Thanks for the replies. I am not actually working on a project that needs to do paging. I was really just wonder how to do it best in case I need to do it in the future.
I thought there might be a different way instead of having to manage a list. I guess the main reason why you may not want to use a list is if new records are continually being added or deleted. Keeping a saved list does not guarantee that all the records in that list are still valid. New records could of been created that should be included and old records may be deleted that should not exist in the list anymore. This would mean you would need to continually re perform your search to update your search list. Plus I thought you would have better performance if you could perform a search an only bring back 100 records instead of 1000000 each time. Regards Adrian Halid Senior Analyst/Programmer IT Vision Australia Pty Ltd (ABN: 34 309 336 904) PO Box 881, Canning Bridge WA 6153 Level 3, Kirin Centre, 15 Ogilvie Road, Applecross, WA, 6153 P: (08) 9315 7000 F: (08) 9315 7088 E: adrian.ha...@itvision.com.au W: http://www.itvision.com.au ___________________________________________________________ NOTICE : This e-mail and any attachments are intended for the addressee(s) only and may contain confidential or privileged material. Any unauthorised review, use, alteration, disclosure or distribution of this e-mail (including any attachments) by an unintended recipient is prohibited. If you are not the intended recipient please contact the sender as soon as possible by return e-mail and then delete both messages. ___________________________________________________________ -----Original Message----- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Oaks, Harold Sent: Saturday, 20 February 2010 1:52 AM To: U2 Users List Subject: Re: [U2] TOP/LIMIT Equivalent? Adrian: Here's a dumb little solution that might work for you if you can run a VOC paragraph (I don't know how your interface to the web application is set up). We have Universe. I make use of the standard line editor ED which I think everybody has. In the VOC file create the two paragraphs PREGET and GETRECS: PREGET is these 3 lines: PA SELECT PRODUCTS BY A1 SAVE.LIST LIST1 GETRECS is these 9 lines: PA DELETE.LIST LIST2 EDIT.LIST LIST2 DATA LOAD LIST1 DATA <<I2,FROMLINE>> DATA <<I3,TOLINE>> DATA FILE GET.LIST LIST2 LIST PRODUCTS REQUIRE.SELECT First run PREGET. This selects all the records sorted as you wish and creates the saved list LIST1. You only run this once. Now, for each selection range of records desired, you run GETRECS with 2 parameters, fromline and toline. For example: GETRECS 1 100 will display the first 100 records GETRECS 700 725 will display the 700th thru 725th records Of course the final line "LIST PRODUCTS REQUIRE.SELECT" would display output on the terminal. You'll be doing something else to make the data you wish appear on the website, perhaps using LIST2 as your source for records IDs. Note that REQUIRE.SELECT is necessary in my LIST command to ensure we don't display the entire file if LIST2 is empty. I have a file with about 325000 records. Running PREGET took about 12 seconds. OK, that's a while. But after that, each run of GETRECS was very quick. Even this command: GETRECS 300001 300100 took about 1/3 second to start displaying. Harold Oaks Sr. Analyst/Programmer Clark County WA -----Original Message----- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Adrian Halid Sent: Thursday, February 18, 2010 5:45 PM To: U2 Users List Subject: Re: [U2] TOP/LIMIT Equivalent? If I have a file of 100,000 products and I want to page them in a list of 100 on my website sorted by product name in attribute 1. SELECT PRODUCTS BY A1 This would return me all 100,000 products. So what would be my query when the website loads and I only want the 1st 100 by product name. Then the user would click the "Next" button to display the 2nd 100 (101, 200). What would be my query then? Would I always have to perform the full select and return 100,000 records and then loop through and only show records 101 to 200. Isn't there a better way so that you only return the 100 records you want to view. Regards Adrian Halid Senior Analyst/Programmer IT Vision Australia Pty Ltd (ABN: 34 309 336 904) PO Box 881, Canning Bridge WA 6153 Level 3, Kirin Centre, 15 Ogilvie Road, Applecross, WA, 6153 P: (08) 9315 7000 F: (08) 9315 7088 E: adrian.ha...