[postgis-tickets] [PostGIS] #2260: Benchmarking speed between built-in tiger normalizer and pagc_address_parser

[PostGIS]

#2260: Benchmarking speed between built-in tiger normalizer and
pagc_address_parser
---------------------------------+------------------------------------------
 Reporter:  robe                 |       Owner:  robe         
     Type:  task                 |      Status:  new          
 Priority:  medium               |   Milestone:  PostGIS 2.1.0
Component:  pagc_address_parser  |     Version:  trunk        
 Keywords:                       |  
---------------------------------+------------------------------------------

Comment(by woodbri):

 That is my conclusion also. So I'm working on rewriting this. I have
 figured out how to create the cache. I need to figure out how to hook the
 per query shutdown callback using RegisterExprContextCallback().

 I have decided to change the signature to:

 {{{
 standardize_address(lextab text, gaztab text, rultab text, micro text,
 macro text)
 }}}

 Where *tab are table or view names and internal I'll do a 'select * from
 tab'.

 I don't think I need to cache multiple contexts because these are done on
 a per query contexts and the lex, gaz, and rules will get loaded into the
 standardizer on creation when the first record is processed and will be
 held through out the query then released.

 If there are multiple queries, each query will run in its own memory
 context and have its own standardizer cached in that context and the
 queries will not share contexts.

-- 
Ticket URL: <http://trac.osgeo.org/postgis/ticket/2260#comment:24>
PostGIS <http://trac.osgeo.org/postgis/>
The PostGIS Trac is used for bug, enhancement & task tracking, a user and developer wiki, and a view into the subversion code repository of PostGIS project.