Excellent, it is working for the simpler case, but for oracle 8 (who
isn't as smart when indexing) I also need it to work for
subqueryload().
So the problem is that my FOR UPDATE OF is also being added for
subqueryload selects.
* Can I tell within the compiles context if this is for subqueryload?
(Or can the Query tell?)
* Are there other cases where the query is "reused" that I need to be
careful of?
I restructured this way (as you're original suggestion to fix another
issue):
@compiles(Select)
def compile_forupdateof(select, compiler, **kw):
rendered = compiler.visit_select(select, **kw)
if hasattr(select, '_for_update_of'):
mapper = class_mapper(select._for_update_of)
name = mapper.mapped_table.name
if compiler.dialect.name == 'oracle':
# Oracle makes us specify the column name (for views, I
guess, since it locks entire row)
name += '.' + mapper.primary_key[0].name
rendered = "%s FOR UPDATE OF %s" % (rendered, name)
return rendered
On Feb 3, 9:51 pm, Michael Bayer <mike...@zzzcomputing.com> wrote:
> On Feb 3, 2011, at 9:29 PM, Kent wrote:
>
>
>
> > Yeah, I wanted to apologize because my heart wants to contribute to
> > the project (really), but I'm working overtime like mad swamped
> > because our product is live in use now and I've got a backload of
> > tickets to solve! I also feel my level of understanding currently is
> > more hacking than contributing. I hope to be of more help to the
> > project in the future.
>
> > Is "simpler than you had in mind" a good thing or am I over
> > simplifying and it won't work for bunch of cases?
>
> > I note that the simple case is working, but something like this fails:
>
> > DBSession.query(Order).for_update_of(Order).limit(10).all()
>
> > since I really need to have the for update inside in this case... any
> > advise or is this what you meant by "There's not a great way to
> > intercept the middle of the SELECT compilation with a new kind of
> > clause in this case."?
>
> i think if it works for what you need right now, then its great. @compiles
> is meant to give you what you need to get out of a jam.
>
>
>
> > On Feb 3, 9:07 pm, Michael Bayer <mike...@zzzcomputing.com> wrote:
> >> oh OK this is a little simpler than what I had in mind, you just have to
> >> add the mixin expression.Executable to your ForUpdateOf class.
>
> >> On Feb 3, 2011, at 9:05 PM, Kent wrote:
>
> >>> Here is a crude outline (need to properly escape table name, etc.), of
> >>> what I think might work, and it seems to render properly, but crashes
> >>> with:
>
> >>> File "/home/rarch/tg2env/lib/python2.6/site-packages/
> >>> SQLAlchemy-0.6.4.2kbdev-py2.6-linux-x86_64.egg/sqlalchemy/engine/
> >>> default.py", line 353, in __init__
> >>> raise exc.ArgumentError("Not an executable clause: %s" % compiled)
> >>> ArgumentError: Not an executable clause:
> >>> ...
>
> >>> class MyQuery(Query):
> >>> _for_update_of = None
>
> >>> @_generative()
> >>> def for_update_of(self, arg):
> >>> """Keep track that we want to for update of this"""
> >>> self._for_update_of = class_mapper(arg).mapped_table.name
>
> >>> def _compile_context(self, labels=True):
> >>> context = super(MyQuery, self)._compile_context(labels)
> >>> if self._for_update_of:
> >>> context.statement = ForUpdateOf(context.statement,
> >>> self._for_update_of)
> >>> return context
>
> >>> class ForUpdateOf(ClauseElement):
> >>> def __init__(self, statement, for_update_of):
> >>> self.statement = statement
> >>> self.for_update_of = for_update_of
>
> >>> @compiles(ForUpdateOf)
> >>> def compile_forupdateof(element, compiler, **kw):
> >>> return "%s FOR UPDATE OF %s" %
> >>> (compiler.process(element.statement), element.for_update_of)
>
> >>> --
> >>> You received this message because you are subscribed to the Google Groups
> >>> "sqlalchemy" group.
> >>> To post to this group, send email to sqlalchemy@googlegroups.com.
> >>> To unsubscribe from this group, send email to
> >>> sqlalchemy+unsubscr...@googlegroups.com.
> >>> For more options, visit this group
> >>> athttp://groups.google.com/group/sqlalchemy?hl=en.
>
> > --
> > You received this message because you are subscribed to the Google Groups
> > "sqlalchemy" group.
> > To post to this group, send email to sqlalchemy@googlegroups.com.
> > To unsubscribe from this group, send email to
> > sqlalchemy+unsubscr...@googlegroups.com.
> > For more options, visit this group
> > athttp://groups.google.com/group/sqlalchemy?hl=en.
>
>
--
You received this message because you are subscribed to the Google Groups
"sqlalchemy" group.
To post to this group, send email to sqlalchemy@googlegroups.com.
To unsubscribe from this group, send email to
sqlalchemy+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/sqlalchemy?hl=en.
