BTW, my other idea about was that we can define a very simple logger
API, while "advanced" loggers would take extra information about the
current operation from the injectable environment objects (or maybe
from thread-bound Transaction).
Andrus
On Aug 2, 2010, at 11:50 AM, Andrus Adamchik wrote:
Hi Evgeny,
On Aug 1, 2010, at 6:13 PM, Evgeny Ryabitskiy (JIRA) wrote:
Evgeny Ryabitskiy commented on CAY-1464:
----------------------------------------
+ 1 for flexibility at this level.
Now we should think about what to path to this events, I mean some
interface for it.
And I am still think that default Cayenne logger can be Improved
without "incur much performance overhead", to provide nice logging
"out of box". That was patch about.
Just took a brief look at the patch... I have some minor notes about
error messages and such, but here are a few things that I have
doubts about:
* Looks like you patch will log all fetched objects? If this is the
case, it is definitely a bad idea. There can be *lots* of them.
* What is the benefit of passing start time instead of time delta in
the logger? IMO the "old" API was cleaner, separating time
measurement from logging.
+ There could be several build-in loggers that could be "turned on"
just by one simple option.
1) Default Logger
2) Formatted logger (move format issues here).
3) Some supper logger that could provide lot's of output.
Yep. DI container is already a reality, so switching to a DI-based
logger should be fairly starightforward (we'll need to figure out
injection into lower level objects underneath DataDomain). This is
on my TODO list. If somebody decides to do it before I get to it,
please open a Jira and let me know.
Andrus
