Re: [Axiom-developer] MonoidWithZero

[Franz Lehner]

> What you wrote appears to be correct (I have not tried to compile it
> yet). You have modified 'mring.spad.pamphlet' to define the new
> category, so you need to ensure that when you re-build Axiom it finds
> this new code. How this works depends on what version of the Axiom
> sources you are using.
I still use wh-sandbox although this will be obsolete at some point as I 
understand.

> So to keep it as simple as possible, I would recommend that after
> making this change, you do a complete re-build of Axiom.
This is what I was trying to do.
MRING.spad is generated in the weave step but the MRCategory chunk does 
not make it there for some reason.
It appears however in 
src/algebra/extract-spad.mk
Grepping around I saw that e.g. AbelianMonoidRing
also appears in various other places of the source,

src/doc/bookvol1.pamphlet
src/doc/topics.data
src/doc/book.pamphlet
src/share/algebra/compress.daase
src/share/algebra/libaxiom.al
src/share/algebra/command.list
src/interp/daase.lisp.pamphlet

to name a few which may be relevant.
Do I have to insert the new category into these databases per hand to
make it visible?

I tried recompiling modules with )compile with other patches earlier but 
this somehow confuses the precedence in the interpreter.
For example, taking inverses in  Matrix Integer  called the inverse 
function from  FreeGroup Matrix Integer ...
Anyways I do want to have it in the library, because I compile my Aldor 
domains externally with a Makefile.

> When using
> build improvements the simplest way to do this is to modify the local
> working copy of the repository and then do a new out-of-source build.
that's what I did with the results noted above.

> Of course instead of modifying the original source, for testing at
> least, you could just re-compile these modules using ')compile' in the
> Axiom interpreter.
yes, manual notangle + )compile works.

regards
Franz