Feature #8842: Integer#[] with range

• Status: Closed
• Assignee: mame (Yusuke Endoh)
• Target version: -

[mame (Yusuke Endoh)]

=begin
I propose to extend Integer#[] accepting a range.

0b01001101[2, 4] == 0b0011
0bHGFEDCBA[2, 4] == 0bFEDC

== Use case

I believe that everyone has written a code like this:

if (n >> 2) & 0xf == 0x3
  ...
end

because this is a very common idiom in C.
But it is less readable, writable, extendable and optimizable.

if n[2, 4] == 0x3
  ...
end

is much better in the all aspects.

== Corner cases

The current Integer#[] (and shift operators) handle an integer as "a bit array with infinity length";
it returns 0 for any negative index and an (extended) sign bit for any index greater than MSB.
We also can use this standard to define the spec for a range argument.
For example:

15[-1, 42] #=> 30  (equivalent to (15 << 1) && (2 ** 42 - 1))
15[3, 42]  #=>  1  (equivalent to (15 >> 3) && (2 ** 42 - 1))
15[3..Float::INFINITY]  #=> 1  (equivalent to 15 >> 3)
15[-3..Float::INFINITY] #=> 2  (equivalent to 1 << 3)

-1[0..Float::INFINITY]   #=> -1
-1[1..Float::INFINITY]   #=> -1
-1[-1..Float::INFINITY]  #=> -2

1[-Float::INFINITY..0] #=> failed to allocate memory
2[-Float::INFINITY..0] #=> 0

Only tricky case that I thought of is a range (beg..end) whose "end" is smaller than "beg".
I think it should be handled as (beg..Float::INFINITY).

15[-3..-4] #=> 2  (equivalent to 1 << 3)
-1[0..-1] #=> -1
-1[0..-2] #=> -1

What do you think?
=end

[mame (Yusuke Endoh)]

• File integer-with-range.pdf integer-with-range.pdf added

[mame (Yusuke Endoh)]

• Tracker changed from Bug to Feature

[matz (Yukihiro Matsumoto)]

• Assignee changed from matz (Yukihiro Matsumoto) to mame (Yusuke Endoh)

Accepted.

Matz.

I take it that the use of '&&' operator in the first 2 corner cases is a typo. But, pardon my ignorance,
in the 4th corner case, why 15[-3..Float::INFINITY] #=> 2 (equivalent to 1 << 3)?
1 << 3 is 8, and even if I assume a typo, 15 << 3 is 120. Is there something I misunderstood?

[mame (Yusuke Endoh)]

• Target version changed from 2.6 to 2.1.0

I take it that the use of '&&' operator in the first 2 corner cases is a typo.

Yes, sorry.

in the 4th corner case, why 15[-3..Float::INFINITY] #=> 2 (equivalent to 1 << 3)?
1 << 3 is 8, and even if I assume a typo, 15 << 3 is 120. Is there something I misunderstood?

No, you are right.

15[-3..Float::INFINITY] #=> 120 (equivalent to 15 << 3)

is correct.

Also, the attached slide has a bug: n[2..6] should be n[2..5] or n[2...6].

I'm sorry. Haste is from the devil. I'll make a patch more carefully.

--
Yusuke Endoh [e-mail]

[hsbt (Hiroshi SHIBATA)]

• Target version changed from 2.1.0 to 2.2.0

[naruse (Yui NARUSE)]

• Target version deleted (2.2.0)

[mame (Yusuke Endoh)]

• Status changed from Assigned to Closed

Applied in changeset git|6bedbf462544a7917fdc8d8c44276079a6e156cf.

---

numeric.c: Extend Integer#[] to support range arguments

0b01001101[2, 4]  #=> 0b0011
0b01001100[2..5]  #=> 0b0011
0b01001100[2...6] #=> 0b0011
    ^^^^

[Feature #8842]