From fa9b00ee64a7ab2c775169550558ab7e35ac30e4 Mon Sep 17 00:00:00 2001
From: John Kerl <kerl.john.r@gmail.com>
Date: Mon, 27 Mar 2023 00:37:33 -0400
Subject: [PATCH] `index` DSL function [WIP] (#1247)

---
 docs/src/manpage.md                           | 22 +++++++++-----
 docs/src/manpage.txt                          | 22 +++++++++-----
 docs/src/reference-dsl-builtin-functions.md   | 13 +++++++-
 internal/pkg/bifs/strings.go                  | 24 +++++++++++++++
 .../pkg/dsl/cst/builtin_function_manager.go   | 12 ++++++++
 man/manpage.txt                               | 22 +++++++++-----
 man/mlr.1                                     | 30 ++++++++++++++-----
 test/cases/dsl-index/0001/cmd                 |  1 +
 test/cases/dsl-index/0001/experr              |  0
 test/cases/dsl-index/0001/expout              | 13 ++++++++
 test/cases/dsl-index/0001/input               | 13 ++++++++
 test/cases/dsl-index/0001/mlr                 |  1 +
 12 files changed, 143 insertions(+), 30 deletions(-)
 create mode 100644 test/cases/dsl-index/0001/cmd
 create mode 100644 test/cases/dsl-index/0001/experr
 create mode 100644 test/cases/dsl-index/0001/expout
 create mode 100644 test/cases/dsl-index/0001/input
 create mode 100644 test/cases/dsl-index/0001/mlr

diff --git a/docs/src/manpage.md b/docs/src/manpage.md
index 120b4b550..e1a0ac63a 100644
--- a/docs/src/manpage.md
+++ b/docs/src/manpage.md
@@ -210,12 +210,12 @@ MILLER(1)                                                            MILLER(1)
        capitalize cbrt ceil clean_whitespace collapse_whitespace concat cos cosh
        depth dhms2fsec dhms2sec erf erfc every exec exp expm1 flatten float floor
        fmtifnum fmtnum fold format fsec2dhms fsec2hms get_keys get_values
-       gmt2localtime gmt2sec gssub gsub haskey hexfmt hms2fsec hms2sec hostname int
-       invqnorm is_absent is_array is_bool is_boolean is_empty is_empty_map is_error
-       is_float is_int is_map is_nan is_nonempty_map is_not_array is_not_empty
-       is_not_map is_not_null is_null is_numeric is_present is_string joink joinkv
-       joinv json_parse json_stringify latin1_to_utf8 leafcount leftpad length
-       localtime2gmt localtime2sec log log10 log1p logifit lstrip madd mapdiff
+       gmt2localtime gmt2sec gssub gsub haskey hexfmt hms2fsec hms2sec hostname index
+       int invqnorm is_absent is_array is_bool is_boolean is_empty is_empty_map
+       is_error is_float is_int is_map is_nan is_nonempty_map is_not_array
+       is_not_empty is_not_map is_not_null is_null is_numeric is_present is_string
+       joink joinkv joinv json_parse json_stringify latin1_to_utf8 leafcount leftpad
+       length localtime2gmt localtime2sec log log10 log1p logifit lstrip madd mapdiff
        mapexcept mapselect mapsum max md5 mexp min mmul msub os pow qnorm reduce
        regextract regextract_or_else rightpad round roundm rstrip sec2dhms sec2gmt
        sec2gmtdate sec2hms sec2localdate sec2localtime select sgn sha1 sha256 sha512
@@ -2387,6 +2387,14 @@ MILLER(1)                                                            MILLER(1)
    1mhostname0m
         (class=system #args=0) Returns the hostname as a string.
 
