would be able to load Polymorph, then it should run, but I am not aware of anyone investing in this direction. > And as Polymorph is a *patch* and not a modular package, this would be exactly the wrong directin (polymorph needs to be merged into Morphic and cleaned up, not made loadable in Squeak) In Squeak, ToolBuilder is the solution to everything. So someone should port Glamour to toolbuilder for Squeak (not for Pharo, though) Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD. |
In reply to this post by Dale Henrichs
Hi Dale,
MetacelloBrowser-AlexandreBergel.58 contains all the commands implemented using a mechanism similar to OB. If you want to implement a new command, the expression "MBCommand newCommand" will be useful. I've merged MetacelloBrowser-AlexandreBergel.58 and your last version MetacelloBrowser-dkh.62 ConfigurationOfMetacelloBrowser defines version 1.45 that loads the resulting package: MetacelloBrowser-AlexandreBergel.63 The package now contains several categories (Core, Test, Commands, ...). I just remarked, when I create a new version, all the modified packages are saved using the toolkit facilities. But no comments are requested? I am continuing to work on the commands. Things need to be cleaned up. Cheers, Alexandre On 26 Feb 2011, at 22:41, Dale Henrichs wrote: > > On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: > >>>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >>> >>> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... >> >> 100% Agree >> >>>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >>> >>> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... >> >> Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. > > Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) > > BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... > >> >>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>> >>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>> >>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >> >> >> Maybe Doru can say a bit more. >> >> I will commit the OB-like commands in a couple of hours. > > I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( > > I've got some more changes myself so I'll get them committed pronto... > > Dale -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
I just produced 1.46 with some cleaning. Maybe you want to double check that I did not miss something. Not all the commands are under unit tests.
Chers, Alexandre On 27 Feb 2011, at 11:49, Alexandre Bergel wrote: > Hi Dale, > > MetacelloBrowser-AlexandreBergel.58 contains all the commands implemented using a mechanism similar to OB. If you want to implement a new command, the expression "MBCommand newCommand" will be useful. > I've merged MetacelloBrowser-AlexandreBergel.58 and your last version MetacelloBrowser-dkh.62 > > ConfigurationOfMetacelloBrowser defines version 1.45 that loads the resulting package: MetacelloBrowser-AlexandreBergel.63 > > The package now contains several categories (Core, Test, Commands, ...). > > I just remarked, when I create a new version, all the modified packages are saved using the toolkit facilities. But no comments are requested? > > I am continuing to work on the commands. Things need to be cleaned up. > > Cheers, > Alexandre > > > > On 26 Feb 2011, at 22:41, Dale Henrichs wrote: > >> >> On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >> >>>>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >>>> >>>> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... >>> >>> 100% Agree >>> >>>>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >>>> >>>> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... >>> >>> Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. >> >> Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) >> >> BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... >> >>> >>>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>>> >>>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>>> >>>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >>> >>> >>> Maybe Doru can say a bit more. >>> >>> I will commit the OB-like commands in a couple of hours. >> >> I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( >> >> I've got some more changes myself so I'll get them committed pronto... >> >> Dale > > -- > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: > Alexandre Bergel http://www.bergel.eu > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > > > > > > -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
Alexandre,
I noticed that two new versions were added again ... were you testing again? If you use the 'save modified packages' menu item the mcz files a committed and a comment is added to the configuration, but the metacello version number isn't incremented ... Dale On Feb 27, 2011, at 7:04 AM, Alexandre Bergel wrote: > I just produced 1.46 with some cleaning. Maybe you want to double check that I did not miss something. Not all the commands are under unit tests. > > Chers, > Alexandre > > > On 27 Feb 2011, at 11:49, Alexandre Bergel wrote: > >> Hi Dale, >> >> MetacelloBrowser-AlexandreBergel.58 contains all the commands implemented using a mechanism similar to OB. If you want to implement a new command, the expression "MBCommand newCommand" will be useful. >> I've merged MetacelloBrowser-AlexandreBergel.58 and your last version MetacelloBrowser-dkh.62 >> >> ConfigurationOfMetacelloBrowser defines version 1.45 that loads the resulting package: MetacelloBrowser-AlexandreBergel.63 >> >> The package now contains several categories (Core, Test, Commands, ...). >> >> I just remarked, when I create a new version, all the modified packages are saved using the toolkit facilities. But no comments are requested? >> >> I am continuing to work on the commands. Things need to be cleaned up. >> >> Cheers, >> Alexandre >> >> >> >> On 26 Feb 2011, at 22:41, Dale Henrichs wrote: >> >>> >>> On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >>> >>>>>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >>>>> >>>>> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... >>>> >>>> 100% Agree >>>> >>>>>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >>>>> >>>>> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... >>>> >>>> Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. >>> >>> Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) >>> >>> BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... >>> >>>> >>>>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>>>> >>>>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>>>> >>>>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >>>> >>>> >>>> Maybe Doru can say a bit more. >>>> >>>> I will commit the OB-like commands in a couple of hours. >>> >>> I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( >>> >>> I've got some more changes myself so I'll get them committed pronto... >>> >>> Dale >> >> -- >> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >> Alexandre Bergel http://www.bergel.eu >> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >> >> >> >> >> >> > > -- > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: > Alexandre Bergel http://www.bergel.eu > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > > > > > |
This makes me think that we should rethink the behavior of +Version again ...
