2012/6/24 Benjamin <benjamin.vanryseghem.ph...@gmail.com> > > On Jun 24, 2012, at 3:55 AM, Hernán Morales Durand wrote: > > Thanks for your answers Benjamin. After reading the chapter I should note > the plugin architecture seems very promising, as you've said it, very > different from OB. So to create my own browser should I subclass Nautilus? > which hook methods should I check first? > > > You should create your own plugin I guess :) > In a plugin, you can access the UI, and put another morph instead of the > text area, or whatever :) >
cool, now is clearer to me the scope of plugins > Have a look a MondrianPlugin, it shows a bit what you can do :) > > > > Some usability issues: > > When I do "Find class...", type Object and Ok, it takes some time to > display the class on the browser. Maybe it would be nice to preload first N > methods in the method list? (i'm thinking in performance when browsing > large lists). > When I bring the menu over a class, there is no "browse hierarchy" option, > is this option missing on purpose? > > > This method exists, I just forget to invoke it :) > > I don't know other people, but I use it a lot to browse whole hierarchies. Hernán > Ben > > > Cheers, > > Hernán > > 2012/6/23 Benjamin <benjamin.vanryseghem.ph...@gmail.com> > >> You can found some doc here: >> http://rmod.lille.inria.fr/web/pier/software/Tools-Improvement/NautilusDocumentation >> >> For Pharo 1.4, I try to maintain it a bit, but my focus is on the >> integration of Nautilus in Pharo 2.0 >> >> Ben >> >> On Jun 23, 2012, at 4:32 PM, Hernán Morales Durand wrote: >> >> For Nautilus I've read a review blog post at >> http://marianopeck.wordpress.com/2012/02/26/nautilus-the-reborn-of-the-systembrowser/but >> I didn't found anything else, maybe I'm missing something? >> >> I've installed Nautilus with the following script in a clean Pharo 1.4 >> core >> >> Gofer it >> url: 'http://ss3.gemstone.com/ss/Nautilus'; >> package: 'ConfigurationOfNautilus'; >> load. >> >> (Smalltalk at: #ConfigurationOfNautilus) perform: #loadRelease >> >> but after some random clicks I've received a SubscriptOutOfBounds >> exception. I've attached the stack from the debugger and a screenshot. >> >> Hernán >> >> 2012/6/23 Benjamin <benjamin.vanryseghem.ph...@gmail.com> >> >>> What have you read already ? >>> >>> Ben >>> >>> On Jun 23, 2012, at 7:18 AM, Hernán Morales Durand wrote: >>> >>> Hello Benjamin, >>> >>> 2012/6/22 Benjamin <benjamin.vanryseghem.ph...@gmail.com> >>> >>>> Hello :) >>>> >>>> This is a nice challenge :) >>>> Firstly, you have to know that the model behind Nautilus is really >>>> different that the one behind OB. Let's say nothing in common. >>>> On the other hand, the code pane modification can be made within an >>>> hour (maybe two if you are not confident with morphic, but it doesn't seems >>>> to be the case) using the plugin mecanism (Johan Fabry can testify ^^). >>>> >>> >>> Ok, let us know when more Nautilus documentation is available. >>> >>> >>>> Then, Nautilus is not cross platform, but we have developed Spec with >>>> the goal to use it for Nautilus, and Spec is(will be ^^) cross platform. So >>>> one day, Nautilus will be cross platform. Sadly, I will not be soon (I have >>>> to rewrite the whole ui). >>>> >>>> >>> thanks for the clarification and status update >>> >>> >>> >>>> About the command definition, I am not an OB expert, but I think that >>>> Command are used for menu entries. >>>> In Nautilus the menus definition are based on pragma, so any class can >>>> extend Nautilus menus. >>>> >>>> >>> Yes, commands in OB are used to define menu entries and their behavior. >>> >>> >>> >>>> In a nutshell, what I have seen in the video is doable in roughly a >>>> couple of hours :) >>>> If you decide to give a try to Nautilus, let me know, and I will be >>>> glad to help/answer question/provide pieces of code etc :) >>>> >>>> >>> Thanks for your reply, I will be in touch. >>> >>> Hernán >>> >>> >>> >>>> Ben >>>> >>>> >>>> On Jun 22, 2012, at 8:14 AM, Hernán Morales Durand wrote: >>>> >>>> First let me apologize for asking without properly evaluating first. I >>>> took some time and read about Nautilus which it seems a very nice work. >>>> I've developed a phylogenetics classifier using OmniBrowser, but I wondered >>>> if Nautilus could save me some effort, and I'm planning to test Pharo 1.4 >>>> or 2.0 in some future in which AFAIK OmniBrowser will not be supported >>>> anymore. >>>> >>>> - Some time ago I've modified the OmniBrowser framework to browse >>>> anything I wanted in the "code pane" (the lower pane typically used for >>>> displaying/editing code), as we had a complex OODB it was very useful, see >>>> http://www.youtube.com/watch?v=VxUaOFRHFPk for a demo. However it was >>>> not easy to understand the OB internals and make the modifications. Have >>>> you tried to implement such feature in Nautilus? >>>> - There is documentation for developing browsers with Nautilus? >>>> Something which explains the basic architecture, how it differs from OB? >>>> for example: >>>> --Do you have to define a metagraph? >>>> --May I have multiple definition panels? >>>> -Do you have a mapping for each node type (class?) with each column? >>>> -OB is not cross-UI (it depends on Morphic). Can I use a Nautilus >>>> browser within a web browser? >>>> - Another thing which is really easy is to define Commands in OB, >>>> (although the #isActive logic could become a mess without some care). How >>>> it is different in Nautilus? >>>> >>>> I'm sure I would rewrite my browsers with Nautilus, my key question is >>>> how much of the OB knowledge I can re-use? >>>> >>>> Cheers, >>>> >>>> Hernán >>>> >>>> -- >>>> Hernán Morales >>>> Institute of Veterinary Genetics. >>>> National Scientific and Technical Research Council (CONICET). >>>> La Plata (1900), Buenos Aires, Argentina. >>>> Telephone: +54 (0221) 421-1799. >>>> Internal: 422 >>>> Fax: 425-7980 or 421-1799. >>>> >>>> >>>> >>> > >
