Skip to content

Honor byte-order mark when parsing UTF-16 and UTF-8 input #65

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
SAY-5 wants to merge 3 commits into
base: main
Choose a base branch
from
+39 −0
Open

Honor byte-order mark when parsing UTF-16 and UTF-8 input #65

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
0c67195
Honor byte-order mark when parsing UTF-16 and UTF-8 input
SAY-5 Jun 18, 2026
3b091a0
test: select root element when parsing declaration-prefixed BOM input
SAY-5 Jun 18, 2026
fbf9065
test: add read(file) round-trip for BOM-prefixed XML inputs
SAY-5 Jun 20, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 16 additions & 0 deletions src/raw.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,23 @@ struct Raw
ctx::Vector{Bool} # Context for xml:space (Vector to support inheritance of context)
has_xml_space::Bool # Whether data contains `xml:space` attribute at least once
end
# Honor a leading byte-order mark (XML 1.0 §4.3.3): decode UTF-16 to UTF-8 and
# strip a UTF-8 BOM so the byte-oriented tokenizer always sees UTF-8.
function _normalize_bom(data::Vector{UInt8})
n = length(data)
if n >= 2 && data[1] == 0xFF && data[2] == 0xFE
units = reinterpret(UInt16, data[3:end])
return Vector{UInt8}(transcode(String, units))
elseif n >= 2 && data[1] == 0xFE && data[2] == 0xFF
units = bswap.(reinterpret(UInt16, data[3:end]))
return Vector{UInt8}(transcode(String, units))
elseif n >= 3 && data[1] == 0xEF && data[2] == 0xBB && data[3] == 0xBF
return data[4:end]
end
return data
end
function Raw(data::Vector{UInt8})#, ctx::Vector{Bool}=Bool[false])
data = _normalize_bom(data)
needle = Vector{UInt8}("xml:space")
has_xml_space = findfirst(needle, data) !== nothing
return Raw(RawDocument, 0, 0, 0, data, [false], has_xml_space)
Expand Down
23 changes: 23 additions & 0 deletions test/runtests.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -642,5 +642,28 @@ end
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'))

# https://github.com/JuliaComputing/XML.jl/issues/62 (UTF-16/BOM input)
text = """<?xml version="1.0"?><root>hello</root>"""
units = transcode(UInt16, text)
le = vcat(UInt8[0xFF, 0xFE], reinterpret(UInt8, units))
be = vcat(UInt8[0xFE, 0xFF], reinterpret(UInt8, bswap.(units)))
utf8_bom = vcat(UInt8[0xEF, 0xBB, 0xBF], Vector{UInt8}(text))
for bytes in (le, be, utf8_bom)
Comment thread
mathieu17g marked this conversation as resolved.
doc = Node(XML.Raw(bytes))
root = only(filter(n -> nodetype(n) == Element, children(doc)))
@test tag(root) == "root"
@test value(only(children(root))) == "hello"

# read(file) round-trip exercises the Mmap.mmap path (the real .xlsx case)
mktemp() do path, io
write(io, bytes)
close(io)
file_doc = read(path, Node)
file_root = only(filter(n -> nodetype(n) == Element, children(file_doc)))
@test tag(file_root) == "root"
@test value(only(children(file_root))) == "hello"
end
end
end

Loading