Caching Issue in Pipe, Makes it appear core Flex.io value is broken

[wanderslth]

Here's a pipe that will fail:

Step 1:

input file: https://raw.githubusercontent.com/flexiodata/sample-data/master/contacts-2016-11.csv 

convert from: delimited to: table delimiter: comma header: true 

select col: givenname, surname, streetaddress, city, state, zipcode, birthday 

filter where: strpart(birthday, "/", 1) = "1980" 

execute lang: python code: aW1wb3J0IHN5cwoKaW5wdXQgPSAnJzsKZm9yIGxpbmUgaW4gc3lzLnN0ZGluOgogICAgaW5wdXQgKz0gbGluZS51cHBlcigpOwogICAgCnN5cy5zdGRvdXQud3JpdGUoaW5wdXQpCg== 

limit value: 5 

OK, looking good, now Step 2:

convert from: delimited to: table delimiter: comma header: true 

Save step returns preview of full file, not the limited (by both rows and fields) file.

You can "Run Pipe" to clear cache and make it work, but a user just thinks the tool fails at its core promise to run logic in sequence.

[alwillia]

Same as caching issue with error codes in #2 and #47.