I agree that it would be great to have a nice way to handle these
kinds of queries. I'm struggling a bit with the particular syntax
you've proposed, but maybe I just need to stare at it a bit longer.
Mark.
Ted Han wrote:
> Well... actually the important part is that it's block based querying. The
> semicolons are just logical line breaks.
> Human.all{ sex == "M"; any{ last_name == "Obama"; last_name == "Clinton" } }
>
> could just as well be
>
> Human.all do
> sex == "M"
> any do
> last_name == "Obama"
> last_name == "Clinton"
> end
> end
>
> One nice additional benefit there, being the fact that you can execute
> whatever you want within the block (warning there though, the scope for the
> block internals isn't the same as the Model's scope).
>
> Also, one effect that i'm not sure i'm entirely happy with, is the fact that
> you can actually chain the statements together... Since the query is built
> via side-effect, rather than return value (i know, my LISP prof is weeping
> inside right now, though he knows not why). Example:
>
> Human.all{ sex == "M" and any{ last_name == "Obama" or last_name ==
> "Clinton" } }
>
> So if you don't like semicolons, you could make it all one logical
> statement.
>
> Lastly, i'm not actually overriding Enumerable's #any? method, cause it's
> #any? not #any. I'm concerned that will cause confusion, as we have just
> encountered. If someone can propose a better name, i'm more than willing to
> use it. However, all vs any is sort of the sensible pair.
>
> The goal here really is to make querying DM more like using normal ruby.
> Rather than throwing around these hashes with all these meaningless =>
> arrows all over the place, you can use the actual operators you're used to.
>
> On Fri, Oct 2, 2009 at 3:10 PM, Brian Smith <[email protected]> wrote:
>
> > I really like the idea of overloading the any method. Not sure if using the
> > semicolon in the middle of it is the best way to go about it. It seems odd
> > to me. Might I suggest just passing n number of hashes to where each hash is
> > a clause of the or query. Some of the syntax is a bit confusing in the
> > current implementation.
> >
> >
> > On Fri, Oct 2, 2009 at 7:34 AM, Ted Han <[email protected]> wrote:
> >
> >> I've got one half completed. It's a library called dm-sugar-glider.
> >> Right now it allows you to do block based queries, and nested conditions,
> >> using ANDs or ORs.
> >>
> >> I'll post to the list when it's done. (I've got the in model property
> >> matching down, i'm wrestling with some internals in core to sort out the
> >> matching w/ relationships) It will require people to bump up to edge
> >> though.
> >>
> >> Examples:
> >>
> >> original = Human.all(:last_name => "Obama", :first_name => /a/)
> >> mixed = Human.all(:last_name => "Obama"){ first_name =~ /a/}
> >> block = Human.all{ last_name == "Obama"; first_name =~ /a/}
> >> mixed.query.should == original.query
> >> block.query.should == original.query
> >>
> >> Human.all{ first_name =~ /ar/; sex == "F" } # => Hillary, Barbra,
> >> Barb
> >> Human.any{ first_name =~ /ar/; sex == "F" } # => Barack + ALL women
> >> Human.any{ sex == "M"; all{ first_name == "Barbra"; dob == "June 8,
> >> 1925" }} # All men + the elder Barbra Bush
> >> Human.all{ sex == "M"; any{ last_name == "Obama"; last_name ==
> >> "Clinton" } } # Barack Obama + Bill Clinton
> >>
> >> On Fri, Oct 2, 2009 at 8:09 AM, Jacques Crocker <[email protected]>wrote:
> >>
> >>>
> >>> Ok, so the error that is being thrown is a result of a different issue
> >>> (isolated and discussed in another thread).
> >>>
> >>> Question is still open though, is there a better way to write OR
> >>> queries in datamapper 0.10?
> >>>
> >>> Thanks!
> >>>
> >>>
> >>> On Oct 2, 4:34 am, Jacques Crocker <[email protected]> wrote:
> >>> > Via Datamapper 0.10, is there a mechanism to better handle
> >>> > conditionals with OR (instead of AND)?
> >>> >
> >>> > I had my ghetto search query break from 0.9 to 0.10:
> >>> > all(["(title like ? OR description_source like ? OR company_name
> >>> > like ? OR outside_location like ?)", search, search, search, search])
> >>> >
> >>> > I know there's the much fancier :name.like => "" , however I
> >>> > couldn't figure out how to make it work with an OR query. Can anyone
> >>> > help?
> >>> >
> >>> > Here's the stack trace I get on the above query. the search variable
> >>> > is just a simple string
> >>> > ~ Binding mismatch: 2 for 1
> >>> > INTERNAL ERROR!!! Binding mismatch: 2 for 1
> >>> >
> >>> /Users/jc/Merb/testapp/gems/gems/data_objects-0.10.0/lib/data_objects/
> >>> > command.rb:73:in `escape_sql'
> >>> >
> >>> /Users/jc/Merb/testapp/gems/gems/dm-core-0.10.0/lib/dm-core/adapters/
> >>> > data_objects_adapter.rb:88:in `execute_reader'
> >>> >
> >>> /Users/jc/Merb/testapp/gems/gems/dm-core-0.10.0/lib/dm-core/adapters/
> >>> > data_objects_adapter.rb:88:in `read'
> >>> >
> >>> /Users/jc/Merb/testapp/gems/gems/dm-core-0.10.0/lib/dm-core/adapters/
> >>> > data_objects_adapter.rb:266:in `with_connection'
> >>> >
> >>> /Users/jc/Merb/testapp/gems/gems/dm-core-0.10.0/lib/dm-core/adapters/
> >>> > data_objects_adapter.rb:84:in `read'
> >>>
> >>>
> >>
> >>
> >>
> >
> > >
> >
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"DataMapper" 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/datamapper?hl=en
-~----------~----~----~----~------~----~------~--~---
