Change attribute internal representation from Dict to OrderedDict

Author: joshday

Project.toml

@@ -5,6 +5,8 @@ version = "0.2.3"
 
 [deps]
 Mmap = "a63ad114-7e13-5084-954f-fe012c677804"
+OrderedCollections = "bac558e1-5e72-5ebc-8fee-abe8a469f55d"
 
 [compat]
 julia = "1.7"
+OrderedCollections = "1.4, 1.5"

README.md

@@ -17,7 +17,7 @@ This package offers fast data structures for reading and writing XML files with
 ```
 nodetype(node)      →   XML.NodeType (an enum type)
 tag(node)           →   String or Nothing
-attributes(node)    →   Dict{String, String} or Nothing
+attributes(node)    →   OrderedDict{String, String} or Nothing
 value(node)         →   String or Nothing
 children(node)      →   Vector{typeof(node)}
 is_simple(node)     →   Bool (whether node is simple .e.g. <tag>item</tag>)

src/XML.jl

@@ -1,6 +1,7 @@
 module XML
 
 using Mmap
+using OrderedCollections: OrderedDict
 
 export
     # Core Types:
@@ -71,7 +72,7 @@ A Lazy representation of an XML node.
 mutable struct LazyNode <: AbstractXMLNode
     raw::Raw
     tag::Union{Nothing, String}
-    attributes::Union{Nothing, Dict{String, String}}
+    attributes::Union{Nothing, OrderedDict{String, String}}
     value::Union{Nothing, String}
 end
 LazyNode(raw::Raw) = LazyNode(raw, nothing, nothing, nothing)
@@ -128,14 +129,14 @@ A representation of an XML DOM node.  For simpler construction, use `(::NodeType
 struct Node <: AbstractXMLNode
     nodetype::NodeType
     tag::Union{Nothing, String}
-    attributes::Union{Nothing, Dict{String, String}}
+    attributes::Union{Nothing, OrderedDict{String, String}}
     value::Union{Nothing, String}
     children::Union{Nothing, Vector{Node}}
 
     function Node(nodetype::NodeType, tag=nothing, attributes=nothing, value=nothing, children=nothing)
         new(nodetype,
             isnothing(tag) ? nothing : string(tag),
-            isnothing(attributes) ? nothing : Dict(string(k) => string(v) for (k, v) in pairs(attributes)),
+            isnothing(attributes) ? nothing : OrderedDict(string(k) => string(v) for (k, v) in pairs(attributes)),
             isnothing(value) ? nothing : string(value),
             isnothing(children) ? nothing :
                 children isa Node ? [children] :

src/raw.jl

@@ -52,7 +52,7 @@ Useful functions:
     - next(o::Raw) --> Raw of the next chunk (or `nothing`).
     - prev(o::Raw) --> Raw of the previous chunk (or `nothing`).
     - tag(o::Raw) --> String of the tag name (or `nothing`).
-    - attributes(o::Raw) --> Dict{String, String} of the attributes (or `nothing`).
+    - attributes(o::Raw) --> OrderedDict{String, String} of the attributes (or `nothing`).
     - value(o::Raw) --> String of the value (or `nothing`).
     - children(o::Raw) --> Vector{Raw} of the children (or `nothing`).
     - parent(o::Raw) --> Raw of the parent (or `nothing`)
@@ -118,7 +118,7 @@ end
 function get_attributes(data, i, j)
     i = name_start(data, i)
     i > j && return nothing
-    out = Dict{String, String}()
+    out = OrderedDict{String, String}()
     while !isnothing(i) && i < j
         key, i = get_name(data, i)
         # get quotechar the value is wrapped in (either ' or ")
@@ -152,7 +152,7 @@ function tag(o::Raw)
 end
 
 """
-    attributes(node) --> Dict{String, String} or Nothing
+    attributes(node) --> OrderedDict{String, String} or Nothing
 
 Return the attributes of `Element`, `Declaration`, or `ProcessingInstruction` nodes.
 """

test/runtests.jl

@@ -1,5 +1,5 @@
 using XML
-using XML: Document, Element, Declaration, Comment, CData, DTD, ProcessingInstruction, Text, escape, unescape
+using XML: Document, Element, Declaration, Comment, CData, DTD, ProcessingInstruction, Text, escape, unescape, OrderedDict
 using Downloads: download
 using Test
 import AbstractTrees
@@ -200,7 +200,7 @@ end
 
 #-----------------------------------------------------------------------------# Issues
 @testset "Issues" begin
-    #12: DTD content was cut short
+    # https://github.com/JuliaComputing/XML.jl/issues/12: DTD content was cut short
     s = """
     <!DOCTYPE note [
     <!ENTITY nbsp "&#xA0;">
@@ -210,9 +210,10 @@ end
     """
 
     doc = parse(Node, s)
-    @test value(only(doc)) == """note [
-        <!ENTITY nbsp "&#xA0;">
-        <!ENTITY writer "Writer: Donald Duck.">
-        <!ENTITY copyright "Copyright: W3Schools.">
-        ]"""
+    @test value(only(doc)) == s[11:end-2]  # note [...]
+
+    # https://github.com/JuliaComputing/XML.jl/issues/14 (Sorted Attributes)
+    kw = NamedTuple(OrderedDict(Symbol(k) => Int(k) for k in 'a':'z'))
+    xyz  = XML.Element("point"; kw...)
+    @test collect(keys(attributes(xyz))) == string.(collect('a':'z'))
 end