Re: [Pytables-users] problem with "where" clause

[David L]

The formatting was lost in the original email. Hopefully the attachment will work better.

From: "David L" <[EMAIL PROTECTED]>
To: pytables-users@lists.sourceforge.net
Subject: [Pytables-users] problem with "where" clause
Date: Tue, 15 May 2007 17:15:03 -0400

Hi,











Thanks for any help you can provide.

David


#!/usr/bin/python

from tables import *
import sys

# Define a user record to characterize some kind of particles
class Particle(IsDescription):
   name      = StringCol(16, indexed = True)   # 16-character String

def write (start, stop):
        print "start, stop = %d, %d" % (start, stop)
        if start == 0:
                open_mode = "w"
        else:
                open_mode = "a"
        # Open a file in "w"rite mode
        h5file = openFile("test.h5", mode = open_mode, title = "Test file")

        if start == 0:
                # Create a new group under "/" (root)
                group = h5file.createGroup("/", 'detector', 'Detector 
information')
                # Create one table on it
                table = h5file.createTable(group, 'readout', Particle, "Readout 
example")
        else:
                table = h5file.root.detector.readout
        # Fill the table with 10 particles
        particle = table.row
        for i in xrange(start, stop):
            particle['name']  = 'Particle: %6d' % (i)
            # Insert a new particle record
            particle.append()
        # Close (and flush) the file
        if start > 0:
                table.reIndex()
        h5file.close()

def where (particle_name):
        h5file = openFile("test.h5", mode = "r", title = "Test file")
        index = h5file.root.detector.readout
        count = 0
        for row in index.where (index.cols.name == particle_name):
                count += 1
                print row ["name"]

        print "indexed search found %d matches" % count

def seq (particle_name):
        h5file = openFile("test.h5", mode = "r", title = "Test file")
        index = h5file.root.detector.readout
        count = 0
        for row in index:#.where (index.cols.name == particle_name):
                if row ['name'] == particle_name:
                        count += 1
                        print row ["name"]

        print "sequential search found %d matches" % count

if __name__ == "__main__":
        if sys.argv [1] == "good" or sys.argv [1] == "bad":
                if sys.argv [1] == "good":
                        start = 0
                        stop = 40000
                        end = 1
                else:
                        if sys.argv [1] == "bad":
                                start = 0
                                stop = 1000
                                end = 40

                for x in range (0, end):
                        write (start, stop)
                        start = stop
                        stop += 1000
        else:
                particle_name = "Particle:  36001" # works only with "good" 
write
        #       particle_name = "Particle:  25001" # always works

                where (particle_name)
                seq (particle_name)

_________________________________________________________________

I've been having problems using the "where" clause and I would appreciate any suggestions. The problem occurs when I try to use a "where" clause to access records that are near the end of the file and when the file has been built from multiple runs. The code below demonstrates the problem. If the script below is started with the command line parameter of "good", it generates a table of 40,000 records. Then if the script is run with the command line parameter of "read", the record is found with both the sequential and indexed searches. If the script is started with the command line parameter of "bad", the table is generated in separate passes. Each pass writes opens the file, writes 1000 records and closes the file. Then if the script is run with the command line parameter of "read", the record is only found in the sequential search. A possibly related problem is that when the "bad" version of the file is generated, ptdump crashes. Make every IM count. Download Messenger and join the i’m Initiative now. It’s free. http://im.live.com/messenger/im/home/?source=TAGHM_MAY07

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

_______________________________________________
Pytables-users mailing list
Pytables-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/pytables-users

_________________________________________________________________

Catch suspicious messages before you open them—with Windows Live Hotmail. http://imagine-windowslive.com/hotmail/?locale=en-us&ocid=TXT_TAGHM_migration_HM_mini_protection_0507

h5_test.py
Description: Binary data

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Pytables-users mailing list
Pytables-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/pytables-users