| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
- correct process call (to check version of guile) to make sure that it's executed in the remote host
- implement `geiser-guile-ensure-scheme-dir` that will (somehow) make sure the scheme files that need sourcing will be available to remote process
- use `geiser-guile-ensure-scheme-dir` instead of `geiser-guile-scheme-dir` in the rest of the code
- cache the guile files being sourced in `geiser-guile-scheme-local-dir` ensure process is called in remote host
|
|
|
|
|
|
And precludes cancellation of asynchronous evaluations.
|
|
|
|
|
|
Fixes issue #5, and then more.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Via a custom variable, as usual, and defaulting to not use them, as it
provides a fuller interactive experience. Very tempted to name this
flag heed-mathias-counsel-p...
|
|
|
|
|
|
|
|
|
|
|
|
Fixes emacs-geiser/guile#9
In geiser-eval REPL meta-command:
All `mod`, `form` and `args` are now syntax objects. The
geiser-guile's logic will handle `mod` and `form` as is because
they're just passed to guile's eval and compile procedures.
`args` are processed by geiser-eval meta-command itself, so
it's necessary to convert it back to a datum. We lose some metadata,
but all elements in the `args` list are also syntax objects so I don't
think it's a big deal.
`syntax->datum` was introduced before guile 2 so this change is
backward compatible.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
See gitlab issue #303, where Sergey Trofimov kindly described not only
the symptons, but this cure.
|
|
|
|
Starting with Emacs 27 cl is fully deprecated, including at
compile-time.
|
|
|
|
It's the convention and by following it we make a big step towards
supporting outline navigation.
The convention doesn't say much about what parts of the code are
supposed to be part of that sections and what parts belong in a
subsequent section. Here we put the `require' forms in this section
and maybe some setup code, that's a popular approach.
In most cases there was " " where we now insert "Code:". They both
serve a similar purpose and we keep the former because some users
depend on that for navigation. We even add this " " in libraries
where it previously was missing.
In some cases the permission statement was followed by a commentary,
which obviously does not belong in the "Code:" section. In such cases
add the conventional "Commentary:" section.
|
|
It's the convention and by following it we make a big step towards
supporting outline navigation.
|
|
A similar idea should probably be used with other schemes, but right
now i feel ashamed of having taken so long to fix this one (assuming
it's fixed!), so let's rush this commit for a change.
|
|
And by a new pair of glasses in the process.
|
|
Those two procedures are gone in these 2.2 times, and things like
autodoc and xref were broken as a result. With Andy's help,
apparently good enough approximations of their functionality are now
in place: let's see how they go.
|
|
Those two procedures are gone in these 2.2 times, and things like
autodoc and xref were broken as a result. With Andy's help,
apparently good enough approximations of their functionality are now
in place: let's see how they go.
|
|
* elisp/geiser-guile.el (guile--manual-look-up): Change parameters for
info-lookup-symbol to string and 'scheme-mode. Fixes lookup.
|
|
With that in the documentation, i'd taken less time to remember the
very existence of geiser-guile-load-path, and the fact that paths are
added also to the compiled load path... but then i guess it's nice to
re-read my code once in a while.
|
|
We were expanding the path of files to be loaded at the wrong place in
the wrong way. This should be better and address bug #196.
|
