The timestamp directory now uses SHA1 hashed versions of the path to each publishing file. This should be a consistent and system-independent way to handle things. The change means that your next publishing command will publish each and every file again, but just once, until new time stamps are in place.
In order to create a more CSS-like setup of the HTML style information, the following changes have been made:
org-export-html-style-default and should not be changed
anymore.
org-export-html-style is now
just the empty string. This variable should receive
settings that are Org-wide. When using org-publish, this
variable is associated with the :style property and can
be used to establish project-wide settings.
org-export-html-style-extra that
should be used for file-local settings. Org-publish can, if
necessary, access this variable with the :style-extra
property.
are all inserted into the HTML header, in the given sequence.
This follows a proposal by Rustom Mody.
You can now set attributes in hyperlinks that will be used when publishing to HTML. For example, if you want to use the ALT and TITLE attributes of an inlined image, here is who to do this:
[[./img/a.jpg{{alt="This is image A" title="Image with no action"}}]]
Thanks to Charles Chen for this idea.
In a link, you can now leave out the "file:" prefix if you
write an absolute file name like /Users/dominik/.emacs or
~/.emacs, or if you write a relative file name by using
./ or ../ to start the file path. You cannot write a
plain file name, because plain text is interpreted as an
internal link.
So for example, a link to an image A.jpg with a thumbnail B.jpg can now be written like
[[./A.jpg][./B.jpg] ]
Deadline and scheduling time stamps are now treated differently in iCalendar export. The default behavior is now the following:
Of course this would not be Emacs if you could not configure
exactly what you want. Take a look at the variables
org-icalendar-use-deadlines and
org-icalendar-use-scheduled if you want to go back to the
old behavior or even do something completely different.
Thanks to Karen Cooke for triggering this change.
If the comparison value in a property search is a string that is enclosed in angular brackets, a time comparison will be done. For example
+DEADLINE>="<2008-12-24 15:20>"
looks for entries with a deadline on or after that time. Special allowed values are "<now>" (with time) and "<today>" (date only).
This is based on a request by Manish.
When set, a link pointing to a directory will actually open the index.org file in that directory. This is a good setting inside a publishing project. When not set, you get a finder/explorer window for that directory, or dired, depending on system and setup.
This follows a request by Richard Riley.
These parameters specify prefixes for each line of included text. :prefix1 is only for the first line, :prefix for all other lines.
This follows a proposal by Richard Riley.
This option can be used to switch the style of the index produced by org-publish. Can be `list' (index is just an itemized list of the titles of the files involved) or `tree' (the directory structure of the source files is reflected in the index). The default is `tree'.
Thanks to Manuel Hermenegildo for the patch.
v will toggle inclusion of trees with the
ARCHIVE tag, this includes obviously the archive sibling.
This was triggered by a proposal by Manuel Hermenegildo.
If I were to name my releases, this one would be called "Adam". Adam, you definitely owe me a beer :-). And I owe you one, too - thanks for all the great ideas.
It used to call org-cut-special, but that is also at bound
to the key C-c C-x C-w.
The date at the cursor in the agenda (and also in the calendar) can now be used to schedule entries, or to set the date in a remember template correctly. It is also designed to make it easier to move an entry to a date picked in the agenda. Thanks to Thomas Baumann for starting the thread that led to this development.
If you want to use the date at the agenda cursor in a
remember template, start remember from the agenda with the
keys k r. While the template is being filled in, the
default date for all time stamps, and also for all
interactive escapes like %^t is now the date at the cursor
in the agenda. The exact same command can also be used from
the calendar if you prefer that.
You may now pick the date for scheduling an item or for setting a deadline in the agenda, where you have the best overview over free time slots. This is a two step process.
k m. In an org-mode file,
this is on C-c C-x C-k.
k s to schedule, or k d to put a deadline. The
agenda is not updated immediately, press r if you want
it to show the affected entry in the right place.
Org has sophisticated mapping capabilities to find all entries satisfying certain criteria. Internally, this functionality is used to produce agenda views, but there is also an API that can be used to execute arbitrary functions for each or selected entries. The main entry point for this API is:
-- Function: org-map-entries func &optional match scope &rest skip
Call FUNC at each headline selected by MATCH in SCOPE.
FUNC is a function or a lisp form. The function will be
called without arguments, with the cursor positioned at
the beginning of the headline. The return values of all
calls to the function will be collected and returned as
a list.
MATCH is a tags/property/todo match as it is used in the
agenda tags view. Only headlines that are matched by
this query will be considered during the iteration.
When MATCH is nil or t, all headlines will be visited by
the iteration.
SCOPE determines the scope of this command, it can
specify a file, all agenda files, the current tree and
much more.
The remaining args are treated as settings for the
skipping facilities of the scanner.
The function given to that mapping routine can really do anything
you like. Here is a simple example that will turn all entries in
the current file with a tag TOMORROW into TODO entries with the
keyword UPCOMING. Entries in comment trees and in archive
trees will be ignored.
(org-map-entries '(org-todo "UPCOMING") "+TOMORROW" 'file 'archive 'comment)
The following example counts the number of entries with TODO
keyword WAITING, in all agenda files.
(length (org-map-entries t "/+WAITING" nil 'agenda))
k r to start remember from the agenda, with enforcing
the cursor date as default for any time stamps created by
the template.
%& escape anywhere in the template. An
interesting combination now is to use %!%&, which will
immediately file and visit the note, which is equivalent to
generating the note directly in the target location. Thanks
to Adam Spiers for this proposal.
If you are using %%(bbdb-anniversaries) to list
anniversaries in the agenda, you can now directly access the
entry that triggered a listed anniversary from the agenda.
Just click the anniversary - it is a link now. Thanks to
Thomas Baumann for a patch to this effect.
See the new variable
org-agenda-columns-remove-prefix-from-item. Thanks to Adam
Spiers for this proposal.
See the new variable `org-export-section-number-format'. Thanks to Adam Spiers for this proposal.
In column view, if you press a key 1-9 or 0, the corresponding values from the list of allowed values for that field at point will be directly selected. Thanks to Levin Du for this proposal and a patch to this effect.
The new hook `org-before-save-iCalendar-file-hook' runs just before the buffer with a created iCalendar export is saved. This is what I settled for after a long discussion with Adam Spiers about doing some special filtering automatically.
When turning on log mode in the agenda with l, clock lines
will now also list the end time, not only the starting time.
Thanks to Tian Qiu for bringing this up again.
Thanks to Manuel Hermenegildo for a patch to this effect.
The [/] and [%] cookies have already provided statistics for checkboxes. Now they do the same also for TODO entries. If a headline contains either cookie, changing the TODO state of any direct child will trigger an update of this cookie. Children that are neither TODO nor DONE are ignored.
There have already been requests to automatically switch the parent headline to DONE when all children are done. I am not making this a default feature, because one needs to make many decisions about which keyword to use, etc. Instead of a complex customization variable, I am providing a hook that can be used. This hook will be called each time a TODO statistics cookie is updated, with the cursor in the corresponding line. Each function in the hook will receive two arguments, the number of done entries, and the number of not-done entries, and you can use the hook to change the state of the headline. Here is an example implementation:
(defun org-summary-todo (n-done n-not-done) "Switch entry to DONE when all sub-entries are done, to TODO otherwise." (let (org-log-done org-log-states) ; turn off logging (org-todo (if (= n-not-done 0) "DONE" "TODO")))) (add-hook 'org-after-todo-statistics-hook 'org-summary-todo)
If you are writing a document with source code examples, you can
include these examples into a #+BEGIN_SRC lang ... #+END_SRC or
(with the org-mtags module loaded) a <src... structure. lang
stands for the Emacs mode used for editing the language, this
could be emacs-lisp for Emacs Lisp mode examples, or org for
Org mode examples. You can now use the key "C-c '" (that is C-c
followed by the single quote) to edit the example in its native
mode. This works by creating an indirect buffer, narrowing it to
the example and setting the appropriate mode. You need to exit
editing by pressing "C-c '" again. This is important, because
lines that have syntactic meaning in Org will be quoted by
calling this command.
"C-c '" also edits include files, the setupfile in a #+setufile
line, and all those little foreign snippets like:
#+HTML: this code can be edited in html-mode #+BEGIN_HTML Same here #+BEGIN_HTML #+LaTeX: this code can be edited in latex-mode #+BEGIN_LaTeX Same here #+BEGIN_LaTeX #+BEGIN_SRC fortran Here we can edit in fortran-mode #+END_SRC
This is necessary for synchronization services. The UIDs are created using the the org-id.el module which is now part of the Org core. If you set the variable
(setq org-icalendar-store-UID t)
then all created UIDs will be stored in the entry as an :ID:
property. This is off by default because it creates lots of
property drawers even if you only play with iCalendar export.
But if you plan to use synchronization, you really need to turn
this on.
Diary sexp entries do not yet receive proper persistent UIDs, because they are transformed to iCalendar format by icalendar.el which creates fresh UIDs each time, based on the current time.
An interesting aspect of Org is that a single outline node can give rise to multiple iCalendar entries (as a timestamp, a deadline, a scheduled item, and as a TODO item). Therefore, Org adds prefixes "TS-", "DL-" "CS-", and "TD-" to the UID during iCalendar export, depending on what triggered the inclusion of the entry. In this way the UID remains unique, but a synchronization program can still figure out from which entry all the different instances originate.
When exporting a subtree by selecting it before calling the
export command, you can now use the properties EXPORT_TITLE,
EXPORT_TEXT, and EXPORT_OPTIONS to overrule the global
#+TITLE, #+TEXT, and #+OPTIONS settings. You can also set
an export file name with EXPORT_FILE_NAME that will overrule
the file name derived from the buffer's file name. As far as the
options are concerned, the global #+OPTIONS will still be read,
and only the options you give in the property will be
overwritten. For example:
#+OPTIONS: skip:nil * Computer Tricks :PROPERTIES: :EXPORT_FILE_NAME: ct.html :EXPORT_TITLE: Steve's collected computer tricks :EXPORT_OPTIONS: h:2 toc:nil :END:
Tags that are defined in a line like
#+FILETAGS: work urgentare inherited by all entries in the file.
Thanks to Manuel Hermenegildo for this proposal.
Previously, the default was to not include text in an org-mode
buffer before the first headline. From now on, the default it to
include it. If you like the old default better, customize the
variable org-export-skip-text-before-1st-heading or set the
value on a per-file basis with
#+OPTIONS: skip:t
A plain list will be exported as a description list if the first item in the list has a term and the description, separated by " :: ". For example
Emacs software by Carsten Dominik
- RefTeX :: Support for LaTeX Labels, References, Citations
- CDLaTeX :: more LaTeX functionality for Emacs
- TeXmathp :: checking LaTeX buffers for Math mode.
- ORG :: An Emacs mode for notes and projet planning.
- CONSTANTS :: An Emacs package for inserting the definition of
natural constants and units into a buffer.
- IDLWAVE :: The Emacs modes for editing and
running IDL and WAVE CL files.
will be rendered as
Emacs software by Carsten Dominik
This works now in the HTML exporter, we still need to supoort it with the LaTeX and ASCII exporters.
For quoting an entire paragraph as a citation, use
#+BEGIN_QUOTE Everything should be made as simple as possible, but not any simpler -- Albert Einstein #+BEGIN_QUOTE
which will render as
Everything should be made as simple as possible, but not any simpler – Albert Einstein
You can now get code examples fontified like they would be fontified in an Emacs Buffer, and export the result to HTML. To do so, wrap the code examples into the following structure:
#+BEGIN_SRC emacs-lisp (defun org-xor (a b) "Exclusive or." (if a (not b) b)) #+END_SRC
In the export, this will then look like this (if you are now looking at the ASCII export and do not see anything interesting, go and check out the HTML version at http://orgmode.org/Changes.html).
(defun org-xor (a b) "Exclusive or." (if a (not b) b))
The string after the BEGIN_SRC is the name of the major emacs
mode that should be used to fontify the code example, without the
"-mode" at the end of the mode name. For example, if you are
writing an Org tutorial with Org examples included, you would use
"org" as the language identifier - in fact, I have used just
that in the example above.
Currently this works only for HTML export, and requires the htmlize.el package, version 1.34 or later. For other backends, such structures are simply exported as EXAMPLE.
A line like
#+INCLUDE "file" markup lang
will lead to the inclusion of the contents of FILE at the moment of publishing. FILE should be surrounded by double quotes, this is obligatory if it contains space characters. The parameters MARKUP and LANG are optional. MARKUP can be "example", "quote", or "src". If it is "src", LANG should be the name of the Emacs mode to be used for fontifying the code. For example:
Here is my /.emacs/ file: #+INCLUDE "~/.emacs" src emacs-lisp
Previously, the default was to not include text in an org-mode
buffer before the first headline. From now on, the default it to
include it. If you like the old default better, customize the
variable org-export-skip-text-before-1st-heading or set the
value on a per-file basis with
#+OPTIONS: skip:t
If you would like to share the Org setup between a number of files, you can now store in-buffer setup in a file and simply point to that file from each file that should read it. If you write in a buffer
#+SETUPFILE: "path/to/setup.org"
then this file will be scanned for in-buffer options like
#+STARTUP, #+TITLE, or #+OPTIONS.
From now on, it makes no difference is you write #+STARTUP or
#+startup, to make these lines less imposing. Similarly for all
other in-buffer keywords.
As a new experimental feature, Org now supports completion of
structural elements like #+BEGIN_EXAMPLE in a special way.
It work by typing, for example "<e" and then pressing TAB, on
an otherwise empty line. "<e" will expand into a complete
EXAMPLE template, with the cursor positioned in the middle.
Currently supported templates are:
<s #+begin_src <e #+begin_example <q #+begin_quote <v #+begin_verse <l #+begin_latex <L #+latex: <h #+begin_html <H #+html: <a #+begin_ascii <i #+include
This is an experimental feature, please comment! See also below under org-mtags.el.
When Emacs opens an Org mode buffer, the outline visibility
is set to a startup value that is taken from the variable
org-startup-folded, or from a #+STARTUP setting in the
buffer. After this has happened, the buffer will now also be
scanned for entries with a VISIBILITY property. Wherever
such a property is found, the corresponding subtree will get
its visibility adjusted. Allowed values for the property
are:
For example, I am using this for the huge Changes.org file that is the source for the list of visible changes you are reading right now. The top-most entry in this file always describes the changes in my current working version. The start of this section currently looks like this:
* Version 6.03 :PROPERTIES: :VISIBILITY: content :END: ** Overview
This was a proposal by Ben Alexander.
The command C-u C-u TAB will switch back to the startup
visibility of the buffer.
Org now remembers the last 5 tasks that you clocked into, to make it easier to clock back into a task after interrupting it for another task.
C-u C-u C-c C-x C-i (or C-u C-u I from the agenda) will
clock into that task and mark it as current default task.
C-u C-c C-x C-i (or C-u I from the agenda) will offer a
list of recently clocked tasks, including the default task,
for selection. d selects the default task, i selects
the task that was interrupted by the task that is currently
being clocked. 1,… selects a recent task. When you
select a task, you will be clocked into it.
C-u C-c C-x C-j to jump to any of these
tasks.
When moving an entry using structure editing commands,
archiving commands, or the special subtree cut-and-paste
commands C-c C-x C-w and C-c C-x C-y, the running clock
marker and all clock history markers will be moved with the
subtree. Now you can start a clock in a remember buffer and
keep the clock running while filing the note away. See also
the variable `org-remember-clock-out-on-exit'.
bbdb-anniversaries is now much faster, thanks to a new
approach using a hash for birthdays. Thanks to Thomas
Baumann for a patch to this effect.
Do people think any of these should become core?
<lisp> tag
of Emacs Wiki and Muse.
#+BEGIN_EXAMPLE ... #+END_EXAMPLEyou can write
<example> ... </example>In fact, I myself find these easier to type and to look at. Also, it will allow you to more easily move text and files back and forth between Org and Muse. For a list of supported structure elements, see the commentary in the file commentary in the file org-mtags.el.
If you load this module and use the "<i" etc completion described above, the Muse form will automatically be inserted.
Many bug fixes again. Will this ever stop?
I had already given up on this, but Greg Chernev (who implemented noutline.el for XEmacs and in this way kept Org alive on XEmacs) has done it again and provided the patches to make column view work under XEmacs. There are still some problems, but the basics work and we will iron out the remaining issues, hopefully soon.
If any of the columns has a summary type defined, turning on
column view in the agenda will show summaries for these
columns. Org will first visit all relevant agenda files and
make sure that the computations of this property are up to
date. This is also true for the special CLOCKSUM property.
Org will then sum the values displayed in the agenda. In the
daily/weekly agenda, the sums will cover a single day, in all
other views they cover the entire block. It is vital to
realize that the agenda may show the same entry multiple
times (for example as scheduled and as a deadline), and it
may show two entries from the same hierarchy (for example a
parent and it's child). In these cases, the summation in
the agenda will lead to incorrect results because some values
will count double.
If you want to plan your work in a very detailed way, or if
you need to produce offers with quotations of the estimated
work effort, you may want to assign effort estimates to
entries. If you are also clocking your work, you may later
want to compare the planned effort with the actual working
time. Effort estimates can now be stored in a special
property Effort, displayed side-to-side with clock sums,
and also be summed over a day, in order to show the planned
work load of a day. See the manual for more details.
Property searches can now choose a number of different operators for comparing values. These operators are `=', `<>', `<', `<=', `>', and `>='.
When the search term uses the operator with plain number like
+Effort>=2.7, then the property value is converted to a
number and a numerical comparison takes place.
When the search term uses a string on the right hand side of
the operator, a string comparison is done: +PRIORITY<"C".