Dale On Feb 27, 2011, at 8:27 AM, Dale Henrichs wrote: > Alexandre, > > I noticed that two new versions were added again ... were you testing again? If you use the 'save modified packages' menu item the mcz files a committed and a comment is added to the configuration, but the metacello version number isn't incremented ... > > Dale > > On Feb 27, 2011, at 7:04 AM, Alexandre Bergel wrote: > >> I just produced 1.46 with some cleaning. Maybe you want to double check that I did not miss something. Not all the commands are under unit tests. >> >> Chers, >> Alexandre >> >> >> On 27 Feb 2011, at 11:49, Alexandre Bergel wrote: >> >>> Hi Dale, >>> >>> MetacelloBrowser-AlexandreBergel.58 contains all the commands implemented using a mechanism similar to OB. If you want to implement a new command, the expression "MBCommand newCommand" will be useful. >>> I've merged MetacelloBrowser-AlexandreBergel.58 and your last version MetacelloBrowser-dkh.62 >>> >>> ConfigurationOfMetacelloBrowser defines version 1.45 that loads the resulting package: MetacelloBrowser-AlexandreBergel.63 >>> >>> The package now contains several categories (Core, Test, Commands, ...). >>> >>> I just remarked, when I create a new version, all the modified packages are saved using the toolkit facilities. But no comments are requested? >>> >>> I am continuing to work on the commands. Things need to be cleaned up. >>> >>> Cheers, >>> Alexandre >>> >>> >>> >>> On 26 Feb 2011, at 22:41, Dale Henrichs wrote: >>> >>>> >>>> On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >>>> >>>>>>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >>>>>> >>>>>> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... >>>>> >>>>> 100% Agree >>>>> >>>>>>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >>>>>> >>>>>> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... >>>>> >>>>> Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. >>>> >>>> Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) >>>> >>>> BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... >>>> >>>>> >>>>>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>>>>> >>>>>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>>>>> >>>>>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >>>>> >>>>> >>>>> Maybe Doru can say a bit more. >>>>> >>>>> I will commit the OB-like commands in a couple of hours. >>>> >>>> I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( >>>> >>>> I've got some more changes myself so I'll get them committed pronto... >>>> >>>> Dale >>> >>> -- >>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>> Alexandre Bergel http://www.bergel.eu >>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>> >>> >>> >>> >>> >>> >> >> -- >> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >> Alexandre Bergel http://www.bergel.eu >> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >> >> >> >> >> > |
In reply to this post by Dale Henrichs
No, it is okay I think. I save often. At each enhancement I realize in the code, I create a new version. Even if 1 line has changed.
