Sponsored Content
Full Discussion: which is better?
Top Forums UNIX for Advanced & Expert Users which is better? Post 302240610 by risk_sly on Friday 26th of September 2008 06:22:26 AM
Old 09-26-2008
which is better?

Hi,

I would like to know which works better in this scenario:

I have a file which contains about 300K to 400K records. This file will be loaded into an oracle table but will be filtered first. I have two ways of doing this.

1st, I filtered the original file and created a new file that contains only the selected records (using awk)

2nd, I included a WHERE clause in the sqlloader control file to filter the records

with these two options, which do you think is more efficient. (some explanations will be much appreciated) thank you.
 
PX_GET_RECORD2(3)					     Library Functions Manual						 PX_GET_RECORD2(3)

NAME
PX_get_record2 -- Returns record in Paradox file SYNOPSIS
#include <paradox.h> int PX_get_record2(pxdoc_t *pxdoc, int recno, char *data, int *deleted, pxdatablockinfo_t *pxdbinfo) DESCRIPTION
This function is similar to PX_get_record(3) but takes two extra parameters. If *deleted is set to 1 the function will consider any record in the database, even those which are deleted. If *pxdbinfo is not NULL, the function will return some information about the data block where the record has been read from. You will have to allocate memory for pxdbinfo before calling PX_get_record2. On return *deleted will be set to 1 if the requested record is deleted or 0 if it is not deleted. The struct pxdatablockinfo_t has the fol- lowing fields: blockpos (long) File positon where the block starts. The first six bytes of the block contain the header, followed by the record data. recordpos (long) File position where the requested record starts. size (int) Size of the data block without the six bytes for the header. recno (int) Record number within the data block. The first record in the block has number 0. numrecords (int) The number of records in this block. number (int) The number of the data block. This function may return records with invalid data, because records are not explizitly marked as deleted, but rather the size of a valid data block is modified. A data block is a fixed size area in the file which holds a certain number of records. If for some reason a data block has newer been completely filled with records, the algorithmn anticipates deleted records in this data block, which are not there. This often happens with the last data block in a file, which is likely to not being fully filled with records. If you accessing several records, do it in ascending order, because this is the most efficient way. Note: This function is deprecated. Use PX_retrieve_record(3) instead RETURN VALUE
Returns 0 on success and -1 on failure. SEE ALSO
PX_get_field(3), PX_get_record(3) AUTHOR
This manual page was written by Uwe Steinmann uwe@steinmann.cx. PX_GET_RECORD2(3)
All times are GMT -4. The time now is 04:18 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy