org-mode/ORGWEBPAGE/Changes.org

Org-mode list of user-visible changes

Version 6.25

Major new features

DocBook export

We now do have a fully functional DocBook exporter, contributed by Baoqiu Cui. Simple press C-c e D to export the current file to DocBook format. You can also get direct conversion to PDF if you have made the correct setup, please see the manual for details.

Kudos to Baoqiu for this fantastic addition, and my personal thanks for doing this in a such a smooth way that I did not have to do anything myself.

Protocols for external access to Emacs and Org

org-protocol.el is a new module that supersedes both org-annotation-helper.el and org-browser.el and replaces them with a more abstracted interface. org-protocol intercepts calls from emacsclient to trigger custom actions without external dependencies. Only one protocol has to be configured with your external applications or the operating system, to trigger an arbitrary number of custom actions. Just register your custom sub-protocol and handler with the new variable org-protocol-protocol-alist.

org-protocol comes the with three standard protocol handlers (in parenthesis the name of the sub-protocol):

org-protocol-remember (remember)

Trigger remember

org-protocol-store-link (store-link)

Store a link

org-protocol-open-source (open-source)

Find the local source of a remote web page.

Passing data to emacs is now as easy as calling

emacsclient org-protocol://sub-protocol://data

For more information see the online documentation

Thanks to Sebastian Rose for this really beautiful module.

Inline tasks

Inline tasks are tasks that have all the properties of normal outline nodes, including the ability to store meta data like scheduling dates, TODO state, tags and properties. But these tasks are not meant to introduce additional outline structure, at least as far as visibility cycling and export is concerned. They are useful for adding tasks in extensive pieces of text where interruption of the flow or restructuring is unwanted.

This feature is not turned on by default, you need to configure org-modules to turn it on, or simply add to you .emacs file:

(require 'org-inlinetask)

After that, tasks with level 15 (30 stars when using org-odd-levels-only) will be treated as inline tasks, and fontification will make obvious which tasks are treated in this way.

Input from RSS feeds

Org can now collect tasks from an RSS feed, a great method to get stuff from online call and note-taking services into your trusted system. You need to configure the feeds in the variable org-feed-alist. The manual contains a short description, more detailed information is available on Worg.

Full credit goes to Brad Bozarth who really paved the way for this exciting new feature.

Export

Allow modification of table attributes in HTML export

The #+ATTR_HTML line can now be used to set attributes for a table. Attributes listed in that line will replace existing attributes in org-export-html-table-tag, or will add new ones. For example

#+ATTR_HTML: border="2" rules="all" frame="all"
#+CAPTION: Finally a table with lines!
| a | b |
|---|---|
| 1 | 2 |

LaTeX low levels are now exported as itemize lists

LaTeX export now treats hierarchy levels 4,5, etc as itemize lists, not as description lists as before. This is more consistent with the behavior of HTML export. You can configure this behavior using the variable org-export-latex-low-levels.

Markup for centering.

Text can be exported centered with

#+BEGIN_CENTER
,Everything should be made as simple as possible, \\
,but not any simpler
#+END_CENTER

Sitemap file is now sitemap.org

Org-publish can produce a list of all files in a project. Previously the file containing this list was called "index.org", really a brain-dead default because during publication it would overwrite the "index.html" file of the website.

The default file name is now "sitemap.org"

Protect explicit target links in HTML export

If a link is [[#name] [desc]], the href produced when exporting the file will be exactly href="#name". So starting a link target with # will indicate that there will be an explicit target for this.

HTML export: Allow "- _" to explicitly terminate a list

If a list contains "- _" (three underscores) as an item, this terminates the list, ignoring this item. This is an experimental feature, it may disappear again if we find other ways to deal with literal examples right after lists.

See this mailing list thread for context.

Agenda

Changing the time of an entry from the agenda

We now have a way to change not only the date, but also the start time of an entry from the agenda. The date is normally changed with S-right/left. Now, if you add a C-u prefix, the hour will be changed. If you immediately press S-right/left again, hours will continue to be changed. A double prefix will do the same for minutes. If the entry has a time range like 14:40-16:00, then both times will change, preserving the length of the appointment.

Show saved PDF agenda view with prefix arg

When writing an agenda view to a PDF file, supplying a a prefix argument (C-u C-x C-w) will get the new file displayed immediately.

This was a request by Alan E Davis.

Filter for entries with no effort defined

During secondary agenda filtering, pressing "?" now will install a filter that selects entries which do not have an effort defined.

This new model was necessary because we needed to stop interpreting entries with no effort defines as 0 effort. This was inconsistent, because for normal agenda sorting, the treatment of these entries depends on the variable org-sort-agenda-noeffort-is-high. Now this variable is also respected during filtering.

This new feature resulted from a discussion with Matt Lundin and Bernt Hansen.

Introduce user-defined sorting operators

The new variable org-agenda-cmp-user-defined can contain a function to test how two entries should be compared during sorting. The symbols user-defined-up and user-defined-down can then be part of any sorting strategy.

This was a request by Samuel Wales.

Indentation of subitems in the agenda

When a tags/property match does match an entry and it's sublevels, the sublevels used to be indented by dots, to indicate that the matches likely result from tag inheritance. This is now no longer the default, so the subitems will not get special indentation. You can get this behavior back with

(setq org-tags-match-list-sublevels 'indented)

Stuck projects search now searches subtrees of unstuck projects

When, during a stuck-project search, a project tree is identified as not stuck, so far the search would continue after the end of the project tree. From now on, the search continues in the subtree, so that stuck subprojects can still be identified.

Miscellaneous

Citations: Use RefTeX to insert citations

RefTeX can now be used to create a citation in Org-mode buffers. Setup the buffer with

#+BIBLIOGRAPHY: bibbase style

and create citations with C-c C-x [.