Finally, if the right hand side is enclosed in curly braces,
a regexp match is done: aaa={regexp}. In this case you
should use only the `=' or `<>' operators, meaning "does
match" or "does not match", respectively.
This was a triggered with a request by Dan Davison.
If the value of the customization variable
org-agenda-text-search-extra-files contains the symbol
agenda-archives as the first element in the list, all
archive files of all agenda files will be added to the list
of files to search. This is relevant for the search view
C-c a s, as well as for the agenda files multi-occur
command C-c a /.
There are new values for the :scope parameter of a clock
table. This can now be file-with-archives and
agenda-with-archives, in order to collect information not
only from the current file or all agenda files, but also from
all archive files that are currently used by these files.
The options available for radio tables using orgtbl-mode have been expanded. You may use several reception points and formats for the same table, you may have special formatting in the last line of the table, and many table parameters may be functions, so that more general transformations are possible. Jason Riedy provided a patch for this, and he will hopefully come up with some examples. Thanks!
This is a new major release, mostly because of structural changes in Org. However, since this took a while, there is also a long list of small improvements and some new significant features.
In the distribution files as well as in the GIT repository, the lisp files are now located in a subdirectory "lisp", and the documentation files are located in a subdirectory "doc". If you are running Org directly from the unpacked distribution archive (zip or tar file, or GIT repository), you need to modify your settings for load-path accordingly.
In the distribution files as well as in the GIT repository, the lisp files are now located in a subdirectory "lisp", and the documentation files are located in a subdirectory "doc". If you are running Org directly from the unpacked distribution archive (zip or tar file, or GIT repository), you need to modify your settings for load-path accordingly.
Org-mode has now a system for loading modules by simply configuring an option that lists all the modules you want to use. Customize the variable `org-modules'. That variable lists both modules that are part of the Org-mode core (and in this way part of Emacs), and modules that are contributed packages. Contributed modules will only be available when you have installed them properly (most likely by downloading the distribution and adding /path/to/orgdir/contrib/lisp to your load path).
There is a new method to archive entries in the current file: By moving it to a sibling called the Archive Sibling. That sibling has the heading "Archive" and also carries the ARCHIVE tag. This can be a great way to do archiving inside a project, to get parts of the project out of the way and to wait with true archiving (moving to another file) until the entire project is done. Archiving to a sibling keeps much of the context, for example inherited tags and approximate tree position in tact.
The key binding for the is "C-c C-x A", and from the agenda buffer you can simply use "A".
Thanks to Ilya Shlyakhter for this rather clever idea.
This fascinating program allows a completely new viewing experience for web pages created from Org files. The same document can be viewed in different ways, and switching between the views as well as navigation uses single-key commands.
One of the view types is an Info-like interface where you can jump through the sections of the document with the `n' and `p' keys (and others). There is also a folding interface where you can fold the document much like you can fold it in org-mode in Emacs, and cycle through the visibility both locally and globally.
To set this up, all you need to do is to make sure that org-infojs.el gets loaded (customize the variable org-modules to check). Then add this line to the buffer:
#+INFOJS_OPT: view:info
In that line, you can configure the initial view and other
settings. Available views are info for the info-like
interface, and overview, content, and showall for the
folding interface. See the manual for more details. The
JavaScript program is served from
http://orgmode.org/org-info.js, and your exported HTML files
will automatically get it from there. However, you may want
to be independent of the existence and stability of
orgmode.org and install a copy locally. Then you need to
change the path from which the script is loaded, either by
using something like
#+INFOJS_OPT: view:info path:../scripts/org-info.js
or by configuring the variable org-infojs-options.
For details see the documentation provided by Sebastian Rose together with org-info.js.
Both these are based on requests by Ilya Shlyakhter.
The :block parameter to the table can now look like any
of these:
| :block | meaning |
|---|---|
| 2008 | The entire year 2008 |
| 2008-04 | The month April 2008 |
| 2008-04-02 | The day April 2, 2008 |
| 2008-W14 | ISO-Week 14 in 2008 |
| today | Today |
| today-5 | The day five days ago |
| thisweek | The current week |
| thisweek-2 | Two weeks ago |
| thismonth | The current month |
| thismonth-12 | Same month, last year |
| lastmonth | Same as thismonth-1 |
What is more, you can now use the S-left and S-right
keys to shift the time block around. The cursor needs to
be in the #+BEGIN: clocktable line for this to work. If
the current block is today, S-left with switch to
yesterday. If the current block is 2008-W14, S-right
will switch to the following week.
org-agenda-start-with-clock report-mode. To modify the
properties of the table, in particular the :maxlevel
depth, configure org-agenda-clockreport-parameter-plist.
