On 10/10/07, Ben Sizer <[EMAIL PROTECTED]> wrote: [...]
> But if I start with 1.0 today, as I intend to, what are my SQLAlchemy > options like? I'm aware that it's possible to use it, and has been > possible for a long time, but is the way in which it's used going to > change significantly for 1.0.4beta1? Your SQLAlchemy options are simple: if you use 1.0.4beta now, you'll have the latest and greatest nearly ready for 1.1. I mean nearly because you will already use the new SA 0.4 ready API which will be the one used in 1.1. (This is if you use a project quickstarted by TG 1.0.4beta1) The only 'breaking' thing when moving to 1.1 will be the transaction management (we'll speak of this a few lines down in this mail) > Is it simple to start working with explicit transactions already? If you start now without any change a quickstarted project you use transactions without knowing it. You can modify your dburi to include the "notrans" option in order to disable the transactions in TG. Since we will certainly remove implicit transactions from TG this means you should think about your data integrity policy and begin/commit/rollback yourself in the controller. All of this is doable today in 1.0. If you control your transactions yourself today [1] you'll be safe tomorrow. If you rely on the implicit transactions of TG to rollback your inserted data for you in case of error in the controller then you'll need to review your code when moving to 1.1 to manage your data integrity yourself in your controller code. My advice would by to manage your transactions by yourself even today, even in 1.0.1... > What's the situation with using Genshi in 1.0? Again, I'm aware it's > possible, but how practical is it, and how compatible will it be with > the way 1.1 does it? If you want to use Genshi today, just go with it right now, it works well. If you prefer to stick with Kid for the moment, no problem, converting kid templates to Genshi templates is quite easy. Using Genshi today will be compatible with tomorrow, but using kid is also compatible. It is more a question of taste on your side, I don't perceive too much trouble doing so. > I've not taken a detailed look at it, but at first glance I think the > organisation is still quite poor, and reference materials are still > treated as secondary to tutorial materials. [...] > Not to put down the hard work of those who've contributed, just to > note that there's still some way to go to be first class. You are right to say that there is room for improvement. We would be interested to hear about a more detailed version of the comment you made here... we always welcome people like you with interesting suggestions, please go on. I suggest you subscribe to the docs mailing list [2] and start a new thread about your impressions and what is missing. Chris Arndt who is our documentation über master (with also being our sysadmin) is a really helpful person who will be glad to see people actually care about the docs :) Cheers, Florent. [1] http://www.sqlalchemy.org/docs/04/session.html#unitofwork_managing [2] http://groups.google.com/group/turbogears-docs --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears?hl=en -~----------~----~----~----~------~----~------~--~---