Together with org-exp-bibtex.el by Taru Karttunen (available as a contributed package), this provides a great environment for including citations into HTML and LaTeX documents.

Changing time ranges as a block

When using the S-cursor keys to change the first time in a time range like

<2009-04-01 Wed 14:40-16:40>

then the end time will change along, so that the duration of the event will stay the same.

This was a request by Anupam Sengupta.

New sparse tree command

A new sparse tree command shows entries with times after a certain date. Keys are C-c / a, this command is for symmetry with C-c / b.

Cloning tasks

A new command allows to create clone copies of the current entry, with shifted dates in all stamps in the entry. This is useful to create, for example, a series of entries for a limited time period. I am using it to prepare lectures, for example.

New face for checkboxes

Checkboxes now have their own face, org-checkbox. This can be used for nice effects, for example choosing a face with a box around it:

(custom-set-faces
 (org-checkbox ((t (:background "#444444" :foreground "white"
                :box (:line-width 1 :style released-button)))))

M-a and M-e for navigation in a table field

In tables fields, the sentence commands M-a and M-e are redefined to jump to the beginning or end of the field.

This was a request by Bastien Guerry.

Backup files for remember buffers

Sometimes users report that they lost data when not immediately storing a new remember note, and then later exiting Emacs or starting a new remember process.

Now you can set the variable org-remember-backup-directory. Each remember buffer created will then get its own unique file name in that directory, and the file will be removed only if the storing of the note to an Org files was successful.

org-mac-message.el: New functions to access flagged mail

Christopher Suckling has added functionality to org-mac-message.el. In particular, you can now select a number of messages and easily get links to all of them with a single command. For details, see the online documentation.

Read-date: New hook

The new hook org-read-date-minibuffer-setup-hook is called when setting up the minibuffer for reading a date. If can be used to install new keys into the temporary keymap used there.

Version 6.24

Incompatible changes

Tag searches are now case-sensitive

From this release on, tag searches will be case sensitive. While I still think it would be nice to have them case-insensitive, this was both an inconsistency (TODO keyword searches have always been case-sensitive), and trouble for coding some efficient algorithms. So please make sure that you give the tags with correct casing when prompted for a match expression.

New key for creating tags/property sparse trees

The key to produce a sparse tree matching tags and properties is now C-c / m instead of C-c a T. This is also more consistent with the C-c a m key for the corresponding agenda view. C-c / T will still work for now, but it is no longer advertised in the documentation and may go away at any time in the future.

IDs in HTML have "ID-" prefix when generated by uuidgen

uuidgen generates IDs that often start with a number, not a latter. However, IDs and names in XHTML must start with a letter. Therefore, IDs in HTML files will now get an "ID-" prefix if they have been generated by uuidgen. This means that id links from one file to another may stop working until all files have been exported again.

In agenda, only priority cookies get the special face

So far, an entire task would get a special face when org-agenda-fontify-priorities was set. Now, the default value for this variable is the symbol cookies, which means that on the cookie is fontified. Set it to t if you want the entire task headline to be fontified.

Details

PDF export of agenda views

Agenda views can now be exported to PDF files by writing them to a file with extension ".pdf". Internally this works by first producing the postscript version and then converting that to PDF using the ghostview utility ps2pdf. Make sure that this utility is installed on your system.

The postscript version will not be removed, it will stay around.

Inline some entry text for Agenda View export

When exporting an agenda view to HTML or PDF for printing or remote access, one of the problems can be that information stored in entries below the headline is not accessible in that format.

You can now copy some of that information to the agenda view before exporting it. For this you need to set the variable org-agenda-add-entry-text-maxlines to a number greater than 0.

(setq org-agenda-add-entry-text-maxlines 20)

Or you can do this with the settings in a custom agenda view, for example

("A" "" agenda ""
 ((org-agenda-ndays 1)
  (org-agenda-add-entry-text-maxlines 5))
 ("agenda-today.pdf"))

Improved ASCII export of links

ASCII export of links works now much better. If a link has a link and a description part which are different, then the description will remain in the text while the link part will be moved to the end of the current section, before the next heading, as a footnote-like construct.

Configure the variable org-export-ascii-links-to-notes if you prefer the links to be shown in the text. In this case, Org will make an attempt to wrap the line which may have become significantly longer by showing the link.

Thanks to Samuel Wales for pointing out the bad state of ASCII link export.

Custom agenda commands can specify a filter preset

If a custom agenda command specifies a value for org-agenda-filter-preset in its options, the initial view of the agenda will be filterd by the specified tags. Applying a filter with / will then always add to that preset filter, clearing the filter with / / will set it back to the preset. Here is an example of a custom agenda view that will display the agenda, but hide all entries with tags FLUFF or BLUFF:

("A" "" agenda ""
 ((org-agenda-filter-preset '("-FLUFF" "-BLUFF"))))

This is in response to a thread on the mailing list, started by Daniel Clemente and with great contributions by Bernt Hansen and Matt Lundin.

Exporting of citations to LaTeX and HTML, using BibTeX

Citations can now me made using BibTeX, and will be exported to LaTeX and HTML. This is implemented in a contributed package by Taru Karttunen, org-exp-bibtex.el. Kudos to Taru for this really nice addition.

Finally a way to specify keywords and description for HTML export

Use something like

#+DESCRIPTION: This page is all about ....
#+KEYWORDS: org-mode, indexing, publishing

To specify the content of the description and keywords meta tags for HTML output.

org-collector.el is now a contributed package

org-collector.el provides functions to create tables by collecting and processing properties from entries in a specific scope like the current tree or file, or even from all agenda files. General lisp expressions can be used to manipulate the property values before they are inserted into an org-mode table, for example as a dynamic block that is easy to update.