The agenda now shows the ISO week for the displayed dates, in
the form W08 for week 8.
The keys d, w, m, and y in the agenda view now accept
prefix arguments. Remember that in the agenda, you can
directly type a prefix argument by typing a number, no need
to press C-u first. The prefix argument may be used to
jump directly to a specific day of the year, ISO week, month,
or year, respectively. For example, 32 d jumps to February
1st, 9 w to ISO week number 9. When setting day, week, or
month view, a year may be encoded in the prefix argument as
well. For example, 200712 w will jump to week 12 in the
year 2007. If such a year specification has only one or two
digits, it will be mapped to the interval 1938-2037.
When entering a date at the date prompt, you may now also specify an ISO week. For example
w4 Monday of week 4 fri w4 Friday of week 4 w4-5 Same as above 2012 w4 fri Friday of week 4 in 2012. 2012-W04-5 Same as above
So far I have not implemented the effect of `org-read-date-prefer-future' on this functionality, because it seemed too magic for me. I'd appreciate comments on this issue: Should `org-read-date-prefer-future' also push dates into the next year if the week you are entering has already passed in the current year? For consistency I guess this should be the case, but I cannot quite wrap my head around it.
I hope but am not entirely convinced that this will behave sanely also during the first/last week of a year. Please test extensively and report back.
This was a request by Thomas Baumann.
org-use-tag-inheritance. This
variable may now be a regular expression or a list to
select the inherited tags. Thanks to Michael Ekstrand for
this excellent proposal.
The regexp option is also implemented for
org-use-property-inheritance, so that you can now select
properties for inheritance my name.
org-entry-get can be set
to the symbol selective. If this is the case, then the
value of the property will be retrieved using inheritance
if and only if the setting in
org-use-property-inheritance selects the property for
inheritance.
org-agenda-date and
org-agenda-date-weekend. Both these faces are initially
similar to the org-agenda-structure face, but you can
customize them freely.
This was an omission in the earlier implementation, spotted by Wanrong Lin. Thanks!
org-directory.
%c - Now always insert the head of the kill ring, never
the X clipboard.
%x - Insert the content of the X clipboard. This is the
first non-empty value from the PRIMARY, SECONDARY and
CLIPBOARD X clipboards.
%^C - This allows the user to choose between any of the
clipboard values available, the kill ring head, and the
initial region if set.
%^L - Like %^C, but this inserts an org link using the
selected value.
Thanks to James TD Smith for this patch.
org-table-export-default-format. You can use properties
TABLE_EXPORT_FILE and TABLE_EXPORT_FORMAT to specify the
file name to which the export should go, and a local
format. For example:
:PROPERTIES: :TABLE_EXPORT_FILE: ~/xx.txt :TABLE_EXPORT_FORMAT: orgtbl-to-generic :splice t :sep "\t" :END:
Thanks to James TD Smith for this patch.
org-todo-keywords, or in the #+TODO line. Use
the "o" key when sorting with C-c ^.
Thanks to James TD Smith for this patch.
M-RET splits lines again
`C-c a s' now invokes a special agenda view that can be used to search notes by keyword and regular expressions. In particular, it does not require a single regular expression or string to search for, but it can search for a number keywords or regexps that can occur in arbitrary sequence in the entry. The search knows the boundaries of an entry, can use simple Boolean logic and is reasonably fast. For example, the search string
+computer +wifi -ethernet -{8\.11[bg]}
will search for note entries that contain the keywords
computer and wifi, but not the keyword ethernet, and
which are also not matched by the regular expression
"8\.11[bg]", meaning to exclude both 8.11b and 8.11g. If the
first character of the search string is an asterisk, the
search will only look at headlines - otherwise it will look
at the headine and the text below it, up to the next
(possibly sub-) heading.
The command searches all agenda files, and in addition the
files listed in org-agenda-text-search-extra-files.
I find it very useful to define a custom command to do such a search only in a limited number of files (my notes files), like this:
("N" "Search notes" search ""
((org-agenda-files '("~/org/notes.org" "~/org/computer.org"))
(org-agenda-text-search-extra-files nil)))
t or nil. If it is a list of major-mode, the
template will be available only when org-remember is
called from a buffer in one of these modes. If it is a
function, the template will be offered only if the function
returns `t' when called in the current buffer. A value of
t or nil for this element means select this template in
any context.
One possible application for this would be to have several templates all using the same selection letter, and choosing the right one based on context. For example, think of tasks describing a bug in a source code file. With the following configuration we make sure that the bug reports are filed into the appropriate sections of the target file.
