On 31/03/2011 16:09, Chris Withers wrote:
What's the preferred way of working with transactions now, be they just
SQLAlchemy transactions or, more likely, a bunch of things tied together
by a transaction from the 'transaction' package?
repoze.tm2 seems to be the "old skool" BFG way of doing things, but is
this still the "right" way in Pyramid?
The templates/scaffolds certainly use repoze.tm2, but the book uses the
transaction package as an example for "Using Finished Callbacks". What
are the differences between these two approaches?
Trying to answer this myself:
So, repoze.tm2 has the following features missing from the simple
example in "Using Finished Callbacks":
- commit veto: what situations would request.exception be None but the
response status start with '4' or '5'?
- after_end cleanups: but, I guess these could just be passed to
add_finished_callback, right? FWIW, doesn anyone have any real-world
examples where they've needed to use after_end cleanups?
- isDoomed support: I guess this could be added to the example; can
someone explain when isDoomed is True and what happens if that
transaction is committed instead of being manually aborted?
If I understand the above correctly (hopefully someone will point out if
I missed anything!) then a simple Finished Callback should suffice, right?
Related: Any chance of a RequestFinished event to subscribe to rather
than having to (rather convolutedly) subscribe to NewRequest just so
your handler can call add_finished_callback?
...in which case, I'd dearly love that RequestFinished event!
cheers,
Chris
--
Simplistix - Content Management, Batch Processing & Python Consulting
- http://www.simplistix.co.uk
--
You received this message because you are subscribed to the Google Groups
"pylons-devel" group.
To post to this group, send email to pylons-devel@googlegroups.com.
To unsubscribe from this group, send email to
pylons-devel+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/pylons-devel?hl=en.
