Re: make_text related scanning improvements
Marcel Krueger <[email protected]> Fri, 18 May 2018 12:01:49 +0200
| Newsgroups | gmane.comp.tex.metapost |
|---|---|
| Message-ID | <[email protected]> |
---- On Fri, 18 May 2018 11:32:57 +0200 Hans Hagen <[email protected]> wrote ---- > On 5/18/2018 11:22 AM, Taco Hoekwater wrote: > > Hi, > > > >> On 18 May 2018, at 11:15, Hans Hagen <[email protected]> wrote: > >> > >> changing fundamentals like this nearly always have side effects ... you cannot always know how these tools are used (for decades) ... you solve one problem and introduces a new one .. such is the world of tex / mp > > > > As the original ‘author’ of the reimplementation I can attest to that! > > > > It took half a dozen minor adjustments to the reimplementation just to make it compatible with pre-existing metapost source files, due to subtle differences between the Pascal and C handling of whitespace. I am not saying the code cannot be improved at all, but it seems likely that any changes at this point will have side-effects that break already existing metapost source files. > right, so > > btex ... etex > > will stay as it is, but i'll look if > > verbatimtex ... etex > > can optionally (!) deal with newlines Awesome. Regarding breaking existing `make_text` callbacks by allowing newlines: As far as I can tell a make_text callback should already be able to deal with newlines in the passed string, because the current version already allows a user to do something like maketext("abc" & char10 & "def") Anyway, thank you very much for your feedback. Marcel -- http://tug.org/metapost/