(setq org-remember-templates
'(("Elisp" ?b "* %a\n\n%i%?" "~/bugs.org" "Elisp bugs" (emacs-lisp-mode))
("C Bugs" ?b "* %a\n\n%i%?" "~/bugs.org" "C bugs" (cc-mode))))
See (info "(org)Remember templates") for details.
Based on proposals by Bastien.
Based on proposals by Bastien.
org-time-stamp-rounding-minutes is now
a list of two values. The first applies when creating a new
time stamp. The second applies when modifying a timestamp
with S-up/down. The default for this new task is 5 minutes,
but 15 may also be a very good value for many people. If
S-up/down is used on a time stamp where the minute part is
not compatible with this granularity it will be made so.
You can bypass this by using a prefix argument to exactly
specify the number of minutes to shift.
This was a proposal by Adam Spiers.
** TODO Call Father DEADLINE: <2008-02-10 Sun ++1w> Marking this DONE will shift the date by at least one week, but also by as many weeks as it takes to get this date into the future. However, it stays on a Sunday, even if you called and marked it done on Saturday. ** TODO Check the batteries in the smoke detectors DEADLINE: <2005-11-01 Tue .+1m> Marking this DONE will shift the date to one month after today.
Proposed by Wanrong Lin and Rainer Stengle.
:link parameter for the clocktable. When
set, the headlines listed in the table will be links to the
original headlines.
:content parameter that is passed to the
writer function of the dynamic block. Use this parameter
to pass the previous content of the block to the writer
function, in case you want to make the outcome dependent on
the previous content.
This was a request by Kyle Sexton.
M-RET can again be used to split a line so that the rest
of the line becomes the new heading. However, if you do
this in a heading containing tags, the tags will stay in the
old line.
Customize the variable org-M-RET-may-split-line if you
don't want this command to split a line in the middle. The
same variable also influences line splitting in items and in
tables.
org-follow-link-hook: runs after following a link
org-publish-before-export-hook: runs before export
org-publish-after-export-hook: runs after export
There is now more control over which state changes are being logged in what way. Please read carefully the corresponding sections in the manual. Basically:
WAIT(w@) Record a note when entering this state.
WAIT(w!) Record a timestamp when entering this state.
WAIT(w@/!) Recore a note when entering and timestamp
when leaving this state. This is great for
getting a record when switching *back* from
WAIT to TODO.
WAIT(/!) Record a timestamp when leaving this state.
Here we not even define a fast access
character, but just the logging stuff.
This was triggered by requests from Wanrong Lin and Bernt Hansen.
Bug fixes, in particular the long-hunted bug about wrong window positions after pressing SPACE in the agenda. Hopefully this is really fixed.
:step for clocktable, to get daily reports.
This is following a proposal by Piotr Zielinski.
:step. This
parameter can be `day' or `week' and will result in separate
tables for each day or week in the requested time interval.
This was triggered by a proposal by Sacha Chua in her blog.
ARCHIVE_OLPATH. This property contains the "path" in the
outline tree to the archived entry, as it was in the
original file. For example, archiving Fix the door in the
following hierarchy
* Tasks ** HOME *** Garage **** Fix the door
will file is with the following property
:ARCHIVE_PATH: Task/HOME/Garage
Note that you can configure (i.e. limit) the information that gets stored upon archiving with the variable `org-archive-save-context-info'.
#+COLUMNS: %20ITEM %10Time_Estimate{:} %CLOCKSUM
will allow you to compare estimated times (as given in the Time_Estimate property) with the clocked times. This was a request by Bernt Hansen.
*Remember* buffer. But if you
have chosen (e.g. by pressing `C-u C-c C-c') to
interactively select the filing location (file and
headline), the refile interface will be used instead. I am
excited about this change, because the `org-goto' interface
is basically a failure, at least for this application. Note
that in any case the refile interface has to be configured
first by customizing `org-refile-targets'.
Minor fixes.
I believe we have finally nailed this one. Thanks to Daniel Pittman for bring this up again and to Eric Schulte for pointing out that it is the empty lines before an entry that really count.
This change was non-trivial, please give it a good test and let me know about any problems.
%^{Author|Roald Dahl|Thomas Mann|Larry Niven}
will prompt for an author name. Pressing RET without typing anything will select "Roald Dahl". Completion will give you any of the three names. And a history will be kept, so you can use the arrow keys to get to previous input. The history is tied to the prompt. By using the same prompt in different templates, you can build a history across templates. The ideas for this came from proposals by Bastien and Adam.
