Re: How to use full-text search URL parser to filter query results by domain name?

2019-04-10 Thread Jess Wren
On 4/8/19 4:50 AM, Arthur Zakirov wrote: > I think it is normal to use ts_parse(). And I suppose you might use > windows functions. > > For example, you have table links: > > =# create table links (score int, link text); > =# insert into links values >   (1, 'http://www.foo.com/bar'), >   (2, 'http

Re: How to use full-text search URL parser to filter query results by domain name?

2019-04-07 Thread Jess Wren
On 4/6/19 11:42 PM, haman...@t-online.de wrote: > Hi, > > I have no real idea about solving the complete problem, and would probably try > something with a temp table first. > For extracting the hostname from a url you could use > > select regex_replace('https?://(.*=)/.*', '\\1', url) > > instead

How to use full-text search URL parser to filter query results by domain name?

2019-04-06 Thread Jess Wren
I am trying to understand how to use the full-text search parser for URLS and hostnames to filter results from a text field containing URLS based on domain, and also how to index text columns for fast lookup/matching based on domain. I have a PostgreSQL database containing documents and links down