Thanks to Eric Schulte for yet another great contribution to Org.

Update of org2rem.el

org2rem.el has been updated significantly and now does a more comprehensive job of exporting Org events to remind.

Thanks to Sharad Pratap for this update.

New div around the entire page in HTMP export

A new <div id=content> is wrapped around the entire page, everything that is inside <body>.

This means that you need to update org-info.js (if you have a local copy). It will be safe todo so, because the new org-info.js still handles older pages correctly. Thanks to Sebastian Rose for making these changes so quicky.

Clustering characters for undo

When typing in Org-mode, undo will now remove up to 20 characters at a time with a single undo command. This is how things work normally in Emacs, but the special binding of characters in Org-mode made this impossible until now.

Thanks to Martin Pohlack for a patch which mimicks the behavior of the Emacs command loop for the Org version of self-insert-command. Note that this will not work in headlines and tables because typing there will do a lot of extra work.

There might be a small typing performance hit resulting from this change - please report in the mailing list if this is noticeable and annoying.

Separate settings for special C-a and C-e

The variable `org-special-ctrl-a/e' now allows separate settings for C-a and C-e. For example

(setq org-special-ctrl-a/e '(reversed . t))

Thanks to Alan Davis for this proposal.

orgstruct++-mode improvements

In addition to orgstruct-mode which allows to use some Org-mode structure commands in other major modes, there is a more invasive version of this mode: orgstruct++-mode. This mode will import all paragraph and line wrapping variables into the major mode, so that, for example, during typing the auto-fill wrapping of items will work just like in Org-mode. This change is not reversible, so turning off orgstruct++-mode will not remove these settings again. orgstruct++-mode is most useful in text modes like message-mode or magit-log-edit-mode. Furthermore, orgstruct++-mode will recognize plain list context not only in the first line of an item, but also further down, so that M-RET will correctly insert new items.

Thanks to Austin Frank for requesting some of these changes.

Promotion and demotion works for regions now

M-right and M-left now do demote and promote all headlines in an active region.

Match syntax for tags/properties is now described in a single place

The manual chapters about tags and about properties now only refer to the section about agenda views, where the general syntax of tag/property matches is described.

Macro replacement

A string like {{{ title }}} will be replaced by the title of the document, {{{ email }}} by the email setting of the author and similarly for other export settings given in #+... lines. In addition to that, you can define an arbitrary number of macros, for example:

#+MACRO: myaddress 41 Onestreet, 12345 New York, NY
...
,my address is {{{myaddress}}}, see you there.

Macro replacement is the very first thing that happens during export, and macros will be replaced even in source code and other protected regions.

New reload command, with keyboard access

There is now a special command to reload all Org Lisp files, so that you can stay in your Emacs session while pulling and compiling changes to Org. The command to reload the compiled files (if available) is C-c C-x r. If no compiled files are found, uncompiled ones will be loaded. If you want to force loading of uncompiled code (great for producing backtraces), use a prefix arg: C-u C-c C-x r. Both commands are available in the menu as well.

This new command was inspired by one written earlier by Bernt Hansen.

Faces for priority cookies can now be set freely

The new variable org-priority-faces can be used to set faces for each priority.

New key for creating tags/property sparse trees

The key to produce a sparse tree matching tags and properties is now C-c / m instead of C-c a T. This is more consistent with the C-c a m key for the corresponding agenda view. C-c / T will still work for now, but it is no longer advertised in the documentation and may go away at any time in the future.

IDs in HTML have "ID-" prefix when generated by uuidgen

uuidgen generates IDs that often start with a number, not a letter. However, IDs and names in XHTML must start with a letter. Therefore, IDs in HTML files will now get an "ID-" prefix if they have been generated by uuidgen. This means that id links from one file to another may stop working until all files have been exported again, so that both links and targets have the new prefix.

In agenda, only priority cookies get the special face

So far, an entire task would get a special face when org-agenda-fontify-priorities was set. Now, the default value for this variable is the symbol cookies, which means that on the cookie is fontified. Set it to t if you want the entire task headline to be fontified.

Turning off time-of-day search in headline

Some people like to put a creation time stamp into a headline and then get confused if the time-of-day found in there shows up as the time-of-day of the deadline/scheduling entry for this headline. The reason for this is that Org searches the headline for a free-format time when trying to sort the entry into the agenda, and that search accidentally finds the time in the creation time stamp or something else that happens to look like a time. If this is more painful than useful for you, configure the new variable org-agenda-search-headline-for-time.

Version 6.23

Overview

• Capture state change notes into a drawer
• Clock lines are now captured into the LOGBOOK drawer as well
• Added org-R.el to contrib directory
• Allow individual formatting of each TODO keyword in HTML export
• New hooks for add-ons to tap into context-sensitive commands
• Publishing files irrespective of extension
• New variable index in the manual
• The ORDERED property also influences checkboxes
• The ORDERED property can be tracked with a tag
• You may now specify line breaks in the fast tags interface
• When a TODO is blocked by checkboxes, keep it visible in agenda
• LaTeX can import Org's in-buffer definitions for TITLE, EMAIL etc.

Incompatible changes

• CLOCK lines will now be captured into the LOGBOOK drawer. See below for details.

Details

Capture state change notes into a drawer

State change notes can now be captured into a drawer LOGBOOK, to keep the entry tidy. If this is what you want, you will need this configuration:

(setq org-log-into-drawer "LOGBOOK")

Thanks to Wanrong Lin for this proposal.

