Skip to content

Enhance -split operator with RightToLeft splitting (#2) #5270

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
IISResetMe wants to merge 3 commits into from
Closed

Enhance -split operator with RightToLeft splitting (#2) #5270

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
81873b3
Add tests for -Split operator Right-to-Left mode
IISResetMe Oct 28, 2017
5e5e3e2
Add Right-To-Left splitting to SplitWithPattern()
IISResetMe Oct 28, 2017
cf47352
Add Right-To-Left splitting to SplitWithPredicate()
IISResetMe Oct 28, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
103 changes: 79 additions & 24 deletions src/System.Management.Automation/engine/lang/parserutils.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -632,22 +632,54 @@ private static object SplitOperatorImpl(ExecutionContext context, IScriptExtent

private static object SplitWithPredicate(ExecutionContext context, IScriptExtent errorPosition, IEnumerable<string> content, ScriptBlock predicate, int limit)
{
List<string> results = new List<string>();
foreach (string item in content)
// If user supplied a negative Max-substrings argument,
// we employ Right-to-Left splitting instead
bool rightToLeft = limit < 0;

limit = System.Math.Abs(limit);

if (limit == 1)
{
List<string> split = new List<String>();
// Don't bother with looking for any delimiters,
// just return the original string(s).
return new List<String>(content);
}

List<string> results = new List<string>();

if (limit == 1)
List<char> buf = new List<char>();

foreach (string item in content)
{
List<string> split;
if(limit == 0)
{
// Don't bother with looking for any delimiters,
// just return the original string.
results.Add(item);
continue;
// No limit specified
// set capacity to item.Length + 1
// this covers worst common case (predicate == {$true})
split = new List<String>(item.Length + 1);
}
else
{
// Limit was specified by the user
// instantiate list with maximum needed capacity
split = new List<String>(limit);
}

// Clear char buffer
buf.Clear();

StringBuilder buf = new StringBuilder();
for (int strIndex = 0; strIndex < item.Length; strIndex++)
int strIndex = 0;
for (int cursor = 0; cursor < item.Length; cursor++)
{
if(rightToLeft)
{
strIndex = item.Length - 1 - cursor;
}
else
{
strIndex = cursor;
}
object isDelimChar = predicate.DoInvokeReturnAsIs(
useLocalScope: true,
errorHandlingBehavior: ScriptBlock.ErrorHandlingBehavior.WriteToExternalErrorPipe,
Expand All @@ -657,8 +689,12 @@ private static object SplitWithPredicate(ExecutionContext context, IScriptExtent
args: new object[] { item, strIndex });
if (LanguagePrimitives.IsTrue(isDelimChar))
{
split.Add(buf.ToString());
buf = new StringBuilder();
if(rightToLeft)
{
buf.Reverse();
}
split.Add(string.Concat(buf));
buf.Clear();

if (limit > 0 && split.Count >= (limit - 1))
{
Expand All @@ -667,9 +703,16 @@ private static object SplitWithPredicate(ExecutionContext context, IScriptExtent
// and add it as the last item, otherwise
// add an empty string if there was
// a delimiter at the end.
if ((strIndex + 1) < item.Length)
if ((cursor + 1) < item.Length)
{
split.Add(item.Substring(strIndex + 1));
if(rightToLeft)
{
split.Add(item.Substring(0, strIndex));
}
else
{
split.Add(item.Substring(strIndex + 1));
}
}
else
{
Expand All @@ -681,24 +724,34 @@ private static object SplitWithPredicate(ExecutionContext context, IScriptExtent
// If this delimiter is at the end of the string,
// add an empty string to denote the item "after"
// it.
if (strIndex == (item.Length - 1))
if (cursor == (item.Length - 1))
{
split.Add("");
}
}
else
{
buf.Append(item[strIndex]);
buf.Add(item[strIndex]);
}
}

// Add any remainder, if we're under the limit.
if (buf.Length > 0 &&
(limit <= 0 || split.Count < limit))
if (buf.Count > 0 &&
(limit == 0 || split.Count < limit))
{
split.Add(buf.ToString());
if(rightToLeft)
{
buf.Reverse();
}
split.Add(string.Concat(buf));
}

if(rightToLeft)
{
// We want to preserve the order from the
// original string (always output chunks left-to-right)
split.Reverse();
}
ExtendList(results, split);
}

Expand Down Expand Up @@ -728,20 +781,22 @@ private static object SplitWithPattern(ExecutionContext context, IScriptExtent e
separatorPattern = Regex.Escape(separatorPattern);
}

RegexOptions regexOptions = parseRegexOptions(options);

if (limit < 0)
{
// Regex only allows 0 to signify "no limit", whereas
// we allow any integer <= 0.
limit = 0;
// If user-suppplied limit is negative we
// interpret it as "split Right-to-Left".
regexOptions |= RegexOptions.RightToLeft;
limit = System.Math.Abs(limit);
}

RegexOptions regexOptions = parseRegexOptions(options);
Regex regex = NewRegex(separatorPattern, regexOptions);

List<string> results = new List<string>();
foreach (string item in content)
{
string[] split = regex.Split(item, limit, 0);
string[] split = regex.Split(item, limit);
ExtendList(results, split);
}

Expand Down
46 changes: 41 additions & 5 deletions test/powershell/Language/Operators/SplitOperator.Tests.ps1
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,13 +30,18 @@ Describe "Split Operator" -Tags CI {
$res[1] | Should Be "b"
$res[2] | Should Be "c"
$res[3] | Should Be "d"
}

It "Binary split operator works with negative substring limit" {
$res = "a b c d" -split " ", -3
$res.count | Should Be 3
$res[0] | Should Be "a b"
$res[1] | Should Be "c"
$res[2] | Should Be "d"

$res = "a b c d" -split " ", -1
$res.count | Should Be 4
$res[0] | Should Be "a"
$res[1] | Should Be "b"
$res[2] | Should Be "c"
$res[3] | Should Be "d"
$res.count | Should Be 1
$res[0] | Should Be "a b c d"
}

It "Binary split operator can works with freeform delimiter" {
Expand Down Expand Up @@ -103,6 +108,37 @@ Describe "Split Operator" -Tags CI {
$res[2] | Should Be "::d"
}

It "Binary split operator works with script block and substring limit" {
$res = "a::b::c::d" -split {$_ -eq "b" -or $_ -eq "C"}, 2
$res.count | Should Be 2
$res[0] | Should Be "a::"
$res[1] | Should Be "::c::d"
}

It "Binary split operator works with script block and substring limit on arrays" {
$res = "a::b::c::d","e::f::g::h" -split {$_ -eq "b" -or $_ -eq "c" -or $_ -eq "f" -or $_ -eq "g"}, 2
$res.count | Should Be 4
$res[0] | Should Be "a::"
$res[1] | Should Be "::c::d"
$res[2] | Should Be "e::"
$res[3] | Should Be "::g::h"
}

It "Binary split operator works with script block and negative substring limit" {
$res = "a::b::c::d" -split {$_ -eq "b" -or $_ -eq "C"}, -2
$res.count | Should Be 2
$res[0] | Should Be "a::b::"
$res[1] | Should Be "::d"
}

It "Binary split operator works with script block and negative substring limit on arrays" {
$res = "a::b::c::d","e::f::g::h" -split {$_ -eq "b" -or $_ -eq "c" -or $_ -eq "f" -or $_ -eq "g"}, -2
$res.count | Should Be 4
$res[0] | Should Be "a::b::"
$res[1] | Should Be "::d"
$res[2] | Should Be "e::f::"
$res[3] | Should Be "::h"
}
}

Context "Binary split operator options" {
Expand Down