Re: Deployment Tool Recommendation

2009-07-15 Thread martin f krafft
also sprach Shawn Milochik [2009.07.15.1722 +0200]: > > Any reason why you can't do 'sudo -u otheruser bash -l' or even > > 'sudo su > > - otheruser'? Seems strange to be able to sudo to root, but unable to > > sudo to a role account. > > > No, I don't see a reason not to do it that way. I jus

caching generated content from the filesystem with Django

2009-05-22 Thread martin f krafft
Hi folks, I am new to Django, and trying to figure out if Django could be used for a new project which is similar to a news site (stories & syndication feeds), as well as more permanent content. The twist is that we already have thousands of such stories stored as Markdown or RST files on the fil

When do snippets get into Django?

2009-05-22 Thread martin f krafft
Hi folks, I am new to Django and could not find the answer to a very important consideration I need to make before diving into Django more. I used to work with Zope and Plone, and one of the major problems were upgrades, not of Zope or Plone, but of all the products that were needed for a particul

caching generated content from the filesystem with Django

2009-05-22 Thread martin f krafft
Hi folks, I am new to Django, and trying to figure out if Django could be used for a new project which is similar to a news site (stories & syndication feeds), as well as more permanent content. The twist is that we already have thousands of such stories stored as Markdown or RST files on the fil

de-facto tagging (was: When do snippets get into Django?)

2009-05-22 Thread martin f krafft
also sprach Alex Gaynor [2009.05.22.1545 +0200]: > http://jacobian.org/writing/what-is-django-contrib/ here is > a description of what one of the core developers see as the role > of django contrib. It should be noted that 2 of the core devs > said at euro django con that that they prefer the co

Re: de-facto tagging (was: When do snippets get into Django?)

2009-05-22 Thread martin f krafft
also sprach Dougal Matthews [2009.05.22.2352 +0200]: > Pretty must everybody does seem to use django-tagging unless they have some > special weird use-case. > > I've used it every time I've needed tagging and don't have a gripe with it > yet. So how would it find its way into contrib? -- marti

Re: de-facto tagging (was: When do snippets get into Django?)

2009-05-23 Thread martin f krafft
also sprach Dougal Matthews [2009.05.23.0232 +0200]: > It doesn't. Why should it? Well, because django-tagging is, quote Jacob: - optional - de-facto standard - the implementation of a common pattern Tagging is a very common feature nowadays, isn't it? Having tag support integrated with D

Re: de-facto tagging (was: When do snippets get into Django?)

2009-05-23 Thread martin f krafft
also sprach Kai Kuehne [2009.05.23.1048 +0200]: > > Pretty must everybody does seem to use django-tagging unless > > they have some special weird use-case. > > Huh? Can you remove that "everybody", please? I've never used > tagging and I'm confident - many others too. I think he implicitly meant

Re: de-facto tagging (was: When do snippets get into Django?)

2009-05-23 Thread martin f krafft
also sprach Dougal Matthews [2009.05.23.1103 +0200]: > I don't think tagging is a particularly hard or non-trivial. No. All the more reason to push for a standard and prevent dozens of different implementations, no? > Anyway, this conversation is a bit irrelevant because; Core > developers want

validation and the ORM layer

2009-05-25 Thread martin f krafft
Hey folks, while playing around with the DB API last night, we discovered that the ORM layer does not enforce content validation. Since the blank field option defaults to False, we expected to be unable to save ORM objects with empty fields, but the DB API will let us do just that: class Beers

Re: de-facto tagging (was: When do snippets get into Django?)

2009-05-25 Thread martin f krafft
also sprach Dougal Matthews [2009.05.23.1823 +0200]: > Fair enough. I don't really have an issue with the idea, could be > quite handy. > > I can just imagine them saying no :) so thought I would question > it as a move/idea. Look what I found! Check out the fifth point, labeled #1, top-most ann

Re: validation and the ORM layer

2009-05-25 Thread martin f krafft
also sprach Alex Gaynor [2009.05.25.1945 +0200]: > Django does not currently support model-validation, there is > a google summer of code project being done by Honza Kral to add > support for this. Good to know, thanks. This means that the feature won't be in 1.1 though, right? -- martin | htt

Re: de-facto tagging (was: When do snippets get into Django?)

2009-05-25 Thread martin f krafft
also sprach James Bennett [2009.05.25.2047 +0200]: > And, really, this is the sort of thing that leads me to say it > shouldn't be in contrib right now. We do need to have some sort of > document outlining criteria, but I'd like to think that commitment to > a stable API (e.g., the author is willi

Obtaining content from Git

2014-11-21 Thread martin f krafft
Hello, we have a Django project with a few pages that come from Git. Currently, Apache rewrite rules serve those files statically (and they make use of the same template/CSS as Django does, so the user does not actually notice), and it's a massive hack and pain to keep up-to-date. Hence I was thi

Obtaining content from Git

2014-11-22 Thread martin f krafft
Hello, we have a Django project with a few pages that come from Git. Currently, Apache rewrite rules serve those files statically (and they make use of the same template/CSS as Django does, so the user does not actually notice), and it's a massive hack and pain to keep up-to-date. Hence I was thi

Re: Obtaining content from Git

2014-11-23 Thread martin f krafft
also sprach Russell Keith-Magee [2014-11-24 00:42 +0100]: > The problem is that there isn't one. There's several :-) > > I'm aware of at least 3: > > https://pypi.python.org/pypi/GitPython/0.3.2.1 > https://pypi.python.org/pypi/pygit2 > https://pypi.python.org/pypi/dulwich I am aware of those

Re: Obtaining content from Git

2014-11-24 Thread martin f krafft
also sprach Russell Keith-Magee [2014-11-24 07:16 +0100]: > The right place is in the view. […] > If you want to do "interesting" things like retrieve a specific > file version, you're not going to be able to use the storage API, > because most raw Interesting perspective you have there, which I

Re: Obtaining content from Git

2014-11-24 Thread martin f krafft
also sprach Tim Chase [2014-11-24 11:19 +0100]: > urlpatterns = patterns('', > url(r'^blob/([0-9a-f]{40})/$', 'myapp.views.render_blob', ...), > ) … this is straight-forward, thanks. I'd probably try not to expose the SHAs to the user but request from the Git library a blob identified b

Re: Obtaining content from Git

2014-11-24 Thread martin f krafft
also sprach Russell Keith-Magee [2014-11-24 11:38 +0100]: > Perhaps I wasn't clear. No, you were. I am just pushing back a little bit because I come from another angle and I greatly appreciate your explanations in helping me understand the Django way of thinking. So in the hope that you now don'

Re: Obtaining content from Git

2014-11-24 Thread martin f krafft
also sprach Russell Keith-Magee [2014-11-25 01:16 +0100]: > Sure - that could work too. Although I'd question why you'd do > this, rather than just checking out the relevant hash onto the > filesystem, and doing a direct filesystem read. Yeah, and I think you have me 99.5% convinced that I want

Custom field reference flowchart

2016-01-12 Thread martin f krafft
Dear Django people, I'm working on something I called "IndirectGenericForeignKey",¹ and following suggestionsi I received on django-developers, I've made a bit of progress on my implementation, using a custom (pseudo-)field.² My code works, but I can't exactly say I understand all of it, which is