[phpXML] Re: [phpXML] Re: [phpXML] Large amounts of data

[<spr...@ho...>]

Speed isn't much of a concern for me here, but memory usage is. The XML is 
only being used for exporting, importing and copying/pasting objects around 
in the application server's management interface. Content, presentation and 
logic are all stored in a MySQL database. The XML can get large because 
serializing a folder object serializes that object and any objects it 
contains. Again, the XML is being used to move objects around and to export 
objects and object types for backup or for import into another Phratry 
application server.

The project is at phratry.sourceforge.net. It should have an alpha release 
within a couple weeks. It is similar to Zope but, in my biased opinion, much 
easier to install, use and develop in.

----Original Message Follows----
From: ph...@pe...
Reply-To: us...@li...
To: us...@li...
Subject: [phpXML] Re: [phpXML] Large amounts of data
Date: 8 Jul 2001 12:12:47 -0000

On Friday 06 July 2001 17:30, you wrote:
 > I'm using phpXML to export/import object heirarchies in an application
 > server. I haven't run into the memory limit yet, but when the application
 > server starts getting some real use people may want to back up an entire
 > site.
 >
 > I'm thinking of creating an extension of phpXML that stores node content 
in
 > a temporary file instead of keeping it in memory. The method
 > set_external_content() will write to a file pointer and then set some 
node
 > attributes like ext_ctnt_position and ext_ctnt_length. The method
 > get_external_content() will use those attributes to seek to that position
 > in the file and grab the specified content.
 >
 > Any thoughts on this? Pitfalls? Interest?

obvious disadvantage for one-off use is that it would be much slower. If 
this
were reasonably static data, you could store the result in a permanent file
as a sort of Xpath index, and then use that in your scripts - I quite like
that idea, especially as other users could then use the index instead of
rereading the file every time.

xml files are no different from any other flatfile 'database': reading an
entire large file to extract a small amount of data is not a very efficient
thing to do. Without knowing your application, I would say as a rule of
thumb, if xml files start getting large, either move to dbms with proper
indexing or find some way to split into smaller files.

--
This message has been sent through the <phpXML/> user discussion list. To 
unsubscribe, please visit 
https://sslsites.de/mailinglisten/user/us...@li.../

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com


--
This message has been sent through the <phpXML/> user discussion list. To unsubscribe, please visit https://sslsites.de/mailinglisten/user/us...@li.../