I will assume lazy consensus and commit changes to /trunk soon -- after 3.1 is released, I'll port and deprecate
On Wed, Mar 16, 2011 at 12:06 PM, Ryan McKinley <[email protected]> wrote: > any concerns with this proposal? If not, i would like to commit soon. > > After 3.1 is released, i would merge with 3.x branch and add a deprecation. > > > > On Mon, Mar 14, 2011 at 12:57 PM, Ryan McKinley <[email protected]> wrote: >> the default implementation would just use toString() >> >> For things that could use the type directly (Date/Numbers) they check >> instacneof. >> >> This is actually identical to what currently happens in >> DocumentBuilder, but would happen in the FieldType and would not check >> everything if it is: >> 1. instaceof BinaryField >> 2. instanceof Date >> >> ryan >> >> >> On Mon, Mar 14, 2011 at 12:52 PM, Yonik Seeley >> <[email protected]> wrote: >>> On Mon, Mar 14, 2011 at 12:45 PM, Ryan McKinley <[email protected]> wrote: >>>> Any opinions on this? >>> >>> I've been focused on getting this 3.1 release out (reviewing/fixing >>> docs, packaging, etc). >>> I'm not sure about Object... does that mean most FieldTypes would be >>> doing "instanceof" checks? >>> >>> -Yonik >>> http://lucidimagination.com >>> >>> >>>> thanks >>>> ryan >>>> >>>> >>>> On Sat, Mar 12, 2011 at 2:29 AM, Ryan McKinley <[email protected]> wrote: >>>>> I think FieldType should take an Object input rather then String -- >>>>> this gives FieldTypes the option of using (and reusing) explicit types >>>>> in addition to String. For embedded apps that fill SolrInputDocuments >>>>> with real objects, the fields can use objects directly -- this means >>>>> that Date does not have to get converted to a String and then back to >>>>> a Date. >>>>> >>>>> This is a major API change, but I think the value is worth the trouble. >>>>> >>>>> Thoughts? >>>>> >>>>> ryan >>>>> >>>> >>>> --------------------------------------------------------------------- >>>> To unsubscribe, e-mail: [email protected] >>>> For additional commands, e-mail: [email protected] >>>> >>>> >>> >> > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
