From c501ef7a396672f601cceb290aab2af1c34a174a Mon Sep 17 00:00:00 2001
From: "Steven G. Johnson" <stevenj@mit.edu>
Date: Thu, 25 Apr 2019 15:46:41 -0400
Subject: [PATCH 1/2] add findall(pattern, string)

---
 NEWS.md                |  1 +
 base/regex.jl          | 24 ++++++++++++++++++++++++
 test/regex.jl          |  6 ++++++
 test/strings/search.jl | 10 ++++++++++
 4 files changed, 41 insertions(+)

diff --git a/NEWS.md b/NEWS.md
index e5d3619986d08..807eaf14edd52 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -20,6 +20,7 @@ Build system changes
 New library functions
 ---------------------
 
+* New `findall(pattern, string)` method where `pattern` is a string or regex.
 
 Standard library changes
 ------------------------
diff --git a/base/regex.jl b/base/regex.jl
index 3bee6dbd64947..3043de4252a6a 100644
--- a/base/regex.jl
+++ b/base/regex.jl
@@ -310,6 +310,30 @@ findnext(r::Regex, s::AbstractString, idx::Integer) = throw(ArgumentError(
 ))
 findfirst(r::Regex, s::AbstractString) = findnext(r,s,firstindex(s))
 
+
+"""
+    findall(pattern::Union{AbstractString,Regex}, string::AbstractString; overlap::Bool=false)
+
+Return a `Vector{UnitRange{Int}}` of all the matches for `pattern` in `string`.
+Each element of the returned vector is a range of indices where the
+matching sequence is found, like the return value of [`findnext`](@ref).
+
+If `overlap=true`, the matching sequences are allowed to overlap indices in the
+original string, otherwise they must be from distinct character ranges.
+"""
+function findall(t::Union{AbstractString,Regex}, s::AbstractString; overlap::Bool=false)
+    found = UnitRange{Int}[]
+    i, e = firstindex(s), lastindex(s)
+    while true
+        r = findnext(t, s, i)
+        isnothing(r) && return found
+        push!(found, r)
+        j = overlap || isempty(r) ? first(r) : last(r)
+        j > e && return found
+        @inbounds i = nextind(s, j)
+    end
+end
+
 """
     SubstitutionString(substr)
 
diff --git a/test/regex.jl b/test/regex.jl
index cb3fa965f8a50..cb9bd6ff5eab8 100644
--- a/test/regex.jl
+++ b/test/regex.jl
@@ -46,6 +46,12 @@
     @test_throws ArgumentError match(r"test", GenericString("this is a test"))
     @test_throws ArgumentError findfirst(r"test", GenericString("this is a test"))
 
+    # findall:
+    @test findall(r"\w+", "foo bar") == [1:3, 5:7]
+    @test findall(r"\w+", "foo bar", overlap=true) == [1:3, 2:3, 3:3, 5:7, 6:7, 7:7]
+    @test findall(r"\w*", "foo bar") == [1:3, 4:3, 5:7, 8:7]
+    @test findall(r"\b", "foo bar") == [1:0, 4:3, 5:4, 8:7]
+
     # Named subpatterns
     let m = match(r"(?<a>.)(.)(?<b>.)", "xyz")
         @test (m[:a], m[2], m["b"]) == ("x", "y", "z")
diff --git a/test/strings/search.jl b/test/strings/search.jl
index e9b6a678baa3a..61aaea463eae8 100644
--- a/test/strings/search.jl
+++ b/test/strings/search.jl
@@ -333,3 +333,13 @@ end
 s_18109 = "fooα🐨βcd3"
 @test findlast(isequal('o'), s_18109) == 3
 @test findfirst(isequal('d'), s_18109) == 13
+
+# findall (issue #31788)
+@testset "findall" begin
+    @test findall("fooo", "foo") == UnitRange{Int}[]
+    @test findall("ing", "Spinning laughing dancing") == [6:8, 15:17, 23:25]
+    @test findall("", "foo") == [1:0, 2:1, 3:2, 4:3]
+    @test findall("αβ", "blαh blαβ blαββy") == findall("αβ", "blαh blαβ blαββy", overlap=true) == [9:11, 16:18]
+    @test findall("aa", "aaaaaa") == [1:2, 3:4, 5:6]
+    @test findall("aa", "aaaaaa", overlap=true) == [1:2, 2:3, 3:4, 4:5, 5:6]
+end

From 25a56f6a925d0e47c07e3d8f931ef79e893b9ea5 Mon Sep 17 00:00:00 2001
From: "Steven G. Johnson" <stevenj@mit.edu>
Date: Thu, 25 Apr 2019 15:47:58 -0400
Subject: [PATCH 2/2] add PR # to NEWS

---
 NEWS.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/NEWS.md b/NEWS.md
index 807eaf14edd52..84adca46a158f 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -20,7 +20,7 @@ Build system changes
 New library functions
 ---------------------
 
-* New `findall(pattern, string)` method where `pattern` is a string or regex.
+* New `findall(pattern, string)` method where `pattern` is a string or regex ([#31834]).
 
 Standard library changes
 ------------------------