On 7/20/07, Thomas Mueller <[EMAIL PROTECTED]> wrote:
Hi,

> the scenario posted here: searching for a John and getting a
> Joe instead

Ok, so the use case is:

1) Get session and load a contact with name 'Joe'
2) Update the name property from 'Joe' to 'John'
3) Now, query the repo for the contact using the name 'Joe'

I would do this:

1) Node contact = ...getNode(...) // we have the node
2) contact.setProperty("name", "John")
    contact.save()
3) ... query ...

Then everything works fine. No problem.
You _only_ have a problem if if you don't call 'save()'.
So why exactly don't you want to call 'save' first?


Thomas, I don't think I have put the problem this way (I want or I
don't want to).
There are two sides of the story:
- the above workflow must be more clear in the spec
- people may be asking why they need to use a specific workflow, when
other persistence solutions are not requiring it.

We can definitely fix the first, but we must be prepared to have a
good answer for the 2nd too. And for the moment, frankly speaking, I
don't seem to have any other explanation than "that's what the spec
says" (or at most Jukka's explanation in this thread -- which while
valid is still complex for a RDBMS guy).

bests,
./alex
--
.w( the_mindstorm )p.

Thomas

Reply via email to