PICK and POKE of strings have off-by-one error with indexes < 1.

[Siskin-Bot]

Submitted by: BrianH

PICK and POKE of string types skip the 0 index, and negative indexes are shifted accordingly. PICK and POKE of block types work correctly.

; PICK of blocks

>> b: skip [a b c] 2
== [c]
>> pick b 2
== none
>> pick b 1
== c
>> pick b 0
== b
>> pick b -1
== a

; PICK of strings

>> b: skip "abc" 2
== "c"
>> pick b 2
== none
>> pick b 1
== #"c"
>> pick b 0
== none  ; should be #"b"
>> pick b -1
== #"b"  ; should be #"a"

---

^{Imported from: CureCode [ Version: alpha 33 Type: Bug Platform: All Category: n/a Reproduce: Always Fixed-in:alpha 55 ]
Imported from: metaeducation#608}

Comments:

Rebolbot commented on Feb 9, 2009:

Submitted by: BrianH

Note that this bug exists in R2, for blocks as well. This needs to be fixed - it's like picking and poking in Roman numerals.

---

Rebolbot commented on May 28, 2009:

Submitted by: Carl

See also #857

---

Rebolbot commented on May 28, 2009:

Submitted by: BrianH

See also #609 - the same bug for AT.

---

Rebolbot mentioned this issue on Jan 12, 2016:
Make VECTOR! of same index-base as other series
String Pick incompatible with block Pick

---

Rebolbot added Type.bug and Status.important on Jan 12, 2016

---

Oldes added a commit to Oldes/Rebol3 that referenced this issue on Jun 13, 2018:
FEAT/FIX: Picking the right PICK

---

Oldes mentioned this pullrequest on Jun 13, 2018:
FEAT/FIX: Picking the right PICK

---

Hostilefork added a commit to metaeducation/ren-c that referenced this issue on Jun 14, 2018:
Restore Rebol2-era PICKing/PATHing of negative/zero indices

---

Hostilefork mentioned this pullrequest on Jun 14, 2018:
Restore Rebol2-era PICKing/PATHing of negative/zero indices

---