+   1mindex0m
+        (class=string #args=2) Returns the index (1-based) of the second argument within the first. Returns -1 if the second argument isn't a substring of the first. Stringifies non-string inputs. Uses UTF-8 encoding to count characters, not bytes.
+       Examples:
+       index("abcde", "e") gives 5
+       index("abcde", "x") gives 01
+       index(12345, 34) gives 3
+       index("fort", "t") gives 5
+
    1mint0m
         (class=conversion #args=1,2) Convert int/float/bool/string to int. If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix. If the second argument is provided and the first argument is a string, the second argument is used as the base. If the second argument is provided and the first argument is not a string, the second argument is ignored.
        Examples:
@@ -3346,5 +3354,5 @@ MILLER(1)                                                            MILLER(1)
 
 
 
-                                  2023-03-24                         MILLER(1)
+                                  2023-03-27                         MILLER(1)
 </pre>
diff --git a/docs/src/manpage.txt b/docs/src/manpage.txt
index c78f1b415..17b05732b 100644
--- a/docs/src/manpage.txt
+++ b/docs/src/manpage.txt
@@ -189,12 +189,12 @@ MILLER(1)                                                            MILLER(1)
        capitalize cbrt ceil clean_whitespace collapse_whitespace concat cos cosh
        depth dhms2fsec dhms2sec erf erfc every exec exp expm1 flatten float floor
        fmtifnum fmtnum fold format fsec2dhms fsec2hms get_keys get_values
-       gmt2localtime gmt2sec gssub gsub haskey hexfmt hms2fsec hms2sec hostname int
-       invqnorm is_absent is_array is_bool is_boolean is_empty is_empty_map is_error
-       is_float is_int is_map is_nan is_nonempty_map is_not_array is_not_empty
-       is_not_map is_not_null is_null is_numeric is_present is_string joink joinkv
-       joinv json_parse json_stringify latin1_to_utf8 leafcount leftpad length
-       localtime2gmt localtime2sec log log10 log1p logifit lstrip madd mapdiff
+       gmt2localtime gmt2sec gssub gsub haskey hexfmt hms2fsec hms2sec hostname index
+       int invqnorm is_absent is_array is_bool is_boolean is_empty is_empty_map
+       is_error is_float is_int is_map is_nan is_nonempty_map is_not_array
+       is_not_empty is_not_map is_not_null is_null is_numeric is_present is_string
+       joink joinkv joinv json_parse json_stringify latin1_to_utf8 leafcount leftpad
+       length localtime2gmt localtime2sec log log10 log1p logifit lstrip madd mapdiff
        mapexcept mapselect mapsum max md5 mexp min mmul msub os pow qnorm reduce
        regextract regextract_or_else rightpad round roundm rstrip sec2dhms sec2gmt
        sec2gmtdate sec2hms sec2localdate sec2localtime select sgn sha1 sha256 sha512
@@ -2366,6 +2366,14 @@ MILLER(1)                                                            MILLER(1)
    1mhostname0m
         (class=system #args=0) Returns the hostname as a string.
 
+   1mindex0m
+        (class=string #args=2) Returns the index (1-based) of the second argument within the first. Returns -1 if the second argument isn't a substring of the first. Stringifies non-string inputs. Uses UTF-8 encoding to count characters, not bytes.
+       Examples:
+       index("abcde", "e") gives 5
+       index("abcde", "x") gives 01
+       index(12345, 34) gives 3
+       index("fort", "t") gives 5
+
    1mint0m
         (class=conversion #args=1,2) Convert int/float/bool/string to int. If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix. If the second argument is provided and the first argument is a string, the second argument is used as the base. If the second argument is provided and the first argument is not a string, the second argument is ignored.
        Examples:
@@ -3325,4 +3333,4 @@ MILLER(1)                                                            MILLER(1)
 
 
 