Clock lines are now captured into the LOGBOOK drawer as well

The CLOCK drawer will be abandoned, clock lines will now also end up in a drawer LOGBOOK. The reason for this is that it's a bit useless to have two different drawers for state change notes and clock lines. If you wish to keep the old way, use

(setq org-clock-into-drawer "CLOCK")

Added org-R.el to contrib directory

Dan Davison has contributed org-R.el which is now in the contrib directory. Org-R performs numerical computations and generates graphics. Data can come from org tables, or from csv files; numerical output can be stored in the org buffer as org tables, and links are created to files containing graphical output. Although, behind the scenes, it uses R, you do not need to know anything about R. Common operations, such as tabulating discrete values in a column of an org table, are available "off the shelf" by specifying options on lines starting with #+R:. However, you can also provide raw R code to be evaluated. The documentation is currently the worg tutorial at http://orgmode.org/worg/org-tutorials/org-R/org-R.php

Thanks to Dan for this great contribution.

Allow individual formatting of TODO keyword and tags in HTML export

TODO keywords in HTML export have the CSS class todo or done. Tags have the CSS class tag. In addition to this, each keyword has now itself as class, so you could do this in your CSS file:

.todo { font-weight:bold; }
.done { font-weight:bold; }
.TODO { color:red; }
.WAITING { color:orange; }
.DONE { color:green; }

If any of your keywords causes conflicts with CSS classes used for different purposes (for example a tag "title" would cause a conflict with the class used for formatting the document title), then you can use the variables org-export-html-tag-class-prefix and org-export-html-todo-kwd-class-prefix to define prefixes for the class names for keywords, for example "kwd-".

Thanks to Wanrong Lin for this request, and to Sebastian Rose for help with the implementation.

New hooks for add-ons to tap into context-sensitive commands

Some commands in Org are context-sensitive, they will execute different functions depending on context. The most important example is of course C-c C-c, but also the M-cursor keys fall into this category.

Org has now a system of hooks that can be used by add-on packages to install their own functionality into these keys. See the docstring of org-ctrl-c-ctrl-c-hook for details. The other hooks are named like org-metaleft-hook or org-shiftmetaright-hook.

Publishing files irrespective of extension

If you set the :base-extension property for a publishing project to the symbol any, all files in the directory will be published, irrespective of extension.

Thanks to Richard Klinda for a patch to this effect.

New variable index in the manual

A new index in the manual lists all variables mentioned in the manual, about 200 variables in total.

The ORDERED property also influences checkboxes

When an entry has the ORDERED property set, checkboxes in the entry must be completed in order. This was already the case for children TODO items, now it also applies for checkboxes.

Thanks to Rainer Stengele for this proposal.

The ORDERED property can be tracked with a tag

The ORDERED property is used to flag an entry so that subtasks (both children TODO items and checkboxes) must be completed in order. This property is most easily toggled with the command C-c C-x o. A property was chosen for this functionality, because this should be a behavior local to the current task, not inherited like tags. However, properties are normally invisible. If you would like visual feedback on the state of this property, configure the variable org-track-ordered-property-with-tag. If you then use C-c C-x o to toggle the property, a tag will be toggled as well, for visual feedback.

Note that the tag itself has no meaning for the behavior of TODO items and checkboxes, and that changing the tag with the usual tag commands will not influence the property and therefore the behavior of TODO and checkbox commands.

You may now specify line breaks in the fast tags interface

Up to now, the fast tags interface tried to lump as many tags as possible into a single line, with the exception that groups would always be on a line by themselves.

Now, if you use several lines to define your tags, like

#+TAGS: aa(a) bb(b) cc(c)
#+TAGS: dd(d) ee(e) ff(f)

then there will also be a line break after the "cc" tag in the fast tag selection interface. You may also write

#+TAGS: aa(a) bb(b) cc(c) \n dd(d) ee(e) ff(f)

to achieve the same effect, and you can use \n several times in order to produce empty lines. In org-tag-alist, newlines are represented as (:newline).

Thanks to Christopher Suckling for a patch to this effect.

When a TODO is blocked by checkboxes, keep it visible in agenda

When the variable org-agenda-dim-blocked-tasks is set to invisible, tasks that are blocked will not be visible in the agenda. If the blocking is due to child TODO entries, this does make sense because the children themselves will show up in the TODO list.

However, as John Rakestraw has pointed out, if the blocking is done by checkboxes, no trace of these subtasks is left. Therefore, when the blocking is done by checkboxes, we now overrule the invisible setting and replace it with mere dimming of the task.

LaTeX can import Org's in-buffer definitions for TITLE, EMAIL etc.

If you configure org-export-latex-import-inbuffer-stuff, in-buffer definitions like #+TITLE will be made available in the LaTeX file as \orgTITLE.

This was a request by Russel Adams.

Version 6.22

Details

org-choose.el by Tom Breton is now included

Org-choose helps documenting a decision-making process by using TODO keywords for different degrees of chosenness, and by automatically keeping a set of alternatives in a consistent state.

Documentation for org-choose.el is available here.

This package inserts itself into Org using hooks, so if other people would like to do interesting stuff with TODO keywords for special purposes, looking at Tom's code might be a good way to start.

Thanks to Tom for this interesting contribution!

orgmode.org and Worg css works now better on IE

Thanks to Sebastian Rose for making these changes.

When exporting a subtree, headline levels are now relative to parent

This was reported as a bug by William Henney and is fixed now.

Inactive dates in tables can be used for sorting.

When sorting table fields or entries by date, Org first tries to find an active date, and, if none exist, uses a passive date if present.

This was a request by Hsui-Khuen Tang

The default for org-return-follows-link is back to nil

Setting it to t violates Emacs rules to some extent. The internal implementation of this has been improved, so setting it to t should now be pretty stable.

Automatic scheduling of siblings with org-depend.el

The sibling of a DONE task can now automatically be scheduled.

This was a patch by Andrew Hyatt.

New skipping conditions

The functions org-agenda-skip-entry-if and org-agenda-skip-subtree-if now accept timestamp and nottimestamp as additional conditions.

