Comparing data without loosing product position

[LorenzoCapoccia]

Hi,
I have added some description to a triggered error that I done in case where 2 collections row count are different in order to proceed with a comparison.. I am just trying to figure out what's the best way, as I noticed, that the comparison will happen correctly, but the last row ( containing data ) is removed, and so that is not acceptable...
Adding row numbers will that solve the problem or will cause even more errors?


thank you

[PvD_SE]

Hi Lorenzo,

Rather than putting the problem description in a word doc, try and describe it in your thread and insert screenshots if needed. In these days of ransomware, not everyone (I for one) is prepared to risk opening documents from the interweb.

Happy coding!
Paul
Sweden 

[LorenzoCapoccia]

 Hi @PvD_SE,

Thank you for taking your time on this matter.. one more thing I would like to add  for the below comment, as per test, I noticed that if i don't add any note in the empty row, then on the final stage to compare I actually do loose the one row from the main data,... and I rather have an error than loose the data info

I would like to add this option to check if between the main data SDPI and Cognos, there is a mismatch count,  this should not happen in reality.

But if that happens, I have added the option to add (with calculation stage a comment to fill an empty row = Wcode Review Manually

Unfortunately, this doesn’t fix the primary issue, as you see the Material ID Cognos, loses the row and the match NO it is exposed for all rest of material number.

The Wcode word, is added just to ensure no error for missing row appears, but if a solution is found to resolve where the error is

Would you please be able to provide a suggestion and/or a sample that I can view and try to replicate?

MATERIAL_ID SDPI	MATERIAL_ID Cognos	Match?
LCD8281101	LCD8281101	Yes
LCD8280901	LCD8280901	Yes
LCD8280801	LCD8280801	Yes
LCD8280701	LCD8280701	Yes
LCD8280695	LCD8280601	No
LCD8280601	LCD8280501	No
LCD8280501	LCD7521401	No
LCD7521401	LCD7521301	No
LCD7521301	LCD7521201	No
LCD7521201	LCD7521101	No
LCD7521101	LCD7521001	No
LCD7521001	Wcode Review Manually	No

MATERIAL_ID SDPI	MATERIAL_ID Cognos	Match?
LCD8281101	LCD8281101	Yes
LCD8280901	LCD8280901	Yes
LCD8280801	LCD8280801	Yes
LCD8280701	LCD8280701	Yes
LCD8280695	Any comment	No
LCD8280601	LCD8280601	No
LCD8280501	LCD8280501	No
LCD7521401	LCD7521401	No
LCD7521301	LCD7521301	No
LCD7521201	LCD7521201	No
LCD7521101	LCD7521101	No
LCD7521001	LCD7521001	No

Thank you

​​​

[PvD_SE]

Hi Lorenzo,

From what I understand, you have tow collections: one for Cognos data and another for SDPI data. Both have a field called MATERIAL_ID. You want to compare the two collections by matching MATERIAL_ID between the two and write Yes or No in the MATCH? column, depending if a match was made or not.

You write that you loose a row in the comparison. I see no Delete Row action so all rows should be read and compared and none should be removed. As you first count the number of rows and then fill up the lesser collection with dummy rows, you should always have the same number of rows in the collections when you start matching them and therefor not loose any rows.

Your way of matching, by looping through one collection (main) and for each row check if that value also exists in the other collection (secondary), seems solid for your purpose. It will however NOT list any entries in the secondary collection that do not exist in the main collection.
In order to get them too, you might want to update the secondary rows that you were able to match with the main. When you're done, the secondary rows that were not updated represent your unmatched rows.

Hope this helps, otherwise explain more. 🙂

Happy coding!
Paul
Sweden