> > There is a bug in the documentation -- `org-export-babel-evaluate' is > obsolete. It should say `org-export-use-babel’.
I don’t see org-export-use-babel in available options. I cannot customise it. '(org-export-use-babel (quote inline-only)) Seems to have no effect. Is that so because, as you say later, users are not supposed to touch this option? > But that part of the manual is irrelevant to what the `:eval' header does. > There is no `inline-only' value in either the documentation or in the lisp > code. So, `inline-only' acts like `yes' for the reason I stated earlier. > > >> With current org, I get this behaviour only if I globally set the option. >> But that somehow disables ":results”. >> > > Right. `:results' is a babel header. When babel is off, the babel headers > are not acted upon. > > Setting `org-export-use-babel' to `nil' or `inline-only' turns off babel for > src blocks. > > > The behavior was purposely changed. With `inline-only' none of the babel > operations will be executed for src blocks -- i.e. the src blocks and > existing results (if any) will be exported as is. > > Users should avoid touching `org-export-use-babel' for almost all purposes. > It does not do the equivalent of setting `:eval' globally. > I just saw this thread: https://lists.gnu.org/archive/html/emacs-orgmode/2017-02/msg00443.html <https://lists.gnu.org/archive/html/emacs-orgmode/2017-02/msg00443.html> I certainly feel that the current behaviour is a regression. The user should have the possibility of controlling globally or buffer-wide whether the source codes are evaluated at the time of export or not. org-export-use-babel does not do that. It turns off babel completely, exporting codes and existing results as is. This is clearly a regression as far as my use case is concerned. Vikas