Alexandre On 27 Feb 2011, at 13:27, Dale Henrichs wrote: > Alexandre, > > I noticed that two new versions were added again ... were you testing again? If you use the 'save modified packages' menu item the mcz files a committed and a comment is added to the configuration, but the metacello version number isn't incremented ... > > Dale > > On Feb 27, 2011, at 7:04 AM, Alexandre Bergel wrote: > >> I just produced 1.46 with some cleaning. Maybe you want to double check that I did not miss something. Not all the commands are under unit tests. >> >> Chers, >> Alexandre >> >> >> On 27 Feb 2011, at 11:49, Alexandre Bergel wrote: >> >>> Hi Dale, >>> >>> MetacelloBrowser-AlexandreBergel.58 contains all the commands implemented using a mechanism similar to OB. If you want to implement a new command, the expression "MBCommand newCommand" will be useful. >>> I've merged MetacelloBrowser-AlexandreBergel.58 and your last version MetacelloBrowser-dkh.62 >>> >>> ConfigurationOfMetacelloBrowser defines version 1.45 that loads the resulting package: MetacelloBrowser-AlexandreBergel.63 >>> >>> The package now contains several categories (Core, Test, Commands, ...). >>> >>> I just remarked, when I create a new version, all the modified packages are saved using the toolkit facilities. But no comments are requested? >>> >>> I am continuing to work on the commands. Things need to be cleaned up. >>> >>> Cheers, >>> Alexandre >>> >>> >>> >>> On 26 Feb 2011, at 22:41, Dale Henrichs wrote: >>> >>>> >>>> On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >>>> >>>>>>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >>>>>> >>>>>> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... >>>>> >>>>> 100% Agree >>>>> >>>>>>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >>>>>> >>>>>> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... >>>>> >>>>> Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. >>>> >>>> Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) >>>> >>>> BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... >>>> >>>>> >>>>>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>>>>> >>>>>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>>>>> >>>>>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >>>>> >>>>> >>>>> Maybe Doru can say a bit more. >>>>> >>>>> I will commit the OB-like commands in a couple of hours. >>>> >>>> I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( >>>> >>>> I've got some more changes myself so I'll get them committed pronto... >>>> >>>> Dale >>> >>> -- >>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>> Alexandre Bergel http://www.bergel.eu >>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>> >>> >>> >>> >>> >>> >> >> -- >> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >> Alexandre Bergel http://www.bergel.eu >> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >> >> >> >> >> > > -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
In reply to this post by Dale Henrichs
Why, have you noticed a version save more than once?
Alexandre On 27 Feb 2011, at 13:38, Dale Henrichs wrote: > This makes me think that we should rethink the behavior of +Version again ... > > Dale > > On Feb 27, 2011, at 8:27 AM, Dale Henrichs wrote: > >> Alexandre, >> >> I noticed that two new versions were added again ... were you testing again? If you use the 'save modified packages' menu item the mcz files a committed and a comment is added to the configuration, but the metacello version number isn't incremented ... >> >> Dale >> >> On Feb 27, 2011, at 7:04 AM, Alexandre Bergel wrote: >> >>> I just produced 1.46 with some cleaning. Maybe you want to double check that I did not miss something. Not all the commands are under unit tests. >>> >>> Chers, >>> Alexandre >>> >>> >>> On 27 Feb 2011, at 11:49, Alexandre Bergel wrote: >>> >>>> Hi Dale, >>>> >>>> MetacelloBrowser-AlexandreBergel.58 contains all the commands implemented using a mechanism similar to OB. If you want to implement a new command, the expression "MBCommand newCommand" will be useful. >>>> I've merged MetacelloBrowser-AlexandreBergel.58 and your last version MetacelloBrowser-dkh.62 >>>> >>>> ConfigurationOfMetacelloBrowser defines version 1.45 that loads the resulting package: MetacelloBrowser-AlexandreBergel.63 >>>> >>>> The package now contains several categories (Core, Test, Commands, ...). >>>> >>>> I just remarked, when I create a new version, all the modified packages are saved using the toolkit facilities. But no comments are requested? >>>> >>>> I am continuing to work on the commands. Things need to be cleaned up. >>>> >>>> Cheers, >>>> Alexandre >>>> >>>> >>>> >>>> On 26 Feb 2011, at 22:41, Dale Henrichs wrote: >>>> >>>>> >>>>> On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >>>>> >>>>>>>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >>>>>>> >>>>>>> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... >>>>>> >>>>>> 100% Agree >>>>>> >>>>>>>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >>>>>>> >>>>>>> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... >>>>>> >>>>>> Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. >>>>> >>>>> Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) >>>>> >>>>> BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... >>>>> >>>>>> >>>>>>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>>>>>> >>>>>>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>>>>>> >>>>>>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >>>>>> >>>>>> >>>>>> Maybe Doru can say a bit more. >>>>>> >>>>>> I will commit the OB-like commands in a couple of hours. >>>>> >>>>> I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( >>>>> >>>>> I've got some more changes myself so I'll get them committed pronto... >>>>> >>>>> Dale >>>> >>>> -- >>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>>> Alexandre Bergel http://www.bergel.eu >>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>>> >>>> >>>> >>>> >>>> >>>> >>> >>> -- >>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>> Alexandre Bergel http://www.bergel.eu >>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>> >>> >>> >>> >>> >> > > -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
In reply to this post by Alexandre Bergel-5
A new metacello version or a new mcz?