-                                  2023-03-24                         MILLER(1)
+                                  2023-03-27                         MILLER(1)
diff --git a/docs/src/reference-dsl-builtin-functions.md b/docs/src/reference-dsl-builtin-functions.md
index 90bf09e00..ae8f5d517 100644
--- a/docs/src/reference-dsl-builtin-functions.md
+++ b/docs/src/reference-dsl-builtin-functions.md
@@ -74,7 +74,7 @@ is 2. Unary operators such as `!` and `~` show argument-count of 1; the ternary
 * [**Hashing functions**](#hashing-functions):  [md5](#md5),  [sha1](#sha1),  [sha256](#sha256),  [sha512](#sha512).
 * [**Higher-order-functions functions**](#higher-order-functions-functions):  [any](#any),  [apply](#apply),  [every](#every),  [fold](#fold),  [reduce](#reduce),  [select](#select),  [sort](#sort).
 * [**Math functions**](#math-functions):  [abs](#abs),  [acos](#acos),  [acosh](#acosh),  [asin](#asin),  [asinh](#asinh),  [atan](#atan),  [atan2](#atan2),  [atanh](#atanh),  [cbrt](#cbrt),  [ceil](#ceil),  [cos](#cos),  [cosh](#cosh),  [erf](#erf),  [erfc](#erfc),  [exp](#exp),  [expm1](#expm1),  [floor](#floor),  [invqnorm](#invqnorm),  [log](#log),  [log10](#log10),  [log1p](#log1p),  [logifit](#logifit),  [max](#max),  [min](#min),  [qnorm](#qnorm),  [round](#round),  [roundm](#roundm),  [sgn](#sgn),  [sin](#sin),  [sinh](#sinh),  [sqrt](#sqrt),  [tan](#tan),  [tanh](#tanh),  [urand](#urand),  [urand32](#urand32),  [urandelement](#urandelement),  [urandint](#urandint),  [urandrange](#urandrange).
-* [**String functions**](#string-functions):  [capitalize](#capitalize),  [clean_whitespace](#clean_whitespace),  [collapse_whitespace](#collapse_whitespace),  [format](#format),  [gssub](#gssub),  [gsub](#gsub),  [latin1_to_utf8](#latin1_to_utf8),  [leftpad](#leftpad),  [lstrip](#lstrip),  [regextract](#regextract),  [regextract_or_else](#regextract_or_else),  [rightpad](#rightpad),  [rstrip](#rstrip),  [ssub](#ssub),  [strip](#strip),  [strlen](#strlen),  [sub](#sub),  [substr](#substr),  [substr0](#substr0),  [substr1](#substr1),  [tolower](#tolower),  [toupper](#toupper),  [truncate](#truncate),  [unformat](#unformat),  [unformatx](#unformatx),  [utf8_to_latin1](#utf8_to_latin1),  [\.](#dot).
+* [**String functions**](#string-functions):  [capitalize](#capitalize),  [clean_whitespace](#clean_whitespace),  [collapse_whitespace](#collapse_whitespace),  [format](#format),  [gssub](#gssub),  [gsub](#gsub),  [index](#index),  [latin1_to_utf8](#latin1_to_utf8),  [leftpad](#leftpad),  [lstrip](#lstrip),  [regextract](#regextract),  [regextract_or_else](#regextract_or_else),  [rightpad](#rightpad),  [rstrip](#rstrip),  [ssub](#ssub),  [strip](#strip),  [strlen](#strlen),  [sub](#sub),  [substr](#substr),  [substr0](#substr0),  [substr1](#substr1),  [tolower](#tolower),  [toupper](#toupper),  [truncate](#truncate),  [unformat](#unformat),  [unformatx](#unformatx),  [utf8_to_latin1](#utf8_to_latin1),  [\.](#dot).
 * [**System functions**](#system-functions):  [exec](#exec),  [hostname](#hostname),  [os](#os),  [system](#system),  [version](#version).
 * [**Time functions**](#time-functions):  [dhms2fsec](#dhms2fsec),  [dhms2sec](#dhms2sec),  [fsec2dhms](#fsec2dhms),  [fsec2hms](#fsec2hms),  [gmt2localtime](#gmt2localtime),  [gmt2sec](#gmt2sec),  [hms2fsec](#hms2fsec),  [hms2sec](#hms2sec),  [localtime2gmt](#localtime2gmt),  [localtime2sec](#localtime2sec),  [sec2dhms](#sec2dhms),  [sec2gmt](#sec2gmt),  [sec2gmtdate](#sec2gmtdate),  [sec2hms](#sec2hms),  [sec2localdate](#sec2localdate),  [sec2localtime](#sec2localtime),  [strftime](#strftime),  [strftime_local](#strftime_local),  [strptime](#strptime),  [strptime_local](#strptime_local),  [systime](#systime),  [systimeint](#systimeint),  [uptime](#uptime).
 * [**Typing functions**](#typing-functions):  [asserting_absent](#asserting_absent),  [asserting_array](#asserting_array),  [asserting_bool](#asserting_bool),  [asserting_boolean](#asserting_boolean),  [asserting_empty](#asserting_empty),  [asserting_empty_map](#asserting_empty_map),  [asserting_error](#asserting_error),  [asserting_float](#asserting_float),  [asserting_int](#asserting_int),  [asserting_map](#asserting_map),  [asserting_nonempty_map](#asserting_nonempty_map),  [asserting_not_array](#asserting_not_array),  [asserting_not_empty](#asserting_not_empty),  [asserting_not_map](#asserting_not_map),  [asserting_not_null](#asserting_not_null),  [asserting_null](#asserting_null),  [asserting_numeric](#asserting_numeric),  [asserting_present](#asserting_present),  [asserting_string](#asserting_string),  [is_absent](#is_absent),  [is_array](#is_array),  [is_bool](#is_bool),  [is_boolean](#is_boolean),  [is_empty](#is_empty),  [is_empty_map](#is_empty_map),  [is_error](#is_error),  [is_float](#is_float),  [is_int](#is_int),  [is_map](#is_map),  [is_nan](#is_nan),  [is_nonempty_map](#is_nonempty_map),  [is_not_array](#is_not_array),  [is_not_empty](#is_not_empty),  [is_not_map](#is_not_map),  [is_not_null](#is_not_null),  [is_null](#is_null),  [is_numeric](#is_numeric),  [is_present](#is_present),  [is_string](#is_string),  [typeof](#typeof).
@@ -1023,6 +1023,17 @@ gsub("prefix4529:suffix8567", "(....ix)([0-9]+)", "[\1 : \2]") gives "[prefix :
 </pre>
 