This was in response to a request by Saurabh Agrawal.

Version 6.21

Details

Changes to some default values of variables:

Here are the new default values:

(setq org-return-follows-link t)

(setq org-use-fast-todo-selection t)

(setq org-yank-adjusted-subtrees nil)

(setq org-tags-column -77)

(setq org-agenda-sorting-strategy
     '((agenda time-up priority-down category-keep)
	(todo time-up priority-down category-keep)
	(tags time-up priority-down category-keep)
	(search category-keep)))

Final cleanup for Emacs 21.1 pretest

Version 6.20

Details

Support for simple TODO dependencies

John Wiegley's code for enforcing simple TODO dependencies has been integrated into Org-mode. Thanks John!

The structure of Org files (hierarchy and lists) makes it easy to define TODO dependencies. A parent TODO task should not be marked DONE until all subtasks (defined as children tasks) are marked as DONE. And sometimes there is a logical sequence to a number of (sub)tasks, so that one task cannot be acted upon before all siblings above it are done. If you customize the variable org-enforce-todo-dependencies, Org will block entries from changing state while they have children that are not DONE. Furthermore, if an entry has a property ORDERED, each of its children will be blocked until all earlier siblings are marked DONE. Here is an example:

* TODO Blocked until (two) is done
** DONE one
** TODO two

* Parent
,  :PROPERTIES:
,    :ORDERED: t
,  :END:
** TODO a
** TODO b, needs to wait for (a)
** TODO c, needs to wait for (a) and (b)

The command C-c C-x o toggles the value of the ORDERED property.

The variable org-agenda-dim-blocked-tasks controls how blocked entries should appear in the agenda, where they can be dimmed or even made invisible.

Furthermore, you can use the variable org-enforce-todo-checkbox-dependencies to block TODO entries from switching to DONE while any checkboxes are unchecked in the entry.

Support for shift-selection in Emacs 23

Customize the variable org-support-shift-select to use S-cursor key for selecting text. Make sure that you carefully read the docstring of that variable first.

Adding and removing checkboxes from many lines

The command C-c C-x C-b normally toggles checkbox status in the current line, or in all lines in the region. With prefix argument it now either adds or removes the checkbox.

This was a requested by Daniel Clemente.

Version 6.19

Overview

• Improved behavior of conversion commands C-c - and C-c *
• Table formulas may now reference fields in other tables
• A final hline is imagined in each table, for the sake of references
• A tags-todo search can now ignore timestamped items
• \par can be used to force a paragraph break, also in footnotes

Details

Improved behavior of conversion commands C-c - and C-c *

The conversion commands C-c - and C-c * are now better behaved and therefore more useful, I hope.

If there is an active region, these commands will act on the region, otherwise on the current line.

C-c -

This command turns headings or normal lines into items, or items into normal lines. When there is a region, everything depends on the first line of the region:

• if it is a item, turn all items in the region into normal lines.

  • if it is a headline, turn all headlines in the region

  into items.

• if it is a normal line, turn all lines into items.

• special case: if there is no active region and the current line is an item, cycle the bullet type of the current list.

  C-c *

  This command turns items and normal lines into headings, or headings into normal lines. When there is a region, everything depends on the first line of the region:

• if it is a item, turn all items in the region into headlines.

  • if it is a headline, turn all headlines in the region

  into normal lines.

• if it is a normal line, turn all lines into headlines.

Table formulas may now reference fields in other tables

You may now reference constants, fields and ranges from a different table, either in the current file or even in a different file. The syntax is

remote(NAME-OR-ID,REF)

where NAME can be the name of a table in the current file as set by a #+TBLNAME: NAME line before the table. It can also be the ID of an entry, even in a different file, and the reference then refers to the first table in that entry. REF is an absolute field or range reference, valid in the referenced table. Note that since there is no "current filed" for the remote table, all row and column references must be absolute, not relative.

A final hline is imagined in each table, for the sake of references

Even if a table does not end with a hline (mine never do because I think it is not pretty), for the sake of references you can assume there is one. So in the following table

| a | b |
|---+---|
| 1 | 2 |
| 3 | 4 |

a reference like @I$1..@II$2 will now work.

A tags-todo search can now ignore timestamped items

The variables org-agenda-todo-ignore-with-date, org-agenda-todo-ignore-with-date, and org-agenda-todo-ignore-with-date make it possible to exclude TODO entries which have this kind of planning info associated with them. This is most useful for people who schedule everything, and who use the TODO list mainly to find things that are not yet scheduled. Thomas Morgan pointed out that also the tags-todo search may serve exactly this purpose, and that it might be good to have a way to make these variables also apply to the tags-todo search. I can see that, but could not convince myself to make this the default. A new variable must be set to make this happen: org-agenda-tags-todo-honor-ignore-options.

\par can be used to force a paragraph break, also in footnotes

The LaTeX idiom \par will insert a paragraph break at that location. Normally you would simply leave an empty line to get such a break, but this is useful for footnotes whose definitions may not contain empty lines.

Version 6.18

Incompatible changes

Short examples must have a space after the colon

Short literal examples can be created by preceding lines with a colon. Such lines must now have a space after the colon. I believe this is already general practice, but now it must be like this. The only exception are lines that are empty except for the colon.

Details

Include files can now also process switches

The example and src switches like -n can now also be added to include file statements:

#+INCLUDE "~/.emacs" src emacs-lisp -n -r

Thanks to Manish for pointing out that this was not yet supported.

Examples can be exported to HTML as text areas

You can now specify a -t switch to an example or src block, to make it export to HTML as a text area. To change the defaults for height (number of lines in the example) and width of this area (80), use the -h and -w switches.

Thanks to Ulf Stegemann for driving this development.