I commit often as well, but don't feel that it is necessary to create new Metacello versions at that rate.... Dale On Feb 27, 2011, at 2:01 PM, Alexandre Bergel wrote: > No, it is okay I think. I save often. At each enhancement I realize in the code, I create a new version. Even if 1 line has changed. > > Alexandre > > > On 27 Feb 2011, at 13:27, Dale Henrichs wrote: > >> Alexandre, >> >> I noticed that two new versions were added again ... were you testing again? If you use the 'save modified packages' menu item the mcz files a committed and a comment is added to the configuration, but the metacello version number isn't incremented ... >> >> Dale >> >> On Feb 27, 2011, at 7:04 AM, Alexandre Bergel wrote: >> >>> I just produced 1.46 with some cleaning. Maybe you want to double check that I did not miss something. Not all the commands are under unit tests. >>> >>> Chers, >>> Alexandre >>> >>> >>> On 27 Feb 2011, at 11:49, Alexandre Bergel wrote: >>> >>>> Hi Dale, >>>> >>>> MetacelloBrowser-AlexandreBergel.58 contains all the commands implemented using a mechanism similar to OB. If you want to implement a new command, the expression "MBCommand newCommand" will be useful. >>>> I've merged MetacelloBrowser-AlexandreBergel.58 and your last version MetacelloBrowser-dkh.62 >>>> >>>> ConfigurationOfMetacelloBrowser defines version 1.45 that loads the resulting package: MetacelloBrowser-AlexandreBergel.63 >>>> >>>> The package now contains several categories (Core, Test, Commands, ...). >>>> >>>> I just remarked, when I create a new version, all the modified packages are saved using the toolkit facilities. But no comments are requested? >>>> >>>> I am continuing to work on the commands. Things need to be cleaned up. >>>> >>>> Cheers, >>>> Alexandre >>>> >>>> >>>> >>>> On 26 Feb 2011, at 22:41, Dale Henrichs wrote: >>>> >>>>> >>>>> On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >>>>> >>>>>>>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >>>>>>> >>>>>>> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... >>>>>> >>>>>> 100% Agree >>>>>> >>>>>>>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >>>>>>> >>>>>>> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... >>>>>> >>>>>> Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. >>>>> >>>>> Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) >>>>> >>>>> BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... >>>>> >>>>>> >>>>>>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>>>>>> >>>>>>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>>>>>> >>>>>>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >>>>>> >>>>>> >>>>>> Maybe Doru can say a bit more. >>>>>> >>>>>> I will commit the OB-like commands in a couple of hours. >>>>> >>>>> I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( >>>>> >>>>> I've got some more changes myself so I'll get them committed pronto... >>>>> >>>>> Dale >>>> >>>> -- >>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>>> Alexandre Bergel http://www.bergel.eu >>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>>> >>>> >>>> >>>> >>>> >>>> >>> >>> -- >>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>> Alexandre Bergel http://www.bergel.eu >>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>> >>> >>> >>> >>> >> >> > > -- > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: > Alexandre Bergel http://www.bergel.eu > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > > > > > |
In reply to this post by Alexandre Bergel-5
Yes ... there were several metacello versions in a row before I sent the mail and the mcz file was incremented by one each time as well...
