* lisp/ox-latex.el (org-latex-src-block, org-latex-keyword,
org-latex-inline-src-block, org-latex-template,
org-latex--caption/label-string, org-latex-engraved-preamble,
org-latex-listings): Replace `org-latex-listings' with
`org-latex-src-block-backend', which now can be set to listings/verbatim
and no longer advertises t/nil as valid values.
* lisp/ox-beamer.el (org-beamer-template): Update in the same manner as
`org-latex-template'.
* lisp/org-compat.el: Make `org-latex-listings' an obsolete alias for
`org-latex-src-block-backend'.
* testing/lisp/test-ox.el: Replace `org-latex-listings' reference with
`org-latex-src-block-backend'.
* doc/org-manual.org (Footnotes, LaTeX specific properties, Literal
Examples): Replace references to `org-latex-listings' with
`org-latex-src-block-backend'.
* etc/ORG-NEWS: Add a news entry noting this change.
The variable `org-latex-listings' originally indicated whether source
blocks should use the listings LaTeX package, or not. This usage has
evolved over the years, and now it sets one of four different
fontification backends. This renaming should make the variable name a
bit less misleading.
* lisp/ox-latex.el (org-latex-src-block--listings,
org-latex-src-block--engraved, org-latex-src-block--minted,
org-latex-src-block--custom): Refactor these --backend functions to use
cl-defun with keys instead of having an 11-argument signature.
(org-latex-src-block): Adjust for the change in signature of the
--backend functions, and refactor the `cond' statement to use `pcase'.
* lisp/ox-latex.el (org-latex-src-block--listings,
org-latex-src-block--engraved, org-latex-src-block--minted): Use
`string-width' instead of `length' to gauge the displayed width of the
string in LaTeX. This may not be a perfect match, but it should be an
improvement.
* lisp/ox-latex.el (org-latex-src-block, org-latex-src-block--engraved,
org-latex-inline-src-block, org-latex-inline-src-block--engraved,
org-latex-src--engrave-code, org-latex-template, org-latex-listings):
Make use of the engraved-faces package (available on ELPA) to provide an
alternative LaTeX code highlighting backend which functions similarly to
htmlize.el for HTML exports.
(org-latex-engraved-preamble, org-latex-engraved-options): Introduce
variables to construct the preamble for engraved code blocks.
* lisp/ox-beamer.el (org-beamer-template): Modify to add engrave-faces
preamble when applicable.
* lisp/ox-latex.el (org-latex--make-option-string): Support a custom
option seperator string.
(org-latex--make-option-string, org-latex-minted-options,
org-latex-listings-options): The first line of the docstrings for
`org-latex-minted-options` and `org-latex-listings-options` describe the
variables as "association lists", yet `org-latex--make-option-string`
does not handle association lists and an example of two-value lists is
provided. To make the behaviour match the docstring,
`org-latex--make-option-string` is modified to work with either a list
two-value lists or an association list, and the examples in
`org-latex-minted-options` and `org-latex-listings-options` updated
accordingly.
* lisp/ox-latex.el (org-latex-inline-src-block,
org-latex-inline-src-block--minted,
org-latex-inline-src-block--listings): Extract the minted and listings
specific logic out of `org-latex-inline-src-block` into the new
functions `org-latex-inline-src-block--minted` and
`org-latex-inline-src-block--listings`.
* lisp/ox-latex.el (org-latex-src-block): Extract the per-format logic
from `org-latex-src-block' into new dedicated functions:
+ `org-latex-src-block--verbatim'
+ `org-latex-src-block--custom'
+ `org-latex-src-block--minted'
+ `org-latex-src-block--listings'
This makes `org-latex-src-block' much less monolithic, taking it from
175 lines to 30, and I find also makes it easier to understand.
