[ruby-core:81691] [Ruby trunk Bug#13663] `String#upto` doesn't work as expected

[zverok.offline@gmail.com]

Issue #13663 has been updated by zverok (Victor Shepelev).


I believe that problem here is how to provide consistency between `succ` and `<=>` for arbitrary length strings.

1. For most of the real use cases, `'x' > 'ac'` is sane (like sorting strings);
2. Things using `succ` (like `upto` and ranges) should check that begin is lower than end;

So... I believe that it is only reasonable to have (1) and (2), though sometimes it leads to "inconsistencies", like described above. If you do a lot of stuff with making ranges from "x" to "ac" it is probably better to have dedicated value object class, with redefined `<=>` and `succ`

----------------------------------------
Bug #13663: `String#upto` doesn't work as expected
https://bugs.ruby-lang.org/issues/13663#change-65380

* Author: sos4nt (Stefan Schüßler)
* Status: Open
* Priority: Normal
* Assignee: 
* Target version: 
* ruby -v: ruby 2.4.1p111 (2017-03-22 revision 58053) [x86_64-darwin15]
* Backport: 2.2: UNKNOWN, 2.3: UNKNOWN, 2.4: UNKNOWN
----------------------------------------
Given that `String#upto` uses `String#succ` to generate successive values, I'd expect

```ruby
'x'.upto('ac').to_a #=> []
```

to return:

```ruby
["x", "y", "z", "aa", "ab", "ac"]
```

Instead, an empty array is returned.

This seems to depend on whether the the receiver is greater than the argument or not:

```ruby
'x' <=> 'ac' #=> 1
```

It works just fine in this case:

```ruby
'b'.upto('ca').to_a
#=> ["b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n",
#    "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "aa",
#    "ab", "ac", "ad", "ae", "af", "ag", "ah", "ai", "aj", "ak", "al",
#    "am", "an", "ao", "ap", "aq", "ar", "as", "at", "au", "av", "aw",
#    "ax", "ay", "az", "ba", "bb", "bc", "bd", "be", "bf", "bg", "bh",
#    "bi", "bj", "bk", "bl", "bm", "bn", "bo", "bp", "bq", "br", "bs",
#    "bt", "bu", "bv", "bw", "bx", "by", "bz", "ca"]
```

Presumably because of:

```ruby
'b' <=> 'ca' #=> -1
```




-- 
https://bugs.ruby-lang.org/