no more array insensitive dots as recent 8th update corrects array to…

[axtens]

… string conversion issue

So I thought this had been dealt with. But I was just trying to update my repo to the upstream and go sync error

[glennj]

Too much deleted for grains.

[glennj]

too much deleted for pythagorean-triplet

[glennj]

too much deleted for series

[glennj]

sieve too

[axtens]

Okay, so what's the best way of handling this?

[glennj]

https://github.com/exercism/8th/blob/main/libs/README.md

[axtens]

Not quite what I meant. I'll have another look tonight (Currently 14:19 UTC+8)

[glennj]

Did I misunderstand "what's the best way of handling this"? Make the change in libs/exercism/test and copy it to the exercises. That way, you minimize editing errors.

[axtens]

I'm not disputing your advice. It's correct. The difficulty is with my understand of the situation. I have a test library in 8th/libs/exercism/test that has, on line 67-68

   \ otherwise fall back to 'lazy evaluation'
   l: =

and it is this line that the other exercises are bombing on. So will copying this file to those exercises fix the problem?

[axtens]

I didn't write the test library.

[glennj]

That came from Ron:
https://8th-dev.com/forum/index.php?topic=2745.msg15984#msg15984

I see in the words list:

l:	--
	IMMEDIATE needs utils/latebind Makes the next word "delayed lookup". Compiles the string  into the word currently being compiled. At runtime, every time the code is invoked, it looks up the word with that name. So l: + will add two numbers, or concatenate two strings, using the same code.

Try adding needs utils/latebind to the test file