 
+### index
+<pre class="pre-non-highlight-non-pair">
+index  (class=string #args=2) Returns the index (1-based) of the second argument within the first. Returns -1 if the second argument isn't a substring of the first. Stringifies non-string inputs. Uses UTF-8 encoding to count characters, not bytes.
+Examples:
+index("abcde", "e") gives 5
+index("abcde", "x") gives 01
+index(12345, 34) gives 3
+index("forêt", "t") gives 5
+</pre>
+
+
 ### latin1_to_utf8
 <pre class="pre-non-highlight-non-pair">
 latin1_to_utf8  (class=string #args=1) Tries to convert Latin-1-encoded string to UTF-8-encoded string. If argument is array or map, recurses into it.
diff --git a/internal/pkg/bifs/strings.go b/internal/pkg/bifs/strings.go
index 46c86e28d..4cdcdce93 100644
--- a/internal/pkg/bifs/strings.go
+++ b/internal/pkg/bifs/strings.go
@@ -125,6 +125,30 @@ func BIF_substr_0_up(input1, input2, input3 *mlrval.Mlrval) *mlrval.Mlrval {
 	return mlrval.FromString(string(runes[lowerZindex : upperZindex+1]))
 }
 
+// ================================================================
+// index(string, substring) returns the index of substring within string (if found), or -1 if not
+// found.
+
+func BIF_index(input1, input2 *mlrval.Mlrval) *mlrval.Mlrval {
+	if input1.IsAbsent() {
+		return mlrval.ABSENT
+	}
+	if input1.IsError() {
+		return mlrval.ERROR
+	}
+	sinput1 := input1.String()
+	sinput2 := input2.String()
+
+	// Handle UTF-8 correctly, since Go's strings.Index counts bytes
+	iindex := strings.Index(sinput1, sinput2)
+	if iindex < 0 {
+		return mlrval.FromInt(int64(iindex))
+	}
+
+	// Go indices are 0-up; Miller indices are 1-up.
+	return mlrval.FromInt(lib.UTF8Strlen(sinput1[:iindex]) + 1)
+}
+
 // ================================================================
 func BIF_truncate(input1, input2 *mlrval.Mlrval) *mlrval.Mlrval {
 	if input1.IsErrorOrAbsent() {
diff --git a/internal/pkg/dsl/cst/builtin_function_manager.go b/internal/pkg/dsl/cst/builtin_function_manager.go
index 8a6ca2d67..6f616baa4 100644
--- a/internal/pkg/dsl/cst/builtin_function_manager.go
+++ b/internal/pkg/dsl/cst/builtin_function_manager.go
@@ -538,6 +538,18 @@ array and string indices, but, this is a backward-compatibility issue with Mille
 Arrays are new in Miller 6; the substr function is older.`,
 			ternaryFunc: bifs.BIF_substr_0_up,
 		},
+		{
+			name:       "index",
+			class:      FUNC_CLASS_STRING,
+			help:       `Returns the index (1-based) of the second argument within the first. Returns -1 if the second argument isn't a substring of the first. Stringifies non-string inputs. Uses UTF-8 encoding to count characters, not bytes.`,
+			binaryFunc: bifs.BIF_index,
+			examples: []string{
+				`index("abcde", "e") gives 5`,
+				`index("abcde", "x") gives 01`,
+				`index(12345, 34) gives 3`,
+				`index("forêt", "t") gives 5`,
+			},
+		},
 
