Implement char_range() to respect string indices (#486)

Also widen `byte_range()` signatures to accept other integer types

Author: c42f

docs/src/api.md

@@ -52,6 +52,7 @@ JuliaSyntax.last_byte
 JuliaSyntax.filename
 JuliaSyntax.source_line
 JuliaSyntax.source_location
+JuliaSyntax.char_range
 JuliaSyntax.sourcetext
 JuliaSyntax.highlight
 ```

src/JuliaSyntax.jl

@@ -9,7 +9,7 @@ export parsestmt, parseall, parseatom
 # Tokenization
 export tokenize, Token, untokenize
 # Source file handling. See also
-#   highlight() sourcetext() source_line() source_location()
+#   highlight() sourcetext() source_line() source_location() char_range()
 export SourceFile
 # Expression heads/kinds. See also
 #   flags() and related predicates.

src/source_files.jl

@@ -15,11 +15,24 @@ end
 """
     byte_range(x)
 
-Return the range of bytes which `x` covers in the source text.
+Return the range of bytes which `x` covers in the source text. See also
+[`char_range`](@ref).
 """
 function byte_range
 end
 
+"""
+    char_range(x)
+
+Compute the range in *character indices* over the source text for syntax object
+`x`. If you want to index the source string you need this, rather than
+[`byte_range`](@ref).
+"""
+function char_range(x)
+    br = byte_range(x)
+    first(br):thisind(sourcefile(x), last(br))
+end
+
 """
     first_byte(x)
 
@@ -232,11 +245,11 @@ function Base.view(source::SourceFile, rng::AbstractUnitRange)
     SubString(source.code, i, j)
 end
 
-function Base.getindex(source::SourceFile, i::Int)
+function Base.getindex(source::SourceFile, i::Integer)
     source.code[i - source.byte_offset]
 end
 
-function Base.thisind(source::SourceFile, i::Int)
+function Base.thisind(source::SourceFile, i::Integer)
     thisind(source.code, i - source.byte_offset) + source.byte_offset
 end
 

test/syntax_tree.jl

@@ -55,6 +55,15 @@
     @test t.position == 13
     @test child(t,1).position == 19
     @test child(t,1).val == :b
+
+    # Unicode character ranges
+    src = "ab + αβ"
+    t = parsestmt(SyntaxNode, src)
+    @test char_range(t[1]) == 1:2
+    @test char_range(t[2]) == 4:4
+    @test char_range(t[3]) == 6:8
+    # conversely, β takes two bytes so char_range(t[3]) != byte_range(t[3])
+    @test byte_range(t[3]) == 6:9
 end
 
 @testset "SyntaxNode pretty printing" begin

test/test_utils.jl

@@ -36,7 +36,9 @@ using .JuliaSyntax:
     highlight,
     tokenize,
     untokenize,
-    filename
+    filename,
+    byte_range,
+    char_range
 
 if VERSION < v"1.6"
     # Compat stuff which might not be in Base for older versions