Hi Dirk, I have been watching the listserv for replies to this post, but I haven't seen any...
I constantly use pmx and I would be upset if anything happened to prevent me from using it. On the other hand, there is an old saying "Never ask someone to do something which you wouldn't do yourself." Consequently, I would like to volunteer for the project. I have been using LuaLaTeX for about a year (in fact, it was this listserv which put me onto it) and so I have some knowledge of how it works. This was my introduction of Lua, most of my programming has been in C. I would like to know how you envisage the project. It seems to me that it would involve replacing the Pascal and Fortran programs with equivalent ones written in Lau. The user would then load a generic .tex file, insert the name of the file to be processed, and compile that document. TeX would load the Lua script(s) which would translate the .mtx/.pmx files to a .tex file, and then TeX would produce a .pdf file. Does that sound about right? So the project would involve: 1. Produce Lua scripts for .mtx, .pmx. These are best done independently, and one would be useful without the other; and 2. Write the TeX file which calls the Lua scripts and produces the .pdf file (this would be easy). I won't be able to do anything until the end of May, but I would like to start thinking about it. Jim Bailey -----Original Message----- From: TeX-Music [mailto:[email protected]] On Behalf Of Dirk Laurie Sent: February 25, 2017 4:24 AM To: Werner Icking Music Archive <[email protected]> Subject: [Tex-music] Future maintenance of the Pascal source code in M-Tx Although many years ago I posted something on this topic, this post is directly inspired by the discussion on the "r+0.g" error. 2017-02-25 12:51 GMT+02:00 Dirk Laurie <[email protected]>: > 2017-02-25 8:56 GMT+02:00 Andre Van Ryckeghem <[email protected]>: >> 1 1 4 4 4 4 0 0 >> 1 1 16 .0 >> >> t >> ./ >> r8+0.f r.g g.f r+0.g ... > - M-Tx should have no problem with the same input because the dot > shortcut is not passed as-is to PMX, it is preprocessed to make two > notes. However, it does have a problem, since the following PMX line > is produced > > r8+0d r8d g84d f1 r8+0d > > That goes onto my TODO list. I find myself utterly reluctant to do anything more about this one than to start a section of the Manual called "Known bugs and workarounds". The fact that PMX is coded in Fortran, and M-Tx in Pascal, are instances of the Hammer Principle. "It is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail." — Abraham Maslow, 1966 PMX and M-Tx both make heavy use of string processing, lexing and parsing, laboriously coded from scratch in languages offering only not even token support (pun intended) for those tasks. The dot shortcut is a distinctly nontrivial task. The M-Tx source code has these comments to the procedure "extractDotShortcut". { Extracts the dot shortcut part of a note: comma shortcut is no problem because a comma cannot be part of a number. } I.e. the procedure needs some sophisticated awareness of what might be legal note syntax just before the dot. Nowadays TeX has Lua available in just about every installation. I no longer have only a hammer. I have a whole chest filled with modern power tools. It's not fun anymore just pounding things into shape. I am willing to collaborate with anyone who is willing to rewrite M-Tx in Lua. (I had a decent go at doing at alone about a year ago, and found it not to be that easy, though.) I'll also collaborate with anyone who is willing to take over the maintenance of the Pascal source. But ... (verb.sap.) -- Dirk ------------------------------- [email protected] mailing list If you want to unsubscribe or look at the archives, go to http://tug.org/mailman/listinfo/tex-music ------------------------------- [email protected] mailing list If you want to unsubscribe or look at the archives, go to http://tug.org/mailman/listinfo/tex-music
