About mipate

mipate · ‎06-23-2020

Yes indeed it is possible. Step 1: Create a parameter in the data tab and give it some name (for this case lets say you have two columns in your table name A and B where both are text and you want to have the search bar index on both of them). Step 2: Create a new calculated column and name it something along the lines of rowShouldAppearInSearch. We want this to be a text column, 'true' if either A or B in the current row contains the current value of the parameter, false otherwise. To do this, have an if statement. If the length of the parameter is 0, return true (no parameter means we have no search restrictions). Else, if A contains parameter or B contains parameter, return true. Else, return false. Step 3: Update your table to filter based on the calculated columns. So select the parameter to filter on, and then choose to filter to only include values of rowShouldAppearInSearch which are true. Let me know if you get stuck on any of this. Otherwise, I hope I was able to help you out!

mipate · ‎06-19-2020

UPDATE: I have found a solution to the issue. I figure that I will not be the last one to run into the same difficulties, so here is where I went wrong. I was opening my csv files in excel prior to uploading and the column that had the date in it seemed to be formatted fine (which is technically was). However, if I used a text edit to manually inspect the csv itself, I found that each date entry was enclosed in quotations. I didn't initially think this was a problem since VA allows for csv uploads and the example datetime format had a comma in it (which implies that the entire datetime would need to be enclosed in quotations to avoid the comma being treated as the next cell). However, it turns out that this was exactly the problem. So my solution was to change the format to MM/DD/YYYY HH:MM:SS AM on my raw data and then upload that to VA. However, that doesn't quite mark the end of the solution; Visual Analytics still read my datetime as a string. However, making a new calculated column and casting the string to a date time worked as intended this time around. If anyone ever runs into a similar issue, let me know and I can try help! (Would also recommend talking to support, they were the incredible ones who knew this was the issue!)

mipate · ‎06-18-2020

I have a date time in a CSV, and VA recognizes it as a character when I upload the csv. Okay, so then I'm trying to backtrace as to what im doing wrong since even when I attempt to format/parse it into a datetime, it just shows up as an empty dot for each row of the newly created calculated column. So then I go back to the simplest case and use the parse function to parse the example date that VA gives: Parse('June 25, 2014 08:15:00 AM', 'DATETIME40.') But this fails and it still shows a dot in every row for this calculated column! How can this possibly be the case that I am using the example that VA gives and it is unable to parse it?

Online Status	Offline
Date Last Visited	‎06-23-2020 04:38 PM

Re: Text Input linked to multiple categories in SAS VA

Re: VA format is not working with parsing a datetime

VA format is not working with parsing a datetime

Re: Text Input linked to multiple categories in SAS VA

Re: Text Input linked to multiple categories in SAS VA

Re: VA format is not working with parsing a datetime

VA format is not working with parsing a datetime