Bruno Le Floch

I notice another option: tell users to use the already currently working \input filename syntax since when there are no braces the primitive is used, if I understand correctly. I...

For the parsing issue with one-character file names, what could be done is to require that people do not follow such a single-character \input{a} directly by some text, namely we...

Silly me. I forgot about `\aftergroup`. The approach you describe is more general: it does not require hooking into the group's end itself. It is uglier indeed to work globally,...

Perhaps having a function to save a value and one to retrieve it may be safer? I'm worried about cases where there are two groups under the hood, e.g. `xparse`...

My first reaction to your last code Will is that it may be clearer with `\group_execute_after:x` replaced by `\tl_gset:Nx \g_code_tl` and after `\group_end:` put `\tl_use:N \g_code_tl`. The harder part of...

In the special case of running some code just after a group ends perhaps a `\group_end_do:x` (equivalent to `\use:nx \group_end:` would be welcome. To reuse Will's earlier example (notwithstanding the...

Looking at the current code again, I see that `\msg_redirect_class:nn { } { }` (i.e., with an empty target) cancels the redirection, so that there is no redirection left. We...

Simply adding `\DoNotIndex{\cs_if_exist:N}` should be enough (it is enough in an MWE), but my first attempt (adding it before `\part{Bootstrapping}` in `source3body.tex`) didn't seem to work. Of course the issue...

With our newfound ability to provide code to all five engines (well, probably not needed for LuaTeX), we could imagine adding a primitive, or maybe simply extend the syntax to...

One difficulty in current engines IIRC is that xetex (by default at least) doesn't let you write any sequence of bytes that is not valid utf8. So if the original...