From fd25505b8e4e390bfd1d097b7f0747322a4cef8c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ciar=C3=A1n=20O=E2=80=99Mara?= <Ciaran.OMara@utas.edu.au>
Date: Thu, 16 Jun 2022 01:25:10 +1000
Subject: [PATCH 1/4] Documentation for `findall` method

---
 src/BioSequences.jl | 46 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)

diff --git a/src/BioSequences.jl b/src/BioSequences.jl
index aef223a9..77e369d9 100644
--- a/src/BioSequences.jl
+++ b/src/BioSequences.jl
@@ -252,6 +252,52 @@ Base.eltype(::Type{<:Search{Q}}) where {Q<:HasRangeEltype} = UnitRange{Int}
 Base.eltype(::Type{<:Search}) = Int
 Base.IteratorSize(::Type{<:Search}) = Base.SizeUnknown()
 
+"""
+    findall(pattern, sequence::BioSequence[,rng::UnitRange{Int}]; overlap::Bool=true)::Vector
+
+Find all occurrences of `pattern` in `sequence`.
+
+The return value is a vector of ranges of indices where the matching sequences were found.
+If there are no matching sequences, the return value is an empty vector.
+
+The search is restricted to the specified range when `rng` is set.
+
+With the keyword argument `overlap` set as `true`, the start index for the next search gets set to the start of the current match plus one; if set to `false`, the start index for the next search gets set to the end of the current match plus one.
+The default value for the keyword argument `overlap` is `true`.
+
+The `pattern` can be a `Biosymbol` or a predicate.
+
+See also [`ExactSearchQuery`](@ref), [`ApproximateSearchQuery`](@ref), [`PWMSearchQuery`](@ref).
+
+# Examples
+```jldoctest
+julia> seq = dna"ACACACAC"
+8nt DNA Sequence:
+ACACACAC
+
+julia> findall(DNA_A, seq)
+4-element Vector{Int64}:
+ 1
+ 3
+ 5
+ 7
+
+julia> findall(ExactSearchQuery(dna"ACAC"), seq)
+3-element Vector{UnitRange{Int64}}:
+ 1:4
+ 3:6
+ 5:8
+
+julia> findall(ExactSearchQuery(dna"ACAC"), seq; overlap=false)
+2-element Vector{UnitRange{Int64}}:
+ 1:4
+ 5:8
+
+julia> findall(ExactSearchQuery(dna"ACAC"), seq, 2:7; overlap=false)
+1-element Vector{UnitRange{Int64}}:
+ 3:6
+```
+"""
 function Base.findall(pat, seq::BioSequence; overlap::Bool = DEFAULT_OVERLAP)
     return collect(search(pat, seq; overlap))
 end

From 5292edb4c7dd34da596cd7a391bb84f0452cfaa1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ciar=C3=A1n=20O=27Mara?= <Ciaran.OMara@utas.edu.au>
Date: Mon, 1 Aug 2022 21:17:55 +1000
Subject: [PATCH 2/4] Add `findall` to examples

---
 docs/src/sequence_search.md    | 5 +++++
 src/search/ExactSearchQuery.jl | 5 +++++
 2 files changed, 10 insertions(+)

diff --git a/docs/src/sequence_search.md b/docs/src/sequence_search.md
index ca49b70e..ac99ea99 100644
--- a/docs/src/sequence_search.md
+++ b/docs/src/sequence_search.md
@@ -250,6 +250,11 @@ julia> qa = PWMSearchQuery(motifs, 1.0);
 julia> findfirst(qa, subject)
 3
 
+julia> findall(qa, subject)
+3-element Vector{Int64}:
+ 3
+ 5
+ 9
 ```
 
 [Wasserman2004]: https://doi.org/10.1038/nrg1315
diff --git a/src/search/ExactSearchQuery.jl b/src/search/ExactSearchQuery.jl
index 948960fd..fbf08ad7 100644
--- a/src/search/ExactSearchQuery.jl
+++ b/src/search/ExactSearchQuery.jl
@@ -28,6 +28,11 @@ julia> findnext(query, seq, 6)
 julia> findprev(query, seq, 7)
 3:5
 
+julia> findall(query, seq)
+2-element Vector{UnitRange{Int64}}:
+ 3:5
+ 8:10
+
 julia> occursin(query, seq)
 true
 

From 871bba9b669e97bb3ed3ad22da88570ca1daf154 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ciar=C3=A1n=20O=27Mara?= <Ciaran.OMara@utas.edu.au>
Date: Mon, 1 Aug 2022 20:56:56 +1000
Subject: [PATCH 3/4] Suppress output

---
 src/BioSequences.jl | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/src/BioSequences.jl b/src/BioSequences.jl
index 77e369d9..f4d7ac4d 100644
--- a/src/BioSequences.jl
+++ b/src/BioSequences.jl
@@ -271,9 +271,7 @@ See also [`ExactSearchQuery`](@ref), [`ApproximateSearchQuery`](@ref), [`PWMSear
 
 # Examples
 ```jldoctest
-julia> seq = dna"ACACACAC"
-8nt DNA Sequence:
-ACACACAC
+julia> seq = dna"ACACACAC";
 
 julia> findall(DNA_A, seq)
 4-element Vector{Int64}:

From b7223a1c6313c9ee4a717f68d4c11c49b22726ef Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ciar=C3=A1n=20O=27Mara?= <Ciaran.OMara@utas.edu.au>
Date: Mon, 1 Aug 2022 20:58:15 +1000
Subject: [PATCH 4/4] Tweak nomenclature

---
 src/BioSequences.jl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/BioSequences.jl b/src/BioSequences.jl
index f4d7ac4d..4676840e 100644
--- a/src/BioSequences.jl
+++ b/src/BioSequences.jl
@@ -265,7 +265,7 @@ The search is restricted to the specified range when `rng` is set.
 With the keyword argument `overlap` set as `true`, the start index for the next search gets set to the start of the current match plus one; if set to `false`, the start index for the next search gets set to the end of the current match plus one.
 The default value for the keyword argument `overlap` is `true`.
 
-The `pattern` can be a `Biosymbol` or a predicate.
+The `pattern` can be a `Biosymbol` or a search query.
 
 See also [`ExactSearchQuery`](@ref), [`ApproximateSearchQuery`](@ref), [`PWMSearchQuery`](@ref).