TrimLeft doesn't like the word "Egg" ?!?!

Question

Ok, this is a bit of as strange query. I am filtering a dataset of various foods that have been eaten at various times of the day. I end up with a list of strings in the form of "Morning-Vegetables", "Afternoon-Vegetables", "Evening-Vegetables" etc.. A formula I am using needs to trim the time of day before the foodtype, so I have 3 formulas which use the functions TrimLeft(String, "Morning-"), TrimLeft(String, "Afternoon-"), TrimLeft(String, "Evening-").

Everything works fine for various types of food and various times of day. Except for "Evening-Egg". I have attached a small sample showing what happens. I have changed "Egg" to "Eggs" to illustrate the point a bit more clearly.

If I do Trimleft(String, "Afternoon-") on "Afternoon-Eggs" = "Eggs"

If I do Trimleft(String, "Evening-") on "Evening-Eggs" = "s"   ?????

If I do Trimleft(String, "Evening") on "Evening-Eggs" = "-Eggs"

I am using Alteryx 10.06.28907.

Like I said, TrimLeft works fine with Coffee, Beer, Fish, Meat, I just seems to have a problem with "Egg" preceded by the word "Evening-".

I know this seems like the request of a crazy person, but is anyone else getting this? The last value of the second column of the attached reads "s" for me, whereas I am expecting "Eggs".

If I am doing something really silly, then sorry!

Pete

Eggs.yxmd

Joe_Mako · Accepted Answer

The Trim functions, take in two arguments, the string to trim, and a list of character to replace. The Trim functions do not work like a Replace function (where the sequence of characters is respected), the list of characters can be in any order, and they will be trimmed.

for example,

TrimLeft([Names],"Evening")

is the exact same thing as

TrimLeft([Names],"gnievE")

Another function to consider is Regex_Replace, then in the pattern, we can specify that the sequence of character to replace have to be at the beginning of the string (with the ^ character), and this will work similar to the replace function.

For example:

Regex_Replace([Names],"^Evening-","")

and

Regex_Replace([Names],"^Afternoon-","")

I think are closer to what you are looking for.

Another tool to consider is Text-to-Columns, to split the string on the dash "-" character.

Take a look at the attached workflow with these additions.