I don't know how postgres server would optimize the query, but in teory
there should be only a slight overhead as the (I)LIKE results should be
cached. How much takes to order by depends on the number of rows returned.
Dave [Hawk-Systems] wrote:
Appreciate the responses...
We are talking a db o
I use mysql, so this was for mysql - LIKE is case insesitive and returns
0 for false and 1 for true.
Dean E. Weimer wrote:
Try
SELECT , (((keywords LIKE '%$search%') * 5) + ((title LIKE
'%$search%') * 3) + (description LIKE '%$search%')) score FROM .
ORDER BY score DESC
PostgreSQL canno
Appreciate the responses...
>> Try
>> SELECT , (((keywords LIKE '%$search%') * 5) + ((title LIKE
>> '%$search%') * 3) + (description LIKE '%$search%')) score FROM .
>> ORDER BY score DESC
>
>PostgreSQL cannot type cast the Boolean type so you have to use a case
>statement, also changing li
> Try
> SELECT , (((keywords LIKE '%$search%') * 5) + ((title LIKE
> '%$search%') * 3) + (description LIKE '%$search%')) score FROM .
> ORDER BY score DESC
PostgreSQL cannot type cast the Boolean type so you have to use a case
statement, also changing like to ilike will get results regard
Try
SELECT , (((keywords LIKE '%$search%') * 5) + ((title LIKE
'%$search%') * 3) + (description LIKE '%$search%')) score FROM .
ORDER BY score DESC
Dave [Hawk-Systems] wrote:
looking for code snippets or links to examples of the following;
- Have a database with multiple fields that wil
looking for code snippets or links to examples of the following;
- Have a database with multiple fields that will be searched against (happens to
be PostgreSQL in this instance, but we can migrate any MySQL based
examples/code)
- We wish to score search results - ie: a match in "keywords" is worth
6 matches
Mail list logo
