* lisp/org-element.el (org-element-link-parser): "radio" links have
their path downcased to avoid introducing case mismatch with their
relative radio target. With this change it is also necessary to add
contents to them, since `:path' property no longer matches real
value of the link.
(org-element-radio-target-parser): Downcase value as explained above.
Store the initial value in a new `:raw-value' property.
* ox-latex.el (org-latex-link):
* ox-html.el (org-html-link):
* ox-beamer.el (org-beamer-link): Fix links to radio targets.
* ox-ascii.el (org-ascii-link): For links to a radio target,
use the link, not the target.
Thanks to Noah Slater for reporting this.
* lisp/org-agenda.el (org-agenda-get-scheduled): If
`org-agenda-skip-scheduled-if-deadline-is-shown' is set to
'repeated-after-deadline, still show tasks without any deadline
The problem was that the check for the deadline seems to return a
default time even if no deadline is set for the task. This adds a
check if there is a deadline at all, if there is none, the task is
shown in the agenda, otherwise the old semantics apply.
TINYCHANGE
* lisp/org.el (org-entry-get-with-inheritance): Temporarily
let-bind `org-file-properties', `org-global-properties' and
`org-global-properties-fixed' to nil before calling `org-entry-get'
on entries up the hierarchy from the queried entry.
Problem was that when org-entry-get-with-inheritance went up the
hierarchy of entries from a given entry, checking whether the property
has been set in any of the entries, it was calling org-entry-get,
which always looks at file-scope and global-scope properties. So if
our property was set file-wide or system-wide, and somewhere up the
hierarchy there was an entry which set some properties _other_ than
the one we're looking up but did not set ours, org-entry-get would
fill in the global property value and report that our property was in
fact set in that entry. The search would stop, and if the property
was actually set further up the hierarchy (which should override
file-wide or system-wide settings), we would never get to that
up-the-hierarchy setting.
Illustration of fixed problem:
#+PROPERTY: myprop aaa
* headline A
:PROPERTIES:
:myprop: bbb
:END:
*** headline B
:PROPERTIES:
:otherprop: ccc
:END:
#+BEGIN_SRC emacs-lisp
(message (org-entry-get-with-inheritance "myprop"))
#+END_SRC
#+RESULTS:
: aaa
Result should be bbb, which it is after the fix.
* org.el (org-store-link): Fall back on an empty string when
no description is available.
This fixes a bug about tangling a source block outside of any
subtree when `org-id-link-to-org-use-id' is `t'.
Thanks to Sean Allred for reporting this.
* org-capture.el (org-capture-finalize): Ensure to widen the
buffer when the target buffer is not initially narrowed.
Thanks to Adam Spiers and Thomas Proschinger for reporting this bug.
* org-compat.el (org-move-to-column): Fix bug about ignoring
bracket links visibility status in tables with S-RET.
Thanks to Michael Brand for reporting this.
* org.el (org-delete-property): Don't suggest to delete the
CATEGORY property when the category is not explicitely set in
the property drawer. Also enforce matching when completing.
Thanks to Oleh for providing a preliminary patch for this.
* org.el (org-insert-heading): Fix regression: with two
universal prefixes, insert heading at the end of the subtree.
(org-insert-todo-heading): Bugfix: only enforce the first TODO
state when arg is '(4).
* lisp/org-element.el (org-element-context): Fix parsing of bold
objects at the beginning of a headline.
* testing/lisp/test-org-element.el (test-org-element/context): Add
test.
* org.el (org-emphasis-alist): Use `org-verbatim' for =text=
and `org-code' for ~text~.
* org.texi (Emphasis and monospace): Document the use of
=verbatim= and ~code~ to be consistent with
`org-element-text-markup-successor'.
Thanks to Yasushi for reporting this.
* lisp/ox-beamer.el (org-beamer-item): Insert the export snippet right
after the first \item, not all of them.
Thanks to Andreas Leha for reporting it.
* lisp/org-bibtex.el (org-bibtex-read): Check string length before
using aref.
If a field in a bibtex entry is empty:
@article(test,
description = "")
the function org-bibtex-read throws an exception because of
using aref on this empty string.
The solution is to check the length of the string before.
TINYCHANGE
Signed-off-by: Stefan-W. Hahn <stefan.hahn@s-hahn.de>
* lisp/org-element.el (org-element-inlinetask-parser): Fix parsing
when regular and degenerate inlinetasks are mixed in the section.
* testing/lisp/test-org-element.el (test-org-element/inlinetask-parser):
Add tests.
Thanks to Anders Johansson for reporting it.
* org-colview.el (org-columns-display-here): Fix the column
view for numbers with a format specifier (e.g. {+; %5.1f}).
Thanks to OSiUX and Michael Brand for reporting this.
* org-src.el (org-edit-src-code): Throw a warning instead of
an error when loading the mode fails, otherwise the user is
left with unusable buffers.
Thanks to Florian Beck for suggesting this.
* lisp/org-element.el (org-element-normalize-contents): Do not ignore
empty lines when an object follows.
* testing/lisp/test-org-element.el (test-org-element/normalize-contents):
Add test.
* org-table.el (org-table-copy-down)
(org-table-find-dataline, org-table-move-row)
(org-table-insert-hline, org-table-kill-row): Use
`org-move-to-column' with the IGNORE-INVISIBLE arg set to `t',
so that abbreviated rows don't interfer with setting the
cursor back at the correct position.
Thanks to Jason Lewis for reporting one of these bugs.
* lisp/org-element.el (org-element-interpret-data): Do not remove
properties by side-effect when interpreting a string, as it also
removes them from the parse tree, making the string unusable without
its :parent property.
If text properties get in the way, it is also possible to use
`substring-no-properties' but then, a compatibility function would be
required for XEmacs.
* lisp/ob-exp.el (org-babel-exp-process-buffer): Also check
`org-src-preserve-indentation' to know when to preserve indentation.
* testing/lisp/test-ob-exp.el (ob-export/export-and-indentation): New
test.
Thanks to John Hendy for reporting it.