 		{
 			name:      "tolower",
diff --git a/man/manpage.txt b/man/manpage.txt
index c78f1b415..17b05732b 100644
--- a/man/manpage.txt
+++ b/man/manpage.txt
@@ -189,12 +189,12 @@ MILLER(1)                                                            MILLER(1)
        capitalize cbrt ceil clean_whitespace collapse_whitespace concat cos cosh
        depth dhms2fsec dhms2sec erf erfc every exec exp expm1 flatten float floor
        fmtifnum fmtnum fold format fsec2dhms fsec2hms get_keys get_values
-       gmt2localtime gmt2sec gssub gsub haskey hexfmt hms2fsec hms2sec hostname int
-       invqnorm is_absent is_array is_bool is_boolean is_empty is_empty_map is_error
-       is_float is_int is_map is_nan is_nonempty_map is_not_array is_not_empty
-       is_not_map is_not_null is_null is_numeric is_present is_string joink joinkv
-       joinv json_parse json_stringify latin1_to_utf8 leafcount leftpad length
-       localtime2gmt localtime2sec log log10 log1p logifit lstrip madd mapdiff
+       gmt2localtime gmt2sec gssub gsub haskey hexfmt hms2fsec hms2sec hostname index
+       int invqnorm is_absent is_array is_bool is_boolean is_empty is_empty_map
+       is_error is_float is_int is_map is_nan is_nonempty_map is_not_array
+       is_not_empty is_not_map is_not_null is_null is_numeric is_present is_string
+       joink joinkv joinv json_parse json_stringify latin1_to_utf8 leafcount leftpad
+       length localtime2gmt localtime2sec log log10 log1p logifit lstrip madd mapdiff
        mapexcept mapselect mapsum max md5 mexp min mmul msub os pow qnorm reduce
        regextract regextract_or_else rightpad round roundm rstrip sec2dhms sec2gmt
        sec2gmtdate sec2hms sec2localdate sec2localtime select sgn sha1 sha256 sha512
@@ -2366,6 +2366,14 @@ MILLER(1)                                                            MILLER(1)
    1mhostname0m
         (class=system #args=0) Returns the hostname as a string.
 
+   1mindex0m
+        (class=string #args=2) Returns the index (1-based) of the second argument within the first. Returns -1 if the second argument isn't a substring of the first. Stringifies non-string inputs. Uses UTF-8 encoding to count characters, not bytes.
+       Examples:
+       index("abcde", "e") gives 5
+       index("abcde", "x") gives 01
+       index(12345, 34) gives 3
+       index("fort", "t") gives 5
+
    1mint0m
         (class=conversion #args=1,2) Convert int/float/bool/string to int. If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix. If the second argument is provided and the first argument is a string, the second argument is used as the base. If the second argument is provided and the first argument is not a string, the second argument is ignored.
        Examples:
@@ -3325,4 +3333,4 @@ MILLER(1)                                                            MILLER(1)
 
 
 