* lisp/org-fold-core.el (org-fold-core-first-unfold-functions):
Remove. This is very too tricky to implement. In particular, when we
use jit-lock, there can be a race between calls to
`org-fold-core-region' and fontification. For example, consider that
one of the functions in the hook folds drawers and everything is
folded initially. Then, if the user requests unfolding headlines,
drawers, and blocks, it happens before jit-lock is ran in invisible
parts of buffer (out of the display). Those parts will be folded
despite user explicitly requested full unfold, which is wrong. Fixing
this behaviour would involve implementing a full-fledged tracker on
top of jit-lock, which is very too complex and will likely introduce
hard-to-debug errors.
* lisp/org-fold-core.el (org-fold-core-initialize):
(org-fold-core-region):
(org-fold-core-remove-optimisation):
* lisp/org-cycle.el (org-cycle-hide-drawers):
(org-cycle-hide-archived-subtrees): Do not use
`org-fold-core-first-unfold-functions'.
Fixes https://list.orgmode.org/80k0ar1lml.fsf@felesatra.moe/T/#u
* lisp/org.el (org-fast-tag-selection): Disabled tags face
A tag can be either selected in "current", "inherited" or disabled. When
disabled, it should have the default face. Previously, if a tag was
already selected, when deselection it will still have an org-todo face.
https://lists.gnu.org/archive/html/emacs-orgmode/2021-10/msg00350.html
* lisp/org-fold.el (org-fold--reveal-outline-maybe): Widen the buffer
before testing headline validity. Otherwise, `goto-char' may not move
where expected in narrowed buffer.
* lisp/org.el (org-set-regexps-and-options): Change
`org-complex-heading-regexp-format' initialization so that the regexp
it produces also matches a headline if it is commented.
TINYCHANGE
* ob-scheme.el (org-babel-scheme-execute-with-geiser): If available,
use the synchronous `geiser-eval-region/wait' function to get the
evaluation result.
Works with geiser 0.18 (before introduction of async evaluation) and
geiser 0.22 (which introduces `geiser-eval-region/wait').
Modified from a patch proposal by Felipe Lema.
TINYCHANGE
* lisp/org-fold-core.el (org-fold-core-folded-p): Allow list of fold
types to be passed as argument.
* lisp/org-fold.el (org-fold-check-before-invisible-edit--overlays):
(org-fold-check-before-invisible-edit--text-properties): Make
overlay/text-property behaviour consistent. Do not handle edits
inside folded links. Fix inconsistencies.
* testing/lisp/test-org-fold.el:
* testing/lisp/test-org.el: Move folding-related tests to new test
file.
* testing/lisp/test-org-fold.el:
(test-org-fold/org-catch-invisible-edits): New test.
See https://orgmode.org/list/m2o809q170.fsf@gmail.com
* lisp/ob-plantuml.el (org-plantuml-args): Rename
`org-plantuml-executable-args' to `org-plantuml-args'.
(org-babel-execute:plantuml): Use `org-plantuml-args' when calling
jar.
* lisp/org-compat.el (org-plantuml-executable-args): Obsolete old
variable name.
* etc/ORG-NEWS (=org-plantump-executable-args= is renamed and applies
to jar as well): Document change.
See the discussion in https://list.orgmode.org/orgmode/87y23hr045.fsf@localhost/
*
lisp/org-fold.el (org-fold-check-before-invisible-edit--text-properties):
Consider `border-and-ok-direction' value when deciding whether to
throw an error and reveal the fold. Never throw an error when editing
at border without affecting the text inside.
Fixes https://orgmode.org/list/m2ilqle995.fsf@gmail.com
*
lisp/org-fold.el (org-fold-save-outline-visibility--text-properties):
Do not expand list of specs on compile time. Simplify saving and
re-applying folds.
* lisp/org-fold.el (org-fold-save-outline-visibility): Fix typo when
-> if.
