Open cderenburger opened 9 years ago
Build out a search vector for each service, remove search vector from locations. See if that corrects the issue.
Vector to testing stage, but root partition on drive is full. Must complete computer maintenance and try importing again.
Service search vector would require service endpoint. Performance is slower and DB size is quite bloated.
Changed Excel sheets to create a location for each service record. Works, seems as responsive as previous configuration. Results are correct for Keyword and service areas.
Having a location for every service means the locations search returns duplicate matches for an address, with no way to differentiate between them. Need to add service name and short description to make this useful.
cc @monfresh
Field "short-desc" exists, just need to populate it with info from existing service short description. May look at displaying service name on results page.
Issue with this method is searching "near me" or by organization, with many duplicated addresses at one location. Undecided as to what. if anything, to do about that with the map.
For each service with location, prepend service name to location name on locations.csv. 1:1 location_id:service_id and 1:many organization_id location_id
"Household Goods and Furniture - Abraham's House" rather than "Abraham's House"
See issue and discussion https://github.com/codeforamerica/ohana-api/issues/365