pg.ddx.io  pgsql-chat@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Andrew Dunstan <andrew@dunslane.net>
To: PostgreSQL-development <pgsql-hackers@postgresql.org>
Subject: Re: is GiST still alive?
Date: Thu, 23 Oct 2003 07:40:33 -0400
Message-ID: <3F97BE31.1030605@dunslane.net> (raw)
In-Reply-To: <3F97349A.5040300@familyhealth.com.au>
References: <200310220438.h9M4cY232554@hosting.commandprompt.com>
	<200310221219.43179.josh@agliodbs.com>
	<60ptgpf175.fsf@dev6.int.libertyrms.info>
	<3F96FA03.6040009@dunslane.net>
	<3F97349A.5040300@familyhealth.com.au>


Storing the XML text has problems - you have to parse it every time you 
want something - that has to cause a huge performance hit.

I use XML a lot for all sorts of purposes, but it is appropriate for 
data transfer rather than data storage, IMNSHO.

cheers

andrew

Christopher Kings-Lynne wrote:

> Do this:
>
> 1. Create a new type called 'xml', based on text.
>
> 2. The xmlin function for that type will validate what you are 
> enterering is XML
>
> 3. Create new functions to implement XPath, SAX, etc. on the xml type.
>
> 4. Extend the contrib/ltree gist-based tree indexing scheme to work on 
> xml and hence the operations in no.3 above are really fast...
>
> Chris
>
>
> Andrew Dunstan wrote:
>
>> Christopher Browne wrote:
>>
>>> But I think back to the XML generator I wrote for GnuCash; it has the
>>> notion of building up a hierarchy of entities and attributes, each of
>>> which is visible as an identifyable object of some sort.  Mapping that
>>> onto a set of PostgreSQL relations wouldn't work terribly well.
>>>  
>>>
>>
>> *nod* I have tried this several times - it just doesn't work well, 
>> because the maps are too different.
>>
>> You could do something like this:
>>   . a table for each element type, fields being the attributes, plus 
>> the node id.
>>   . a table to tie everything together (parent_id, child_id, 
>> child_order, child_type).
>> In theory you could even generate the DB schema from an XML schema 
>> and evaluate it with XPath-like expressions.
>>
>> But why put yourself to such bother? I have never found a good reason 
>> to do this sort of thing.
>>
>> cheers
>>
>> andrew
>>
>>
>> ---------------------------(end of broadcast)---------------------------
>> TIP 7: don't forget to increase your free space map settings
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 7: don't forget to increase your free space map settings
>




view thread (16+ messages)  latest in thread

Message-ID: <3F97BE31.1030605@dunslane.net>
Permalink:  ../3F97BE31.1030605@dunslane.net/
Also on:    postgresql.org/message-id/3F97BE31.1030605@dunslane.net

 · 

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients from your mail client; the link fills in
  the headers below: reply via email
  Some clients, webmail especially, drop In-Reply-To and References from
  a link, and the reply then starts a new thread: use the mbox below.

  To: pgsql-chat@postgresql.org
  Cc: andrew@dunslane.net, pgsql-hackers@postgresql.org
  Subject: Re: is GiST still alive?
  In-Reply-To: <3F97BE31.1030605@dunslane.net>
  References: <200310220438.h9M4cY232554@hosting.commandprompt.com> <200310221219.43179.josh@agliodbs.com> <60ptgpf175.fsf@dev6.int.libertyrms.info> <3F96FA03.6040009@dunslane.net> <3F97349A.5040300@familyhealth.com.au> <3F97BE31.1030605@dunslane.net>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox