View previous topic :: View next topic
|
Author |
Message |
preethan Warnings : 1 New User
Joined: 15 Dec 2008 Posts: 33 Location: Chennai
|
|
|
|
I am facing some hurdles in SERENA comparex.
Consider I am having two files
SYSUT1 with data :
ABC
DEF
XYZ
SYSUT2 with data :
XYZ
DEF
ABC
Say i have a key = (1,3)
Now what happens here, 'ABC' from SYSUT1 is checked with the keys of SYSUT2. It encounters 'XYZ' and 'DEF', before getting 'ABC' in SYSUT2. Now 'DEF' is checked with keys of SYSUT2. Since SYSUT2 has already reached end of the file, This is considered as an insert. Similar thing happens to 'XYZ' in sysut1 and 'DEF' , 'XYZ' is sysut2. So it will show like 2 inserts in sysut1 and 2 inserts in sysut2.
To overcome this problem i used to sort both files. Is there any option in comparex to overcome the problem without sorting ? |
|
Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007 Posts: 10873 Location: italy
|
|
|
|
why not look at the comparex manual Yourself ? |
|
Back to top |
|
|
preethan Warnings : 1 New User
Joined: 15 Dec 2008 Posts: 33 Location: Chennai
|
|
|
|
I have gone through the manual already. Since am not finding any, I am posting it here to check if any one knows. |
|
Back to top |
|
|
Akatsukami
Global Moderator
Joined: 03 Oct 2009 Posts: 1788 Location: Bloomington, IL
|
|
|
|
preethan wrote: |
I have gone through the manual already. Since am not finding any, I am posting it here to check if any one knows. |
Why do you think that there might be a manual shebichtav and a manual shebe'al peh? |
|
Back to top |
|
|
preethan Warnings : 1 New User
Joined: 15 Dec 2008 Posts: 33 Location: Chennai
|
|
|
|
i am not understanding what you are talking. If people don't know the answer , they can keep quiet instead of posting something irrelevant to the topic |
|
Back to top |
|
|
preethan Warnings : 1 New User
Joined: 15 Dec 2008 Posts: 33 Location: Chennai
|
|
|
|
The examples which i mentioned was purely to illustrate. I have been involved in comparing two big files. So i am fed up comparing both. Key everything looks fine. But i am getting inserts in sysut1 and 2. Now after sorting both files i am not getting any differences. It makes me puzzled and need to know why.
preethan wrote: |
Consider I am having two files
SYSUT1 with data :
ABC
DEF
XYZ
SYSUT2 with data :
XYZ
DEF
ABC
Say i have a key = (1,3)
Now what happens here, 'ABC' from SYSUT1 is checked with the keys of SYSUT2. It encounters 'XYZ' and 'DEF', before getting 'ABC' in SYSUT2. Now 'DEF' is checked with keys of SYSUT2. Since SYSUT2 has already reached end of the file, This is considered as an insert. Similar thing happens to 'XYZ' in sysut1 and 'DEF' , 'XYZ' is sysut2. So it will show like 2 inserts in sysut1 and 2 inserts in sysut2.
To overcome this problem i used to sort both files. Is there any option in comparex to overcome the problem without sorting ? |
|
|
Back to top |
|
|
Akatsukami
Global Moderator
Joined: 03 Oct 2009 Posts: 1788 Location: Bloomington, IL
|
|
|
|
Irrelevant to the topic or exposing your ignorance? |
|
Back to top |
|
|
Bill Woodger
Moderator Emeritus
Joined: 09 Mar 2011 Posts: 7309 Location: Inside the Matrix
|
|
|
|
Try google (for your latest question).
Unless it is your own board, don't try telling people what they can and can't do.
Does Comparex have an option to "Do what I want, not what you do"? Your data is messed up. If it should match, it doesn't. That is not Comparex's fault.
I'm sure you'll find, in the manual, various ways to select records to compare, parts of records to compare. I doubt you'll find something for "the file is totally screwed, give me a clean compare anyway, I wanna get on".
Sorting the file to get a clean compare does not sound like a good idea to me.
The files either should match, subject to what you can specify, or they shouldn't match, so you'd not bother to compare them.
If you can be clear about what you are really trying to do, maybe you get something, though you're the first person I've seen mention Comparex here, so you may even then be out of luck.
I used Comparex a long time ago. If the file was pickled, it would act exactly as you describe. Good indicator to look at the problem causing the pickling, rather than try to mess around with Comparex control cards.
EDIT: I see things have moved on. |
|
Back to top |
|
|
dbzTHEdinosauer
Global Moderator
Joined: 20 Oct 2006 Posts: 6966 Location: porcelain throne
|
|
|
|
oh boy, it is whinny-ass day again.
you have to use the comparex -psychic compare parm
EDIT: i was also a victim of slow fingers. |
|
Back to top |
|
|
Bill Woodger
Moderator Emeritus
Joined: 09 Mar 2011 Posts: 7309 Location: Inside the Matrix
|
|
|
|
OK, you have a very simple situation.
Although the data is the same on both files, the order of the data is not the same. The further into the compare it gets, the more "spurious" inserts and deletes will be reported.
You have to concentrate on where it first went wrong, fogetting that later everything looks a mess. You have at least one record out of sequence, and it is possible that that is all. Or a handful or a hundred. You have to find out. Look at the first few errors, identify what is common, start digging.
If there has been a SORT step along the way in producing the files, and on one side there is EQUAL and the other NOEQUALS, or both NOEQUALS, then that is a very easy way to achieve this effect.
EDIT: I always ran Comparex with a limit to the number of errors to report. If the files were clean, it would get to the end anyway. If not, you don't get the great pile of junk after the useful stuff that you never look at anyway. |
|
Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007 Posts: 10873 Location: italy
|
|
|
|
Quote: |
i am not understanding what you are talking. If people don't know the answer , they can keep quiet instead of posting something irrelevant to the topic |
Akatsukami Sama knows very well what he is talking about ...
anyway He was just asking for what reason You think that there should be two versions of the same manual as there are two versions of the...
Torah Shebichtav vs. Torah Sheb'al Peh |
|
Back to top |
|
|
preethan Warnings : 1 New User
Joined: 15 Dec 2008 Posts: 33 Location: Chennai
|
|
|
|
Bill Woodger wrote: |
OK, you have a very simple situation.
Although the data is the same on both files, the order of the data is not the same. The further into the compare it gets, the more "spurious" inserts and deletes will be reported.
You have to concentrate on where it first went wrong, fogetting that later everything looks a mess. You have at least one record out of sequence, and it is possible that that is all. Or a handful or a hundred. You have to find out. Look at the first few errors, identify what is common, start digging.
If there has been a SORT step along the way in producing the files, and on one side there is EQUAL and the other NOEQUALS, or both NOEQUALS, then that is a very easy way to achieve this effect.
EDIT: I always ran Comparex with a limit to the number of errors to report. If the files were clean, it would get to the end anyway. If not, you don't get the great pile of junk after the useful stuff that you never look at anyway. |
You have gotten my problem. The problem here is both are unloads taken from a table at two instances. So the order in which they are written in the unload file differs in both, eventhough the data is similar. So everytime i compare both i have to run sort for these two files to get exact results. I was so confused as why this was happening this way. Thats the reason I came here. I know that my data is correct, I also know what the problem is (order is not in sync in both), so I thought if there are people who have a specialized knowledge in comparex would clear my doubts.
@ all .. thanks fr replyin back. its not that i never did any kind of research from my side. I did, but i still could not get a clue. So I was seeking the experts help here. |
|
Back to top |
|
|
Bill Woodger
Moderator Emeritus
Joined: 09 Mar 2011 Posts: 7309 Location: Inside the Matrix
|
|
|
|
If you said all this up front, you'd maybe have had an easier ride... |
|
Back to top |
|
|
|