#asdf3 — Public Fediverse posts

#commonLisp #ASDF #asdf3 #remindMe #'ASDF/BACKWARD-INTERFACE:SYSTEM-DEFINITION-PATHNAME

#commonLisp #ASDF #ASDF3 #Mastoquestions #programming hey so how am I meant to know what got :mix'ed into a package as by #'uiop:define-package programmatically?
I'm using :class :package-inferred-system anyway, so I'm referring to packages by identifying the file they are identified with in a given system: Am I meant to just open the file, and then #'ASSOC :mix in (cddr (read)) ?

#commonLisp #ASDF #ASDF3 #Mastoquestions #programming hey so how am I meant to know what got :mix'ed into a package as by #'uiop:define-package programmatically?
I'm using :class :package-inferred-system anyway, so I'm referring to packages by identifying the file they are identified with in a given system: Am I meant to just open the file, and then #'ASSOC :mix in (cddr (read)) ?

@rwxrwxrwx
I also eventually noticed that the ASDF manual recommends that :depends-on or #ASDF3.1(error ".." ), though in the package-inferred-system section they use the feature test one, which I think is also what @svetlyak40wt was thinking.

So now, if I want to have polynomial rings from this package aside from buchberger's, I can load :st-buchberger/src/arithmetic instead to just have the polynomial-ring arithmetic for another package. That's my thinking anyway.

@rwxrwxrwx @julioj @svetlyak40wt @louis @lispi314
I uploaded my small, systematic change to Juan's #:cl-buchberger . You can see logic got moved out of the defsystem and into per-file uiop:define-package. My working name for my patch is :st-buchberger so I could switch.

What do you think of the new style asdf 3.1 packaging? I put some notes at the top of README.md.
gopher://tilde.institute/1/~screwtape/st-buchberger/
https://gopher.floodgap.com/gopher/gw.lite?gopher://tilde.institute:70/1/~screwtape/st-buchberger/

#ASDF3 #CommonLisp #packaging #lisp #algebra #math

@rwxrwxrwx @julioj @svetlyak40wt @louis @lispi314
I uploaded my small, systematic change to Juan's #:cl-buchberger . You can see logic got moved out of the defsystem and into per-file uiop:define-package. My working name for my patch is :st-buchberger so I could switch.

What do you think of the new style asdf 3.1 packaging? I put some notes at the top of README.md.
gopher://tilde.institute/1/~screwtape/st-buchberger/
https://gopher.floodgap.com/gopher/gw.lite?gopher://tilde.institute:70/1/~screwtape/st-buchberger/

#ASDF3 #CommonLisp #packaging #lisp #algebra #math

@mnl In #commonlisp #asdf3 would optionally cook a set of miscellaneous other files generally into special scope variables in a sub/package.

So just have a my-package/templates namespace compiled into the executable with the available blobs interned and exported? So
> my-package/templates:yaml-001
"however:
yaml = goes;
again = I;
forgot:
about = it;
use:
sexps = or;
structs = instead?;"
;; in lisp we would convert configs to sexps (lists) or #S(tructs)

My tidy and rewrite of the initial binry-hop package I made, using #CommonLisp #ASDF3 package-inferred-system. #gopher
gopher://tilde.institute/1/~screwtape/binry-hop/
https://gopher.floodgap.com/gopher/gw.lite?=tilde.institute+70+312f7e7363726577746170652f62696e72792d686f702f
I put the old code in a dated subfolder of old/ on the institute
A little writing about my experiences with using package-inferred-system to recompile specific files in my running image:
gopher://tilde.institiute/0/~screwtape/206491811-package-inferred-system-asdf-niceties.txt
https://gopher.floodgap.com/gopher/gw.lite?=tilde.institute+70+302f7e7363726577746170652f3230363439313831312d7061636b6167652d696e6665727265642d73797374656d2d617364662d6e696365746965732e747874