2010/2/9 Mike Ramirez <gufym...@gmail.com>: > On Tuesday 09 February 2010 13:09:08 Brice Leroy wrote: >> django-admin startproject creates an empty folder... >> >> startproject almost = mkdir >> >> This will generate your models, modelForm, views, templates, urls, and >> soon customized forms and permissions >> > > That's what the web interface does? > > Yes startproject gives you empty files so you can add in your own code, with > the minimal funs. > > For me, I don't see your project as a big advantage, except maybe two minutes > of time where I don't need to do things like: > > class MyModel(models.Model): > ... > > > I don't know, thought about your approach before, but think it's reduntant. > Just because each project requires it's own data and model field attributes > and form funs... And I think django already does a great job of genericizing
It does provide you a skeleton that you're supposed to adapt. I wanted it to be as simple as possible. > this stuff with the code i.e. ModelFoms. I don't know how much of a real > timesaver this really is. Didn't even mention tests.. Which can you auto > generate? Doesn't make so much sense to me to generate test > > Especially if your project's outputed code isn't template driven where each > user of it can change the boiler plate code with minimal or no fuss, so it can > be tailored to each users needs. Like I'm able to add in my most used > imports, which aren't all going to be yours or anyone elses or the same in > each project. I though about it (adding profile, css framework, JS framework...), but doesn't have yet a good idea on how to integrate it right now... and it doesn't feet my first goal which is to quickly create a working project structure that can be easily extended. > > I think the minimal approach of django-admin is a much better way to do these > things. With either way you still have to give the fields, options and the > names, then fill in the functions you need/want. Each function that can be > autogenerated is like two lines of code. > I don't know if your generated code can handle a lot of choices I make, like > making fields not editable in the admin, setting defaults, etc... and even if > you do offer these choices, the time to use your project and code it up > probably isn't much different. > > The big thing for me in terms of tools like this, is the ability to save time, > which I don't think there is a difference between what django-admin has and > your project. Because either way I need to still fill in the fields, the type > of field it is, etc.. and typeing IntegerField() isn't faster than selecting > from a drop down box. I cannot really argue with you on that. If you can please try it first. > > Tho the importing of existing models maybe interesting for refactoring, Tho I > just cp oldversion newversion and make the changes. > > I still have a large feeling I'm going to editting what your project outputs > (blame this on things like UML). When I would just rather write it once and > forget it. > > I also love building development tools, I'm just finishing up something like > this for Go. Where it'll generate a project directory tree of your Go files > (either a package or command, with Makefiels, readme, etc.. all template > driven so you can edit the boilerplate code to be what you want) it'll be > released in a few more days when I finish cleaning up the default templates > and a few optimizations. If you want to see what I mean, I don't mind sending > it today. Just know there are a lot of mistakes in the templates and still > looking at a few tweaks and playing with usability. I would like to see it. that would probably give me some ideas > But in the end, I think the time saved with yours is going to be minimal I > believe unless your can guess what model fields I need and what options they > are it's not saving that much time (maybe 30 minutes?). Otherwise, I think > that altering django-admin startporject to be template driven is a better idea > than this. Sorry for being negative, but without negativity how can we be > challenged and grow? You can be negative but it's hard to take some of your arguments when obviously you didn't try it :p Otherwise I appreciate your argumentation on some points. It's true that optimizing the existent is often the best practice. This more another approach in the spirit of perfectionists with deadlines ( it's far from the perfection ). Thank you Mike > Mike > >> I'll add a copy project so you can extend an existing model if you >> have a big production pace of common project model. >> >> 2010/2/9 Mike Ramirez <gufym...@gmail.com>: >> > On Tuesday 09 February 2010 12:58:32 Brice Leroy wrote: >> >> Hello everybody, >> >> >> >> I've been working for a week on this idea of a generator for django >> >> project. The goal is to provide a simple architecture that can be >> >> after adapted to your project. The system is in alpha stage but allows >> >> you to create project, model (+1 view and 1 form) and application. >> >> Once your project is created you can download a tar.gz with the >> >> complete project already build for you. >> >> >> >> It's supposed to be used with the trunk version of django. >> >> >> >> I plan to add permission, forms and view during the next 15 days. Let >> >> me know if you have any advice or request. >> >> >> >> The project can be found here: http://alpha.djangogenerator.com/ and >> >> is open to registration >> >> The source code is located on github: >> >> http://github.com/debrice/djangogenerator >> >> >> >> All the best, >> >> >> >> -- >> >> Brice Leroy >> >> http://www.debrice.com >> > >> > how is this different than `django-admin startproject` ? >> > >> > Mike >> > -- > -- You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-us...@googlegroups.com. To unsubscribe from this group, send email to django-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/django-users?hl=en.
