* lisp/org.el (org-promote):
(org-demote): Turn off after-change-functions to speed up the
reindentation of text.
This is mainly toob-oz: brought up to date with the rest of Babel
* ob-tangle.el (org-babel-tangle): Make it work in an indirect buffer.
The problem was that the message generated after tangling included the
file name of the current buffer which was nil in an indirect buffer.
TINYCHANGE
* lisp/org-agenda.el (org-agenda-get-timestamps): Fix agenda display when headlines are missing
The following entry breaks the agenda display.
,----
| ****
| :PROPERTIES:
| :DateCreated: <2011-04-13 Mi 10:00>
| :END:
|
| [[gnus:nntp%2BGwene:gwene.org.emacsen.planet][gnus:nntp+Gwene:gwene.org.emacsen.planet]]
`----
Provide the empty string as the headline if the search for the
headline returns nil. org-format-agenda-items requires a string for
the headline parameter.
* lisp/ob.el (org-babel-params-from-buffer): Now looking for header
arguments in #+Properties: as well as #+Babel:. Also, we're no
longer caching these results into a file local variable.
* lisp/ob-python.el (org-babel-python-evaluate-external-process):
Allow parsing as a table in the case of ":results output table".
(org-babel-python-evaluate-session): Allow parsing as a table in
the case of ":results output table".
* lisp/org-agenda.el (org-agenda-bulk-action): Allow bulk scatter
in all possible agenda views. Use `org-agenda-schedule' instead of
`org-agenda-date-later'.
The bulk scatter command so far shifted the date that was causing an
entry to appear in the agenda. However, the true intend was to
reschedule onto dates in the near future. This patch fixes this
issue. A side effect is that you can now bulk scatter tasks that to
not yet have a date, for example also tasks picked from the TODO
list.
* lisp/org.el (org-entry-get):
(org-entry-delete):
(org-entry-put):
(org-property-values):
(org-delete-property-globally): Use org-re-property.
(org-re-property): New function allowing to build a regexp to match a
property.
Signed-off-by: Julien Danjou <julien@danjou.info>
The problem is illustrated by the following example:
* Illustration of bug in html export
- This has a space after the colons :: so will work in latex and html
- This doesn't have a space after the colons ::so is an invalid
description item according to the org manual. Won't work in html
or docbook. Will nevertheless work in latex, provided /first/
description item is valid.
- Has a terminating space ::
- So it works in both html and latex export!
- Even though it's difficult to distinguish from the next example.
- Lacks a terminating space ::
- At present, *doesn't* work in html or docbook export, does in
latex. This is the case that the following patch fixes.
* lisp/org-table.el (org-table-cleanup-narrow-column-properties): Use
point-min rather than 1 when moving in the buffer.
Signed-off-by: Julien Danjou <julien@danjou.info>
* lisp/org-list.el (org-list-in-valid-context-p): renamed from
org-list-in-valid-block-p.
(org-at-item-p,org-list-search-generic): use renamed function.
* lisp/org.el (org-fill-item-nobreak-p): new function.
(org-set-autofill-regexps): modify `fill-nobreak-predicate' to
prevent auto-fill from breaking line just before an item regexp.
* lisp/org.el (org-set-font-lock-defaults): Be more strict when
recognizing description items, and do not fontify wrong cases like:
"- term ::description" or "1. term :: description"
* lisp/org-latex.el: pay attention to end of footnote. Before closing
the command, ensure that list is properly closed or that last link
is separated from the curly brace.
* lisp/org.el (org-indent-line-function): footnote definition must
stay at column 0 to be recognized as such. Body below can have
normal indentation, so it should ignore its definition when
computing indentation.
* org-capture.el (org-capture-clock-keep): New local variable.
(org-capture, org-capture-finalize): Use it and fix a bug when
:clock-keep is set to `t'.
* lisp/org-agenda.el: Remove obsolete code for adding todo list in front
of the agenda
Todo lists should be built with the TODO list or a block agenda instead.
This code was triggered for C-u R in block agendas which is used to
limit the clock report to the current tags filter applied to the
agenda.
* lisp/org-latex.el (org-export-latex-convert-table.el-table): Fix
format-string for insertion of captions.
This fixes a bug introduced in caa5da1 since when the caption itself
is not placed inside the caption command.
* lisp/org-exp.el (org-export-remove-comment-blocks-and-subtrees): Fix regexp.
Rasmus <rasmus.pank@gmail.com> wrote:
> Hi,
> It seems that
>
> ,----
> | #+BEGIN_COMMENT
> | ...
> | #+END_COMMENT
> `----
>
> Might be broken in Org-mode 7.5. According to the manual,
>
> ,----
> | Finally, regions surrounded by
> | #+BEGIN_COMMENT ... #+END_COMMENT will not be exported.
> `----
>
> Consider the following example generated with Org-mode 7.5 in Emacs 24.0.50
> (started without config files).
>
> ,----
> | #+TITLE: this is a test
> |
> | #+BEGIN_COMMENT
> | Don't export me
> | #+END_COMMENT
> |
> | I'd like to be exposed
> `----
>
> The HTML export is:
>
> ,----
> | <snip>
> | <h1 class="title">this is a test</h1>
> |
> | <div class="org-comment">
> | </br>
> | <p>
> | <p>
> | Don't export me
> | </p>
> | </div>
> |
> | I'd like to be exposed
> | <snip>
> `----
>
> The LaTeX export is:
>
> ,----
> | \begin{verbatim}
> | Don't export me
> | \end{verbatim}
> | I'd like to be exposed
> `----
>
Indeed.
It seems that one of the hair-raising regexps that Carsten manages
to come up with is not quite right. Try this patch for now:
--8<---------------cut here---------------start------------->8---
--8<---------------cut here---------------end--------------->8---
Nick
* lisp/ob-lisp.el (org-babel-execute:lisp): Turn vectors into lists
before reading by elisp
(org-bable-lisp-vector-to-list): Stub of a vector->list function,
should be replaced with a cl-vector->el-vector function.
* lisp/org.el (org-end-of-meta-data-and-drawers): New function.
* lisp/org-capture.el (org-capture-place-plain-text): Implement adding
plain text templates to Org nodes.
This is something which came out of a discussion with Philip Rooke, in
the thread
Philip tried to use a capture template with template type plain, but
using a date tree as a target. Plain templates where placed at the
end of the file, not at the end of the entry. I complained first that
mixing headlined entries and plain snippets into the same capture
target is not possible, but I realized that there is a way to make
this work OK. The headlined entries become children, and the plain
text snippets become part of the text before the first child.
* lisp/org-exp.el (org-export-handle-comments): Add the org-protected
property to the replacement string.
Although org-export-handle-comments adds the org-protected property to
the matched string, the subsequent `replace-match' call to change the
comment character does not add this property to the entire format
string. Fix this by propertizing the entirety of the newtext argument
to replace-match.
* lisp/org-html.el (org-export-as-html) (org-html-level-start): Fix
logic for section number printing when NUM is an integer.
Fixes a bug introduced in 9f57b8e which considered all non-integer
values of the num option to be nil.
* lisp/org-latex.el (org-export-latex-special-chars): Fix regexp for
`single' special characters and ellipsis.
Repeated special characters are exported differently depending on
their position in the buffer. A "&&" string at the start of a line is
exported as "&\&" whereas in the middle of a line you get "\&\&". The
former is incorrect. Fix this by matching the beginning of a line
before a character. While we're at it, amalgamate the regexps for the
different special characters.