Hello @andreas_lds,   @andreas_lds wrote: Please don't double post the same question. I have merged your second post with the original one. Thank you for merging the two topics. I'm sorry, this was my first post on this site, and when I created the first topic, I realized 2 hours later that it was gone, so I created a new one, not knowing that the first was considered spam...   Regards, Kevin_T ... View more

Hello @ChrisNZ,   Yes, your snippet also works, I just modified some parts of the code to match my personal code.   @ChrisNZ wrote:   I have not understood how you use ID.   Well, If the ID=A66571, for each new value of the column "name", and if a column (AA for example) is equal to "NO", so all the AA column for ths particular "name" value is set to missing, and so on for BB... So, A66571 is my "Master ID", and the others just follow Master ID's rules.   Thanks for contributing.   Regards, Kevin_T ... View more

Hello @Tom,   I thought I could use the LAG function but it's tricky to use with this kind of problem. I've tried TEMPORARY arrays before posting here, but it didn't work. And I also considered using a BLANK instead of a period.   I see your solution works too, so thanks for your contribution.   Regards, Kevin_T ... View more

Hello @AMSAS, your answer meets my expectations exactly! I was not far from this result, I had not thought of storing the values ​​found in a new variable (here keepVar), indeed, instead of Y or N, I had YES or NO, so I had to adapt the base table with the TRANWRD function, then following your program, repeat the reverse step with TRANWRD.   Thank you very much for your contribution, it helped me a lot!   Regards, Kevin_T ... View more

Hello, I've already seen this topic and the answer, which didn't help me. ... View more

Hello,   I'm trying to find a way to replace the values ​​of some rows, based on a particular row, itself based on the "name" column. I will explain what I would like to achieve (I have tried many possibilities already, but never have the perfect result).   In the "name" column, there are two names, A and B. The "ID" column contains "ID", the important identifier being A66571, so I already put it in the first line, by "name" column. There are then different columns ("AA", "BB"...).   What I want is that if for ID = "A66571", I subsequently have columns = "NO", then I want to enter "." in these same columns, but for EACH ID (including A66571), for this "name" column. But if for the ID A66571, I subsequently have columns = "YES", then the other rows, for these same columns do not change.   Here's what I currently have,      and next what I would like to get : What I never manage to get is, for example, when ID = A66571, and the columns "AA" and "BB" = "YES", for me, when ID = "A08176", the columns "AA" and "BB" are always = "." . I hope you will help me solve this problem, I've been blocking it for quite a few hours (using Arrays, Retain, Lags...). Thank you so much.   Kevin_T ... View more