It's been a while so you've probably learned this, but you can change the tests.

You need to check the length of the output string, not the input string, this will fail on valid inputs with sufficiently large whitespace.

list is a python keyword, so change it to lst or l or something similar that isn't a python keyword or module.

It's alloways great to make more katas for this platform, but I feel like this idea has been done hundereds of times over. Maybe try adding your own little spin and some random tests.

There weren't tests like that?! I just wasted time then.

I honestly expected the test cases to be better with some of the values being "[", "'Hidden'", "'", ",", " " etc.

Seems good now

Can't find it now

Ok, now it's 100% fixed. The problem was that the numbers generated were completely random insted of being from the charaters list used for the other tests. All ASCII codes should now translate to a charater in string.ascii_letters, string.punctuation, or be a space.

I don't see any changes:

input: 6770759978108109981011166512186678767788367731217810077978310377115109887177109898089761209812082103111103110701131119970114821107310011786681188410874116661221119011910788711141171108672118851068510811970118102819811969988998114651198810065791061061111021158685767573117847483751051051068687979098103677772112103111114866510183104112117658411112197113120

'ʥKϥ̪̪ϦetʋÚʛͬ̊̈́̅y̍\x07̝̿%ˇǾϜˍˆϝ̨ϐxϕÐÒķnķFqoϥ\x0bǢn˛\x01²ʜv͉Wƛʚzo΅¿NͧrunͣÓ͵j͓QÅ\x0b̪ęϕÄϦ\u0383̫Ǒwͱ\x06Ƀjjofsͤɀ˵ķ̐ˬŷiij̝ͤ\u038d̪ů̉ÓÒķoǦʋ\x12ĶƛÓ˽͉oÛˇĸ' 
should equal
'ʥKϥ̪̪ϦetʋÚʛͬ̊̈́̅y̍\x07̝̿%ˇǾϜˍˆϝ̨ϐxϕÐÒķnķFqoϥ\x0bǢn˛\x01²ʜv͉Wƛʚzo΅¿NͧrunͣÓ͵j͓QÅ\x0b̪ęϕÄϦ\u0383̫Ǒwͱ\x06Ƀjjofsͤɀ˵ķ̐ˬŷiij̝ͤ\u038d̪ů̉ÓÒķoǦʋ\x12ĶƛÓ˽͉oÛˇĸ\x00'

Definitely not ascii... based on the description the input should look like "067070075099...`

Could you link the duplicate?

Fixed, just found out that some characters in string.punctuation in Python were registered incorrectly like '\x04' (End Of Transmission). All string lengths should be divisible by 3 now.

Fixed, some charaters used were registered incorrectly like '\x04' (End Of Transmission).

Seems to be a good start, although it would need some polishing (e.g. better tests). That is, if it was not a duplicate...

Random tests seem pretty broken, e.g.:

input: "6546747872114478338596911711288681098910310411941109107103427679124113100801046437696767981017197901061026269123478010233939811440451008395893868108784659726284781081206899923637781188611610071367411661624512367728567631251139711864"

'ʎʢ̓Óƿ́͛ʳ«\x80ͤmͻ\x1f)ÂnΎˆŖ˿ΐƛĶPhʃ́ʤ̞eˏ̖jfɲΐś̡\x17Ɖϕ\x90-d͇ɍƂ̪ͮǑόɴǞlxʱϠū̊v͝¡\x07\x88˥¦¢ǃì̄͘˻}qϋº'
should equal
'ʎʢ̓Óƿ́͛ʳ«\x80ͤmͻ\x1f)ÂnΎˆŖ˿ΐƛĶPhʃ́ʤ̞eˏ̖jfɲΐś̡\x17Ɖϕ\x90-d͇ɍƂ̪ͮǑόɴǞlxʱϠū̊v͝¡\x07\x88˥¦¢ǃì̄͘˻}qϋº\x04'

• string size not divisible by 3
• expected output is not ascii