Dale On Feb 27, 2011, at 2:02 PM, Alexandre Bergel wrote: > Why, have you noticed a version save more than once? > > Alexandre > > > On 27 Feb 2011, at 13:38, Dale Henrichs wrote: > >> This makes me think that we should rethink the behavior of +Version again ... >> >> Dale >> >> On Feb 27, 2011, at 8:27 AM, Dale Henrichs wrote: >> >>> Alexandre, >>> >>> I noticed that two new versions were added again ... were you testing again? If you use the 'save modified packages' menu item the mcz files a committed and a comment is added to the configuration, but the metacello version number isn't incremented ... >>> >>> Dale >>> >>> On Feb 27, 2011, at 7:04 AM, Alexandre Bergel wrote: >>> >>>> I just produced 1.46 with some cleaning. Maybe you want to double check that I did not miss something. Not all the commands are under unit tests. >>>> >>>> Chers, >>>> Alexandre >>>> >>>> >>>> On 27 Feb 2011, at 11:49, Alexandre Bergel wrote: >>>> >>>>> Hi Dale, >>>>> >>>>> MetacelloBrowser-AlexandreBergel.58 contains all the commands implemented using a mechanism similar to OB. If you want to implement a new command, the expression "MBCommand newCommand" will be useful. >>>>> I've merged MetacelloBrowser-AlexandreBergel.58 and your last version MetacelloBrowser-dkh.62 >>>>> >>>>> ConfigurationOfMetacelloBrowser defines version 1.45 that loads the resulting package: MetacelloBrowser-AlexandreBergel.63 >>>>> >>>>> The package now contains several categories (Core, Test, Commands, ...). >>>>> >>>>> I just remarked, when I create a new version, all the modified packages are saved using the toolkit facilities. But no comments are requested? >>>>> >>>>> I am continuing to work on the commands. Things need to be cleaned up. >>>>> >>>>> Cheers, >>>>> Alexandre >>>>> >>>>> >>>>> >>>>> On 26 Feb 2011, at 22:41, Dale Henrichs wrote: >>>>> >>>>>> >>>>>> On Feb 26, 2011, at 3:08 PM, Alexandre Bergel wrote: >>>>>> >>>>>>>>> Excellent! I also think it is a good idea to have something that indicate configurations that are okay. Even though, I think it is better that it is impossible to build an invalid configuration. But, if this helps you in your migration and fixing effort, then go for it! >>>>>>>> >>>>>>>> Agreed, but until we reach that goal, I think that the constant feedback from validation is necessary... >>>>>>> >>>>>>> 100% Agree >>>>>>> >>>>>>>>> MetacelloBrowser is done by using the GUI mini-framework offered by Monticello. I find it easy to use. However, if we go for a new framework, then I will be happy to do so. >>>>>>>> >>>>>>>> It wasn't a big change in how the ui is built just using a different builder so that I could easily add the icons. At the moment it isn't clear whether it is possible to use icons in morphic on Squeak the way they are used in Pharo so portability will have to suffer in the short term ... >>>>>>> >>>>>>> Yes. Maybe we could have on the left hand side two sections (-- valid --, -- invalid --) or something like this. >>>>>> >>>>>> Well, I'm pretty sure that OB in Squeak has the icons, so I think it just boils down to a little bit of work:) >>>>>> >>>>>> BTW, the icons make finding and fixing validation errors real easy! A red dot next to the config or version indicates that there are validation issues that need to be addressed ... clicking on the icon brings up a list of validation issues, browsing the version brings up an editor on the config version ... when they're all green you're done ... >>>>>> >>>>>>> >>>>>>>>> Doru suggested to use Glamour. I like the idea. Maybe we could have a version based on Glamour on some point. I see that we could have two different and separate favor of MetacelloBrowser. A bit similar than with the system browser, we have the standard ST-80 browser and OB. >>>>>>>> >>>>>>>> Yeah, for GemStone I will have to have an OmniBrowser versions of the MetacelloBrowser, but I will wait until the interface settles down to take that step:( >>>>>>>> >>>>>>>> I love Glamour but it suffers from the same problem as OB in that it depends upon a lot of stuff not in the core (at least it used to) which is the advantage of the pure morphic-based tool ... >>>>>>> >>>>>>> >>>>>>> Maybe Doru can say a bit more. >>>>>>> >>>>>>> I will commit the OB-like commands in a couple of hours. >>>>>> >>>>>> I assume Glamour also runs on Squeak? Even though there doesn't seem to be a lot of visible support for Metacello from certain areas of the Squeak community, I still get the feeling that there are a large number of developers using Metacello on Squeak and I really want them to be able benefit from having a MetacelloBrowser ... tool support is critical but I really don't have enough cycles to worry about more than one implementation that works everywhere, except GemStone:( >>>>>> >>>>>> I've got some more changes myself so I'll get them committed pronto... >>>>>> >>>>>> Dale >>>>> >>>>> -- >>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>>>> Alexandre Bergel http://www.bergel.eu >>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>> >>>> -- >>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>>> Alexandre Bergel http://www.bergel.eu >>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>>> >>>> >>>> >>>> >>>> >>> >> >> > > -- > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: > Alexandre Bergel http://www.bergel.eu > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > > > > > |
Free forum by Nabble | Edit this page |