This project doesn't work in Emacs 25

[ginqi7]

The EIEIO package has updated in Emacs 25. Some function used in o-blog has been obsolate. It makes this project can not run.

Debugger entered--Lisp error: (invalid-slot-name "#<ob:framework-component ob:framework-component>" lead) signal(invalid-slot-name ("#<ob:framework-component ob:framework-component>" lead))

I checked the codes of o-blog. I find that it use set-slot-valuefunction, and this function hase been obsolate in EIEIO package. There may be other function that can not be used.

[carlhenrikek]

Did you have any luck with sorting this out? I'm still unable to make this run in emacs-25.

[mkrauss]

Updating set-slot-value forms with CLOS style setf forms would be pretty easy, but there seem to be a lot of things failing in Emacs 25. This is too bad, it is making o-blog unusable. This seems to be the only real package of it's kind and I'd be willing to help fix it up but I'm not even sure where to begin, and there has been no activity for a while... @renard - is this project dead?

[jonnay]

#162 at least gets the package loading. I am still having problems, but maybe I can get somewhere with it.

[cviebig]

+1 :-)

[jonnay]

I'm kinda stuck at ob:parse-config. https://github.com/renard/o-blog/blob/master/lisp/o-blog-backend-org.el#L99

ob:get-name just returns the object name which if you're doing an org export is "ob:backend:org". So why would it then treat the result of that as if it were a file?

[jonnay]

Yeah. All the problems I'm having now are related to ob:get-name. There are a number of functions that expect it to return the file it is working on, and instead it returns the classname. I don't understand why you would expect the class name to be a file name as well.

[jonnay]

Sorry Guys, I'm hosed. I don't think I can fix this. It seems like I would be fighting a version change in oblog, as well as emacs. So I'm out. :/