On 7 January 2018 at 23:38, Hans Hagen <pra...@wxs.nl> wrote: > On 1/7/2018 11:15 PM, David Carlisle wrote: > >> Should not \lastnamedcs expand to the token with empty csname after > >> \csname\endcsname ? > > It's by design (an explicit test and nothing done with a null cs, so no > cleanup side effects too) ... btw, it's already tricky enough this code and > we're not going to change such functionality (just like we don't change > other primitives with side effects).
OK, although the resulting behaviour is a bit odd, but can the manual be fixed to describe the behaviour as the current behaviour is not as described there. thanks, David