@itvision.com.au W: http://www.itvision.com.au ___________________________________________________________ NOTICE : This e-mail and any attachments are intended for the addressee(s) only and may contain confidential or privileged material. Any unauthorised review, use, alteration, disclosure or distribution of this e-mail (including any attachments) by an unintended recipient is prohibited. If you are not the intended recipient please contact the sender as soon as possible by return e-mail and then delete both messages. ___________________________________________________________ -----Original Message----- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Dan Fitzgerald Sent: Friday, 19 February 2010 9:29 AM To: 'U2 Users List' Subject: Re: [U2] TOP/LIMIT Equivalent? You could save the list, and edit it. But, really, there will be no logical relationship between these records. The 1st hundred is just as random as the 2nd or 3rd hundred, unless your modulus is an extremely ill-advised one. Again, they are hashed records. What we mean by this is that UV will take your item id, extract a numeric equivalent from it (i.e., a238rs45 becomes 23845), then divides by the modulo (a simplification, but reasonably illustrative for our purposes). It takes the resulting remainder (if your modulus was 101, this will be 9), adds 1 (because you can have a remainder of 0), and puts that record into group #10. Unless you know your precise hashing algorithm (which was proprietary, last time I checked), you simply don't know where the record will land. (OK, if you have sequential numeric id's, and you use group.stat.detail to analyze the file, you might be able to predict it, but whatever algorithm you cook up goes out the window if you resize or change the file type, and that's a lot of unnecessary work, anyway). If you want a random sample, use SAMPLE. If you need a logically related group (the last 100 records added, for example), then you should apply selection criteria (SSELECT FILE WITH TIMESTAMP GE xxxx BY-DSND TIMESTAMP). The logic is in the data. -----Original Message----- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Adrian Halid Sent: Thursday, February 18, 2010 8:06 PM To: U2 Users List Subject: Re: [U2] TOP/LIMIT Equivalent? But how would you get the 2nd 100 (101 - 200) and the 3rd 100 (301 to 400) without writing a subroutine to do it? Regards Adrian Halid Senior Analyst/Programmer IT Vision Australia Pty Ltd (ABN: 34 309 336 904) PO Box 881, Canning Bridge WA 6153 Level 3, Kirin Centre, 15 Ogilvie Road, Applecross, WA, 6153 P: (08) 9315 7000 F: (08) 9315 7088 E: adrian.ha...@itvision.com.au W: http://www.itvision.com.au ___________________________________________________________ NOTICE : This e-mail and any attachments are intended for the addressee(s) only and may contain confidential or privileged material. Any unauthorised review, use, alteration, disclosure or distribution of this e-mail (including any attachments) by an unintended recipient is prohibited. If you are not the intended recipient please contact the sender as soon as possible by return e-mail and then delete both messages. ___________________________________________________________ -----Original Message----- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of fft2...@aol.com Sent: Friday, 19 February 2010 4:55 AM To: u2-users@listserver.u2ug.org Subject: Re: [U2] TOP/LIMIT Equivalent? In a message dated 2/18/2010 10:27:46 AM Pacific Standard Time, cla...@hme.com writes: > You can sort first and then SAMPLE, too.>> Caryl has the idea here if you really want the "Last" 100. You can Sort BY-DSND which means "by descending" so if you want the last alphabetically or the last by ID or the last by some other field you can sort on that and then Sample 100 ? I've NEVER tried this. Does it work? If that doesn't work you can EXECUTE a sort by dsnd and then simply write a brief BASIC routine to pull off the "first" 100 of that list and output it! Will "Where There's a Way there's a Will... or something" Johnson _______________________________________________ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users _______________________________________________ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users _______________________________________________ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users _______________________________________________ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users _______________________________________________ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users _______________________________________________ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
