mirrors/Nim
1
0
Fork 0
mirror of https://github.com/nim-lang/Nim.git synced 2026-04-19 05:50:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

strutils.split/rsplit now return src on an empty sep (#22136)

Browse Source 
This is a rebase of an earlier rejected PR. Following the discussion
around it, this commit provides a valid output for and edge case
of an empty separator for `split` and `rsplit` routines. The empty
separator is interpreted as "split by no separators" and the initial
string is returned. This is consistent with the behaviour of the
`set[char]` version of `split`/`rsplit` routines and unifies them all.

Compared to a commit merged earlier, this one has a benefit of
not using assertions that will be removed in release builds
and thus still not preventing possible infinite loops (which was the
earlier behaviour for this edge case for separator of type `string`).

Co-authored-by: Andreas Rumpf <rumpf_a@web.de>
(cherry picked from commit 5e529b3bfa)
This commit is contained in:
Zoom
2023-06-21 06:52:33 +00:00
committed by narimiran
parent 9d573f4a1d
commit 6863b5d289
4 changed files with 59 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

45
lib/pure/strutils.nim
View File

@@ -346,11 +346,14 @@ func cmpIgnoreStyle*(a, b: string): int {.rtl, extern: "nsuCmpIgnoreStyle".} =
# --------- Private templates for different split separators -----------
func substrEq(s: string, pos: int, substr: string): bool =
var i = 0
# Always returns false for empty `substr`
var length = substr.len
while i < length and pos+i < s.len and s[pos+i] == substr[i]:
inc i
return i == length
if length > 0:
var i = 0
while i < length and pos+i < s.len and s[pos+i] == substr[i]:
inc i
i == length
else: false
template stringHasSep(s: string, index: int, seps: set[char]): bool =
s[index] in seps
@@ -492,12 +495,17 @@ iterator split*(s: string, sep: string, maxsplit: int = -1): string =
## "is"
## "corrupted"
##
## .. note:: Empty separator string results in returning an original string,
## following the interpretation "split by no element".
##
## See also:
## * `rsplit iterator<#rsplit.i,string,string,int,bool>`_
## * `splitLines iterator<#splitLines.i,string>`_
## * `splitWhitespace iterator<#splitWhitespace.i,string,int>`_
## * `split func<#split,string,string,int>`_
splitCommon(s, sep, maxsplit, sep.len)
let sepLen = if sep.len == 0: 1 # prevents infinite loop
else: sep.len
splitCommon(s, sep, maxsplit, sepLen)
template rsplitCommon(s, sep, maxsplit, sepLen) =
@@ -567,6 +575,9 @@ iterator rsplit*(s: string, seps: set[char] = Whitespace,
##
## Substrings are separated from the right by the set of chars `seps`
##
## .. note:: Empty separator set results in returning an original string,
## following the interpretation "split by no element".
##
## See also:
## * `split iterator<#split.i,string,set[char],int>`_
## * `splitLines iterator<#splitLines.i,string>`_
@@ -592,12 +603,17 @@ iterator rsplit*(s: string, sep: string, maxsplit: int = -1,
##
## Substrings are separated from the right by the string `sep`
##
## .. note:: Empty separator string results in returning an original string,
## following the interpretation "split by no element".
##
## See also:
## * `split iterator<#split.i,string,string,int>`_
## * `splitLines iterator<#splitLines.i,string>`_
## * `splitWhitespace iterator<#splitWhitespace.i,string,int>`_
## * `rsplit func<#rsplit,string,string,int>`_
rsplitCommon(s, sep, maxsplit, sep.len)
let sepLen = if sep.len == 0: 1 # prevents infinite loop
else: sep.len
rsplitCommon(s, sep, maxsplit, sepLen)
iterator splitLines*(s: string, keepEol = false): string =
## Splits the string `s` into its containing lines.
@@ -708,6 +724,9 @@ func split*(s: string, seps: set[char] = Whitespace, maxsplit: int = -1): seq[
## The same as the `split iterator <#split.i,string,set[char],int>`_ (see its
## documentation), but is a func that returns a sequence of substrings.
##
## .. note:: Empty separator set results in returning an original string,
## following the interpretation "split by no element".
##
## See also:
## * `split iterator <#split.i,string,set[char],int>`_
## * `rsplit func<#rsplit,string,set[char],int>`_
@@ -716,6 +735,7 @@ func split*(s: string, seps: set[char] = Whitespace, maxsplit: int = -1): seq[
runnableExamples:
doAssert "a,b;c".split({',', ';'}) == @["a", "b", "c"]
doAssert "".split({' '}) == @[""]
doAssert "empty seps return unsplit s".split({}) == @["empty seps return unsplit s"]
accResult(split(s, seps, maxsplit))
func split*(s: string, sep: string, maxsplit: int = -1): seq[string] {.rtl,
@@ -725,6 +745,9 @@ func split*(s: string, sep: string, maxsplit: int = -1): seq[string] {.rtl,
## Substrings are separated by the string `sep`. This is a wrapper around the
## `split iterator <#split.i,string,string,int>`_.
##
## .. note:: Empty separator string results in returning an original string,
## following the interpretation "split by no element".
##
## See also:
## * `split iterator <#split.i,string,string,int>`_
## * `rsplit func<#rsplit,string,string,int>`_
@@ -737,8 +760,7 @@ func split*(s: string, sep: string, maxsplit: int = -1): seq[string] {.rtl,
doAssert "a largely spaced sentence".split(" ") == @["a", "", "largely",
"", "", "", "spaced", "sentence"]
doAssert "a largely spaced sentence".split(" ", maxsplit = 1) == @["a", " largely spaced sentence"]
doAssert(sep.len > 0)
doAssert "empty sep returns unsplit s".split("") == @["empty sep returns unsplit s"]
accResult(split(s, sep, maxsplit))
func rsplit*(s: string, sep: char, maxsplit: int = -1): seq[string] {.rtl,
@@ -788,6 +810,9 @@ func rsplit*(s: string, seps: set[char] = Whitespace,
## .. code-block:: nim
## @["Root#Object#Method", "Index"]
##
## .. note:: Empty separator set results in returning an original string,
## following the interpretation "split by no element".
##
## See also:
## * `rsplit iterator <#rsplit.i,string,set[char],int>`_
## * `split func<#split,string,set[char],int>`_
@@ -815,6 +840,9 @@ func rsplit*(s: string, sep: string, maxsplit: int = -1): seq[string] {.rtl,
## .. code-block:: nim
## @["Root#Object#Method", "Index"]
##
## .. note:: Empty separator string results in returning an original string,
## following the interpretation "split by no element".
##
## See also:
## * `rsplit iterator <#rsplit.i,string,string,int,bool>`_
## * `split func<#split,string,string,int>`_
@@ -829,6 +857,7 @@ func rsplit*(s: string, sep: string, maxsplit: int = -1): seq[string] {.rtl,
doAssert "".rsplit("Elon Musk") == @[""]
doAssert "a largely spaced sentence".rsplit(" ") == @["a", "",
"largely", "", "", "", "spaced", "sentence"]
doAssert "empty sep returns unsplit s".rsplit("") == @["empty sep returns unsplit s"]
accResult(rsplit(s, sep, maxsplit))
result.reverse()