LaTeX_CLASS can be given as a property

When exporting a single subtree by selecting it as a region before export, the LaTeX class for the export will be taken from the LaTeX_CLASS property of the entry if present.

Thanks to Robert Goldman for this request.

Better handling of inlined images in different backends

Two new variables govern which kind of files can be inlined during export. These are org-export-html-inline-image-extensions and org-export-latex-inline-image-extensions. Remember that links are turned into an inline image if they are a pure link with no description. HTML files can inline .png, .jpg, and .gif files, while LaTeX files, when processed with pdflatex, can inline .png, .jpg, and .pdf files. These also represent the default settings for the new variables. Note that this means that pure links to .pdf files will be inlined - to avoid this for a particular link, make sure that the link has a description part which is not equal to the link part.

Links by ID now continue to work in HTML exported files

If you make links by ID, these links will now still work in HTML exported files, provided that you keep the relative path from link to target file the same.

Thanks to Friedrich Delgado Friedrichs for pushing this over the line.

The relative timer can be paused

The new command `C-c C-x ,' will pause the relative timer. When the relative timer is running, its value will be shown in the mode line. To get rid of this display, you need to really stop the timer with `C-u C-c C-x ,'.

Thanks to Alan Davis for driving this change.

The attachment directory may now be chosen by the user

Instead of using the automatic, unique directory related to the entry ID, you can also use a chosen directory for the attachments of an entry. This directory is specified by the ATTACH_DIR property. You can use `C-c C-a s' to set this property.

Thanks to Jason Jackson for this proposal.

You can use a single attachment directory for a subtree

By setting the property ATTACH_DIR_INHERIT, you can now tell Org that children of the entry should use the same directory for attachments, unless a child explicitly defines its own directory with the ATTACH_DIR property. You can use the command `C-c C-a i' to set this property in an entry.

Version 6.17

Overview

• Footnote support
• Line numbers and references in literal examples
• New hooks for export preprocessing
• Capture column view into a different file

Details

Footnote support

Org-mode now directly supports the creation of footnotes. In contrast to the footnote.el package, Org-mode's footnotes are designed for work on a larger document, not only for one-off documents like emails. The basic syntax is similar to the one used by footnote.el, i.e. a footnote is defined in a paragraph that is started by a footnote marker in square brackets in column 0, no indentation allowed. The footnote reference is simply the marker in square brackets inside text. For example:

The Org homepage[fn:1] now looks a lot better than it used to.
...
[fn:1] The link is: http://orgmode.org

Org-mode extends the number-based syntax to named footnotes and optional inline definition. Using plain numbers as markers is supported for backward compatibility, but not encouraged because of possible conflicts with LaTeX syntax. Here are the valid references:

[1]

A plain numeric footnote marker.

A named footnote reference, where `name' is a unique label word or, for simplicity of automatic creation, a number.

A LaTeX-like anonymous footnote where the definition is given directly at the reference point.

An inline definition of a footnote, which also specifies a name for the note. Since Org allows multiple references to the same note, you can then use use `' to create additional references.

Footnote labels can be created automatically, or you create names yourself. This is handled by the variable org-footnote-auto-label and its corresponding #+STARTUP keywords, see the docstring of that variable for details.

The following command handles footnotes:

C-c C-x f

The footnote action command. When the cursor is on a footnote reference, jump to the definition. When it is at a definition, jump to the (first) reference. Otherwise, create a new footnote. Depending on the variable `org-footnote-define-inline' (with associated #+STARTUP options fninline and nofninline), the definitions will be placed right into the text as part of the reference, or separately into the location determined by the variable org-footnote-section. When this command is called with a prefix argument, a menu of additional options is offered:

s

Sort the footnote definitions by reference sequence. During editing, Org makes no effort to sort footnote definitions into a particular sequence. If you want them sorted, use this command, which will also move entries according to org-footnote-section.

n

Normalize the footnotes by collecting all definitions (including inline definitions) into a special section, and then numbering them in sequence. The references will then also be numbers. This is meant to be the final step before finishing a document (e.g. sending off an email). The exporters do this automatically, and so could something like `message-send-hook'.

d

Delete the footnote at point, and all references to it.

C-c C-c

If the cursor is on a footnote reference, jump to the definition. If it is a the definition, jump back to the reference. When called with a prefix argument at either location, offer the same menu as `C-u C-c C-x f'.

C-c C-o or mouse-1/2

Footnote labels are also links to the corresponding definition/reference, and you can use the usual commands to follow these links.

Org-mode's footnote support is designed so that it should also work in buffers that are not in Org-mode, for example in email messages. Just bind org-footnote-action to a global key like C-c f.

The main trigger for this development came from a hook function written by Paul Rivier, to implement named footnotes and to convert them to numbered ones before export. Thanks, Paul!

Thanks also to Scot Becker for a thoughtful post bringing this subject back onto the discussion table, and to Matt Lundin for the idea of named footnotes and his prompt testing of the new features.

Line numbers and references in literal examples

Literal examples introduced with #+BEGIN_EXAMPLE or #+BEGIN_SRC do now allow optional line numbering in the example. Furthermore, links to specific code lines are supported, greatly increasing Org-mode's utility for writing tutorials and other similar documents.

Code references use special labels embedded directly into the source code. Such labels look like "(ref:name)" and must be unique within a document. Org-mode links with "(name)" in the link part will be correctly interpreted, both while working with an Org file (internal links), and while exporting to the different backends. Line numbering and code references are supported for all three major backends, HTML, LaTeX, and ASCII. In the HTML backend, hovering the mouse over a link to a source line will remote-highlight the referenced code line.

The options for the BEGIN lines are:

-n

Number the lines in the example

+n

Like -n, but continue numbering from where the previous example left off.

-r

Remove the coderef cookies from the example, and replace links to this reference with line numbers. This option takes only effect if either -n or +n are given as well. If -r is not given, coderefs simply use the label name.

-l "fmt"

Define a local format for coderef labels, see the variable org-coderef-label-format for details. Use this of the default syntax causes conflicts with the code in the code snippet you are using.

Here is an example:

#+begin_src emacs-lisp -n -r
(defmacro org-unmodified (&rest body)                   (ref:def)
  "Execute body without changing `buffer-modified-p'."
  `(set-buffer-modified-p                              (ref:back)
    (prog1 (buffer-modified-p) ,@body)))
#+end_src
[[(def)][Line (def)]] contains the macro name.  Later at line [[(back)]],
backquoting is used.

When exported, this is translated to:

(defmacro org-unmodified (&rest body)                   (ref:def)
  "Execute body without changing `buffer-modified-p'."
  `(set-buffer-modified-p                              (ref:back)
    (prog1 (buffer-modified-p) ,@body)))

Line (def) contains the macro name. Later at line /tec/org-mode/src/commit/ef2ee3df5d9eb47a171d6bdc86c67708711aa0be/ORGWEBPAGE/(back), backquoting is used.

Thanks to Ilya Shlyakhter for proposing this feature set. Thanks to Sebastian Rose for the key Javascript element that made the remote highlighting possible.

New hooks for export preprocessing

The export preprocessor now runs more hooks, to allow better-timed tweaking by user functions:

org-export-preprocess-hook

Pretty much the first thing in the preprocessor. But org-mode is already active in the preprocessing buffer.

org-export-preprocess-after-include-files-hook

This is run after the contents of included files have been inserted.

org-export-preprocess-after-tree-selection-hook

This is run after selection of trees to be exported has happened. This selection includes tags-based selection, as well as removal of commented and archived trees.

org-export-preprocess-before-backend-specifics-hook

Hook run before backend-specific functions are called during preprocessing.

org-export-preprocess-final-hook

Hook for preprocessing an export buffer. This is run as the last thing in the preprocessing buffer, just before returning the buffer string to the backend.

Capture column view into a different file

The :id parameter for the dynamic block capturing column view can now truly be an ID that will also be found in a different file. Also, it can be like file:path/to/file, to capture the global column view from a different file.

Thanks to Francois Lagarde for his report that IDs outside the current file would not work.

Version 6.16

Cleanup of many small bugs, and one new feature.

Details

References to last table row with special names

Fields in the last row of a table can now be referenced with $LR1, $LR2, etc. These references can appear both on the left hand side and right hand side of a formula.

Version 6.15f

This version reverses the introduction of @0 as a reference to the last rwo in a table, because of a conflict with the use of @0 for the current row.

Version 6.15

Overview

• All known LaTeX export issues fixed
• Captions and attributes for figures and tables.
• Better implementation for entry IDs
• Spreadsheet references to the last table line.
• Old syntax for link attributes abandoned

Incompatible changes

Old syntax for link attributes abandoned

There used to be a syntax for setting link attributes for HTML export by enclosing the attributes into double braces and adding them to the link itself, like

[[./img/a.jpg{{alt="an image"}}] ]

This syntax is not longer supported, use instead

#+ATTR_HTML: alt="an image"
[[./img/a.jpg] ]

Details

All known LaTeX export issues fixed

All the remaining issues with the LaTeX exporter have hopefully been addressed in this release. In particular, this covers quoting of special characters in tables and problems with exporting files where the headline is in the first line, or with an active region.

Captions and attributes for figures and tables.

Tables, and Hyperlinks that represent inlined images, can now be equipped with additional information that will be used during export. The information will be taken from the following special lines in the buffer and apply to the first following table or link.

#+CAPTION:

The caption of the image or table. This string should be processed according to the export backend, but this is not yet done.

#+LABEL:

A label to identify the figure/table for cross references. For HTML export, this string will become the ID for the <div class="figure"> element that encapsulates the image tag and the caption. For LaTeX export, this string will be used as the argument of a \label{...} macro. These labels will be available for internal links like [[label][Table] ].

#+ATTR_HTML:

Attributes for HTML export of image, to be added as attributes into the <img...> tag. This string will not be processed, so it should have immediately the right format.

#+ATTR_LaTeX:

Attributes for LaTeX export of images and tables.
For images, this string is directly inserted into the optional argument of the \includegraphics[...]{file} command, to specify scaling, clipping and other options. This string will not be processed, so it should have immediately the right format, like width=5cm,angle=90.
For tables, this can currently contain the keyword longtable, to request typesetting of the table using the longtable package, which automatically distributes the table over several pages if needed. Also, the attributes line may contain an alignment string for the tabular environment, like longtable,align=l|lrl

For LaTeX export, if either a caption or a label is given, the element will be exported as a float, i.e. wrapped into a figure or table environment.

Better implementation for entry IDs

Unique identifiers for entries can now be used more efficiently. Internally, a hash array has replaced the alist used so far to keep track of the files in which an ID is defined. This makes it quite fast to find an entry by ID.

There is a new link type which looks like this:

id:GLOBALLY-UNIQUE-IDENTIFIER

This link points to a specific entry. When you move the entry to a different file, for example if you move it to an archive file, the link will continue to work.

The file org-id.el contains an API that can be used to write code using these identifiers, including creating IDs and finding them wherever they are.

Org has its own method to create unique identifiers, but if the system has uuidgen command installed (Mac's and Linux systems generally do), it will be used by default (a change compared to the earlier implmentation, where you explicitdly had to opt for uuidgen). You can also select the method by hand, using the variable org-id-method.