-                                  2023-03-24                         MILLER(1)
+                                  2023-03-27                         MILLER(1)
diff --git a/man/mlr.1 b/man/mlr.1
index a56102ec0..1c50f0b02 100644
--- a/man/mlr.1
+++ b/man/mlr.1
@@ -2,12 +2,12 @@
 .\"     Title: mlr
 .\"    Author: [see the "AUTHOR" section]
 .\" Generator: ./mkman.rb
-.\"      Date: 2023-03-24
+.\"      Date: 2023-03-27
 .\"    Manual: \ \&
 .\"    Source: \ \&
 .\"  Language: English
 .\"
-.TH "MILLER" "1" "2023-03-24" "\ \&" "\ \&"
+.TH "MILLER" "1" "2023-03-27" "\ \&" "\ \&"
 .\" -----------------------------------------------------------------
 .\" * Portability definitions
 .\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -236,12 +236,12 @@ asserting_present asserting_string atan atan2 atanh bitcount boolean
 capitalize cbrt ceil clean_whitespace collapse_whitespace concat cos cosh
 depth dhms2fsec dhms2sec erf erfc every exec exp expm1 flatten float floor
 fmtifnum fmtnum fold format fsec2dhms fsec2hms get_keys get_values
-gmt2localtime gmt2sec gssub gsub haskey hexfmt hms2fsec hms2sec hostname int
-invqnorm is_absent is_array is_bool is_boolean is_empty is_empty_map is_error
-is_float is_int is_map is_nan is_nonempty_map is_not_array is_not_empty
-is_not_map is_not_null is_null is_numeric is_present is_string joink joinkv
-joinv json_parse json_stringify latin1_to_utf8 leafcount leftpad length
-localtime2gmt localtime2sec log log10 log1p logifit lstrip madd mapdiff
+gmt2localtime gmt2sec gssub gsub haskey hexfmt hms2fsec hms2sec hostname index
+int invqnorm is_absent is_array is_bool is_boolean is_empty is_empty_map
+is_error is_float is_int is_map is_nan is_nonempty_map is_not_array
+is_not_empty is_not_map is_not_null is_null is_numeric is_present is_string
+joink joinkv joinv json_parse json_stringify latin1_to_utf8 leafcount leftpad
+length localtime2gmt localtime2sec log log10 log1p logifit lstrip madd mapdiff
 mapexcept mapselect mapsum max md5 mexp min mmul msub os pow qnorm reduce
 regextract regextract_or_else rightpad round roundm rstrip sec2dhms sec2gmt
 sec2gmtdate sec2hms sec2localdate sec2localtime select sgn sha1 sha256 sha512
@@ -3357,6 +3357,20 @@ gsub("prefix4529:suffix8567", "(....ix)([0-9]+)", "[\e1 : \e2]") gives "[prefix
 .fi
 .if n \{\
 .RE
+.SS "index"
+.if n \{\
+.RS 0
+.\}
+.nf
+ (class=string #args=2) Returns the index (1-based) of the second argument within the first. Returns -1 if the second argument isn't a substring of the first. Stringifies non-string inputs. Uses UTF-8 encoding to count characters, not bytes.
+Examples:
+index("abcde", "e") gives 5
+index("abcde", "x") gives 01
+index(12345, 34) gives 3
+index("forêt", "t") gives 5
+.fi
+.if n \{\
+.RE
 .SS "int"
 .if n \{\
 .RS 0
diff --git a/test/cases/dsl-index/0001/cmd b/test/cases/dsl-index/0001/cmd
new file mode 100644
index 000000000..557ef51a6
--- /dev/null
+++ b/test/cases/dsl-index/0001/cmd
@@ -0,0 +1 @@
+mlr --icsv --opprint put -f ${CASEDIR}/mlr ${CASEDIR}/input
diff --git a/test/cases/dsl-index/0001/experr b/test/cases/dsl-index/0001/experr
new file mode 100644
index 000000000..e69de29bb
diff --git a/test/cases/dsl-index/0001/expout b/test/cases/dsl-index/0001/expout
new file mode 100644
index 000000000..2b4fc0407
--- /dev/null
+++ b/test/cases/dsl-index/0001/expout
@@ -0,0 +1,13 @@
+a            b  c
+abcde        a  1
+abcde        b  2
+abcde        c  3
+abcde        d  4
+abcde        e  5
+abcde        x  -1
+forêt cachée fo 1
+forêt cachée êt 4
+forêt cachée ê  4
+forêt cachée e  12
+forêt cachée x  -1
+12345        34 3
diff --git a/test/cases/dsl-index/0001/input b/test/cases/dsl-index/0001/input
new file mode 100644
index 000000000..5ded386d3
--- /dev/null
+++ b/test/cases/dsl-index/0001/input
@@ -0,0 +1,13 @@
+a,b
+abcde,a
+abcde,b
+abcde,c
+abcde,d
+abcde,e
+abcde,x
+forêt cachée,fo
+forêt cachée,êt
+forêt cachée,ê
+forêt cachée,e
+forêt cachée,x
+12345,34
diff --git a/test/cases/dsl-index/0001/mlr b/test/cases/dsl-index/0001/mlr
new file mode 100644
index 000000000..a95391469
--- /dev/null
+++ b/test/cases/dsl-index/0001/mlr
@@ -0,0 +1 @@
+$c = index($a, $b)