7
tests/js/tstdlib_various.nim
View File

@@ -150,12 +150,7 @@ block tsplit2:
s.add("#")
s.add(w)
var errored = false
try:
discard "hello".split("")
except AssertionDefect:
errored = true
doAssert errored
doAssert "true".split("") == @["true"]
block txmlgen:
var nim = "Nim"

7
tests/stdlib/tstdlib_various.nim
View File

@@ -26,7 +26,6 @@ Hi Andreas! How do you feel, Rumpf?
[2, 3, 4, 5]
[2, 3, 4, 5, 6]
[1, 2, 3, 4, 5, 6]
true
<h1><a href="http://force7.de/nim">Nim</a></h1>
'''
"""
@@ -206,11 +205,7 @@ block tsplit2:
s.add("#")
s.add(w)
try:
discard "hello".split("")
echo "false"
except AssertionDefect:
echo "true"
doAssert "true".split("") == @["true"]

22
tests/stdlib/tstrutils.nim
View File

@@ -51,6 +51,15 @@ template main() =
doAssert s.split(maxsplit = 4) == @["", "this", "is", "an", "example "]
doAssert s.split(' ', maxsplit = 1) == @["", "this is an example "]
doAssert s.split(" ", maxsplit = 4) == @["", "this", "is", "an", "example "]
# Empty string:
doAssert "".split() == @[""]
doAssert "".split(" ") == @[""]
doAssert "".split({' '}) == @[""]
# Empty separators:
doAssert "".split({}) == @[""]
doAssert "".split("") == @[""]
doAssert s.split({}) == @[s]
doAssert s.split("") == @[s]
block: # splitLines
let fixture = "a\nb\rc\r\nd"
@@ -61,12 +70,21 @@ template main() =
block: # rsplit
doAssert rsplit("foo bar", seps = Whitespace) == @["foo", "bar"]
doAssert rsplit(" foo bar", seps = Whitespace, maxsplit = 1) == @[" foo", "bar"]
doAssert rsplit(" foo bar ", seps = Whitespace, maxsplit = 1) == @[
" foo bar", ""]
doAssert rsplit(" foo bar ", seps = Whitespace, maxsplit = 1) == @[" foo bar", ""]
doAssert rsplit(":foo:bar", sep = ':') == @["", "foo", "bar"]
doAssert rsplit(":foo:bar", sep = ':', maxsplit = 2) == @["", "foo", "bar"]
doAssert rsplit(":foo:bar", sep = ':', maxsplit = 3) == @["", "foo", "bar"]
doAssert rsplit("foothebar", sep = "the") == @["foo", "bar"]
# Empty string:
doAssert "".rsplit() == @[""]
doAssert "".rsplit(" ") == @[""]
doAssert "".rsplit({' '}) == @[""]
# Empty separators:
let s = " this is an example "
doAssert "".rsplit({}) == @[""]
doAssert "".rsplit("") == @[""]
doAssert s.rsplit({}) == @[s]
doAssert s.rsplit("") == @[s]
block: # splitWhitespace
let s = " this is an example "