You can now permanently lock the agenda construction to a certain scope, like a file or a subtree. So instead of pressing "<" for each command in the agenda dispatcher, you only once select a restriction scope. All subsequent agenda commands will than respect this restriction. For example, you can use this at work, to limit agendas to your work file or tree, and at home to limit to the home file or tree. Or you can use it during the day in order to focus in on certain projects.
You select a scope with the command `C-c C-x <', which restricts to the current subtree. When called with a `C-u' prefix, the restriction is to the current file. You can also make restrictions from the speedbar frame, see below.
When making a new restriction and an agenda window is currently visible, it will immediately be updated to reflect the new scope. If you like you can display an agenda view and then watch it change in various scopes.
To get rid of the restriction, use the command "C-c C-x >". Or press ">" in the agenda dispatcher. Also, and use of "<" in the dispatcher will disable the restriction lock and select a new restriction.
Thanks to Rick Moynihan for triggering this development.
(add-hook 'org-mode-hook
(lambda () 'imenu-add-to-menubar "Imenu"))
a menu will be created in each Org-mode buffer that provides access to all level 1 and level 2 headings. The depth of the menu can be set with the variable `org-imenu-depth'.
*a*.
(defun my-remember-template-n ()
(interactive)
(org-remember ?n))
You need to modify this. The character selecting the template must now be the second argument to `org-remember':
(defun my-remember-template-n ()
(interactive)
(org-remember nil ?n))
(setq org-refile-targets '((nil . (:level . 2))))
selects all level 2 headlines in the current buffer as targets. And
(setq org-refile-targets
'((org-agenda-files . (:tag . "refile"))))
searches all agenda files and selects headlines that are explicitly marked with the tag :refile: . Note that the list of targets is built upon first use only, to rebuilt it, call the command `C-c C-w' with a double prefix argument.
This is based on an idea and example implementation by Max Mikhanosha. Many thanks Max.
This was proposed by someone, but I have lost track who. Sorry, and thanks anyway.
%c insert the current clipboard, like C-y would do %(..) evaluate Lisp expression and insert the result %[..] include file
Thanks to Adam Spiers and Tim O'Callaghan.
(require 'remember) (setq remember-annotation-functions '(org-remember-annotation)) (setq remember-handler-functions '(org-remember-handler)) (add-hook 'remember-mode-hook 'org-remember-apply-template))
You might still want to set `org-default-notes-file' to provide a default for templates without a file, and `org-directory' to show where to find other org files.
*/bold italic/* is no longer allowed.
*a*.
=quoted *xxx* a_x = b=. This and the following
point implement a request by Daniel Clemente.
There have been several small but very useful additions to the date prompt.
| Org | Description | HTML |
|---|---|---|
\\- | double backslash followed by minus | ­ |
-- | two dashes (minuses) | – |
--- | three dashes (minuses) | — |
... | three dots | … |
You can turn this globally on or off with `org-export-with-special-strings' or locally with "-:t" or "-:nil" in the #+OPTIONS line. Thanks to Adam Spiers for starting the discussion, and thanks to Daniel Clemente and William Henney for relevant inputs.
g does now the same a "r", refresh current display,
because "g" is the Emacs standard for "refresh"
G toggle the time grid, used to be "g"
e Execute another agenda command, pretty much the same as
`C-c a', but shorter and keep the same agenda window.
You can identify the entry whose column view you want to capture by assigning an :ID: property, and use that property in the dynamic block definition. For example:
* Planning
:PROPERTIES:
:ID: planning-overview
:END:
[...]
* The column view
#+BEGIN: columnview :hlines 1 :id "planning-overview"
#+END:
Use `C-c C-x r' to insert such a dynamic block, and you will be prompted for the ID.
(setq org-agenda-custom-commands
'(("h" . "HOME + Name tag searches") ; describe prefix "h"
("hl" tags "+HOME+Lisa")
("hp" tags "+HOME+Peter")
("hk" tags "+HOME+Kim")))
We are using a new routine for sorting entries, courtesy of John Wiegley. Many thanks to John.
There is now a <div>-based structure in exported HTML.
I am not sure if the class names are the best choice, let me know if there are more "canonical" choices.
Thanks to Mike Newman and Cezar for input, and in particular to Mike for his clearly formulated specification.
#+DRAWERS: HIDDEN STATE PROPERTIES
This will define the drawers :HIDDEN: and :STATE:. The :PROPERTY: drawer should always be part of this list, or your properties will not be folded away. Thanks to Richard G. Riley for this proposal.