* lisp/ob-tangle.el (org-babel-tangle-single-block): If the
argument to :tangle is nil (e.g. when parsing conditional
tangling such as (when condition "yes") the current code
throws an error in `file-name-directory'. This commit
checks if the argument is a string before calling the fun
TINYCHANGE
* lisp/org-macs.el (org-fill-template): Fix a bug in template
expansion: if one key is a substring of another key (like "noweb" and
"noweb-ref", or "tangle" and "tangle-mode"), the second key wouldn't
get expanded properly. The problem was that keys were processed in
order of increasing length; they are now sorted in order of descending
length.
TINYCHANGE
* testing/lisp/test-org-macs.el (ert-deftest test-org/fill-template):
There is a bug in `org-fill-template': it sorts and processes keys in
order of increasing length, so that "noweb" is seen before
"noweb-ref", and "tangle" before "tangle-mode". So in a template that
includes "%noweb-ref", it will substitute the value of "noweb".
This change includes a test for this bug.
TINYCHANGE
* lisp/org-inlinetask.el (org-inlinetask-hide-tasks): Make sure that
we loop over inlinetasks inside current section only.
`org-inlinetask-goto-end' moves point to the end of the inlinetask
element. If an inlinetask is followed by a heading, the function
logic will treat the heading as the beginning of another inline task.
Prevent it by moving the point at the end of the END line of the
inlinetask.
Fixes
https://list.orgmode.org/orgmode/CAP7OBx+rs-7v4jstEFYC4u0=0NZ3xJKfb3CNrWeDvjmwRTvgxw@mail.gmail.com/
* lisp/ob-core.el (org-babel-expand-noweb-references): Add support for
`noweb-prefix' header argument, to not repeat the prefix characters
when expanding a noweb reference.
(org-babel-common-header-args-w-values):
(org-babel-safe-header-args): Add `noweb-prefix' value.
* doc/org-manual.org: Document `noweb-prefix' babel header argument.
* etc/ORG-NEWS: Document `:noweb-prefix'.
* lisp/org.el (org-todo): Insert new todo keyword before
`save-excursion' markers. This avoids apparent point movement when a
new keyword is inserted into an empty heading.
Before: "* <point>" -> "*<point> TODO";
After: "* <point>" -> "* TODO<point>";
Fixes https://orgmode.org/list/87mtg6265i.fsf@yandex.com
* lisp/org-refile.el (org-refile): Do not reveal the refiled headline.
This reverts the new behaviour accidentally introduced by org-fold
patchset.
See https://orgmode.org/list/87r15l5kce.fsf@gnu.org
* lisp/org-fold.el (org-fold--isearch-reveal): Call
`org-fold-show-visibility'. The previously used
`org-fold-show-set-visibility' does not accept context as argument.
Reported in https://orgmode.org/list/8735i5gd8n.fsf@gmail.com
* lisp/org-fold-core.el (org-fold-core-fontify-region): Collect all
the return values of `font-lock-default-fontify-region' on visible
regions and extend jit-lock-bounds accordingly.
* lisp/org-macs.el (org-string-width): Do not remove the STRING faces
unless character width is requested. Faces might alter pixel width of
the string and removing them would yield incorrect results.
* lisp/org-fold-core.el (org-fold-core-fontify-region): Run
`org-fold-core-first-unfold-functions' on all the regions unmarked by
'org-fold-core-fontified property instead of using incorrect
heuristics with `point' position. Make sure that fontified region is
registered as fontified according to the return value of
`font-lock-default-fontify-region'
Fixes https://orgmode.org/list/8735i5gd8n.fsf@gmail.com
* lisp/org-agenda.el: Re-enable native compilation as it does not
cause any issues with agenda rendering without other custom changes
not included in this branch.
* lisp/org-fold-core.el (org-fold-core-fontify-region): Handle FORCE
argument better. Skip unnecessary code parts when FORCE is non-nil.
Assign `fontified' text property manually in the actually fontified
regions. We cannot just supply correct return value since jit-lock
does not allow piecewise fontification.
TEC
Ihor Radchenko
Daniel Fleischer
Ignacio
Tim Van den Langenbergh
Jacopo De Simoi
Andrew Arensburger
Sébastien Miquel
Nick Dokos
Nick Dokos