Lists of properties become difficult to read and write by hand if they get too long:
data answer=42 author="Douglas Adams" work="The Hitchhiker's Guide to the Galaxy"
data answer=42 \
author="Douglas Adams" \
work="The Hitchhiker's Guide to the Galaxy"
Either you get a very long line or you must take care not to forget the continuation character \ and to remove it after the last property (like , in JSON).
As far as I read the spec, a line must not start with a property (because bare identifiers must not contain =). This would be invalid in KDL 1.0.0:
data answer=42
author="Douglas Adams"
work="The Hitchhiker's Guide to the Galaxy"
Why not let the parser assume continuation of a node in this case (instead of syntax error)?
Lists of properties become difficult to read and write by hand if they get too long:
Either you get a very long line or you must take care not to forget the continuation character
\
and to remove it after the last property (like,
in JSON).As far as I read the spec, a line must not start with a property (because bare identifiers must not contain
=
). This would be invalid in KDL 1.0.0:Why not let the parser assume continuation of a node in this case (instead of syntax error)?