Inconsistent treatment of trailing line breaks/spaces in a literal string

[na4zagin3]

In the current implementation (SATySFi 0.0.5), trailing line breaks in a literal trimmed inconsistently.

`string`# style literal removes up to one trailing line break (U+000A), `string`# style literal up to two line breaks. Anyway, spaces (U+0020) followed by the trailing line breaks won't be trimmed.

Hereinafter ⏎ represents a line break character (U+000A) and ␣ a space (U+0020).

Literal	Interpreted
`a␣⏎`# | a␣
`a␣⏎⏎`# | a␣⏎
`a␣⏎⏎⏎`# | a␣⏎⏎
`a␣⏎` | a␣
`a␣⏎⏎` | a␣
`a␣⏎⏎⏎` | a␣⏎
`a␣⏎⏎⏎⏎` | a␣⏎⏎

I would expect one of the following behaviors:

• Both trailing line breaks and spaces are equally treated. I.e., `string`# style literal preserves any trailing line breaks or spaces while `string`# style trims all the trailing line breaks and spaces.

  For example, `a␣⏎␣⏎`# and `a␣⏎␣⏎` will represent a␣⏎␣⏎ and a, respectively.

• When there are trailing line breaks following spaces, the spaces won't be removed.

  • `string`# style literal preserves all the trailing line breaks while `string`# style trims all the trailing line breaks.

  For example, `a⏎⏎⏎`# and `a⏎⏎⏎` will represent a⏎⏎⏎ and a, respectively.

  • `string`# style literal preserves up to one trailing line break while `string`# style trims all the trailing line breaks.

  For example, `a⏎⏎⏎`# and `a⏎⏎⏎` will represent a⏎⏎ and a, respectively.

Original comment: https://satysfi.slack.com/archives/CTGUT75FZ/p1584115832024200

[na4zagin3]

Oops, this duplicates #233. I'm so sorry.