If the ID system ever gets confused about where a certain ID is, it initiates a global scan of all agenda files with associated archives, all files previously known containing any IDs, and all currently visited Org-mode files to rebuild the hash. You can also initiate this by hand: M-x org-id-update-id-locations. Running this command will also dump into the *Messages* buffer information about any duplicate IDs. These should not exist, and Org will never make the same ID twice, but if you copy an entry with its properties, duplicate IDs will inevitably be produced. Unfortunately, this is unavoidable in a plain text system that allows you to edit the text in arbitrary ways, and a portion of care on your side is needed to keep this system clean.

The hash is stored in the file ~/.emacs.d/.org-id-locations. This is also a change from previous versions where the file was ~/.org=id-locations. Therefore, you can remove this old file if you have it. I am not sure what will happen if the .emacs.d directory does not exists in your setup, but in modern Emacsen, I believe it should exist. If you do not want to use IDs across files, you can avoid the overhead with tracking IDs by customizing the variable org-id-track-globally. IDs can then still be used for links inside a single file.

IDs will also be used when you create a new link to an Org-mode buffer. If you use org-store-link (normally at C-c l) inside en entry in an Org-mode buffer, and ID property will be created if it does not exist, and the stored link will be an id: link. If you prefer the much less secure linking to headline text, you can configure the variable org-link-to-org-use-id. The default setting for this variable is create-if-interactive, meaning that an ID will be created when you store a link interactively, but not if you happen to be in an Org-mode file while you create a remember note (which usually has a link to the place where you were when starting remember).

Spreadsheet references to the last table line.

You may now use @0 to reference the last dataline in a table in a stable way. This is useful in particular for automatically generated tables like the ones using org-collector.el by Eric Schulte.

Version 6.14

Overview

• New relative timer to support timed notes
• Special faces can be set for individual tags
• The agenda shows now all tags, including inherited ones.
• Exclude some tags from inheritance.
• More special values for time comparisons in property searches
• Control for exporting meta data
• Cut and Paste with hot links from w3m to Org
• LOCATION can be inherited for iCalendar export
• Relative row references crossing hlines now throw an error

Incompatible Changes

Relative row references crossing hlines now throw an error

Relative row references in tables look like this: "@-4" which means the forth row above this one. These row references are not allowed to cross horizontal separator lines (hlines). So far, when a row reference violates this policy, Org would silently choose the field just next to the hline.

Tassilo Horn pointed out that this kind of hidden magic is actually confusing and may cause incorrect formulas, and I do agree. Therefore, trying to cross a hline with a relative reference will now throw an error.

If you need the old behavior, customize the variable `org-table-error-on-row-ref-crossing-hline'.

Details

New relative timer to support timed notes

Org now supports taking timed notes, useful for example while watching a video, or during a meeting which is also recorded.

C-c C-x .

Insert a relative time into the buffer. The first time you use this, the timer will be started. When called with a prefix argument, the timer is reset to 0.

C-c C-x -

Insert a description list item with the current relative time. With a prefix argument, first reset the timer to 0.

M-RET

Once the time list has been initiated, you can also use the normal item-creating command to insert the next timer item.

C-c C-x 0

Reset the timer without inserting anything into the buffer. By default, the timer is reset to 0. When called with a C-u prefix, reset the timer to specific starting offset. The user is prompted for the offset, with a default taken from a timer string at point, if any, So this can be used to restart taking notes after a break in the process. When called with a double prefix argument C-c C-u, change all timer strings in the active region by a certain amount. This can be used to fix timer strings if the timer was not started at exactly the right moment.

Thanks to Alan Dove, Adam Spiers, and Alan Davis for contributions to this idea.

Special faces can be set for individual tags

You may now use the variable org-tag-faces to define the face used for specific tags, much in the same way as you can do for TODO keywords.

Thanks to Samuel Wales for this proposal.

The agenda shows now all tags, including inherited ones.

This request has come up often, most recently it was formulated by Tassilo Horn.

If you prefer the old behavior of only showing the local tags, customize the variable org-agenda-show-inherited-tags.

Exclude some tags from inheritance.

So far, the only way to select tags for inheritance was to allow it for all tags, or to do a positive selection using one of the more complex settings for `org-use-tag-inheritance'. It may actually be better to allow inheritance for all but a few tags, which was difficult to achieve with this methodology.

A new option, `org-tags-exclude-from-inheritance', allows to specify an exclusion list for inherited tags.

More special values for time comparisons in property searches

In addition to <now>, <today>, <yesterday>, and <tomorrow>, there are more special values accepted now in time comparisons in property searches: You may use strings like <+3d> or <-2w>, with units d, w, m, and y for day, week, month, and year, respectively

Thanks to Linday Todd for this proposal.

Control for exporting meta data

All the metadata in a headline, i.e. the TODO keyword, the priority cookie, and the tags, can now be excluded from export with appropriate options:

Variable	Publishing property	OPTIONS switch
org-export-with-todo-keywords	:todo-keywords	todo:
org-export-with-tags	:tags	tags:
org-export-with-priority	:priority	pri:

Cut and Paste with hot links from w3m to Org

You can now use the key C-c C-x M-w in a w3m buffer with HTML content to copy either the region or the entire file in a special way. When you yank this text back into an Org-mode buffer, all links from the w3m buffer will continue to work under Org-mode.

For this to work you need to load the new file org-w3m.el. Please check your org-modules variable to make sure that this is turned on.

Thanks for Richard Riley for the idea and to Andy Stewart for the implementation.

LOCATION can be inherited for iCalendar export

The LOCATION property can now be inherited during iCalendar export if you configure org-use-property-inheritance like this:

(setq org-use-property-inheritance '("LOCATION"))

Version 6.13

Overview

• Keybindings in Remember buffers can be configured
• Support for ido completion
• New face for date lines in agenda column view
• Invisible targets become now anchors in headlines.
• New contributed file org-exp-blocks.el
• New contributed file org-eval-light.el
• Link translation
• BBDB links may use regular expressions.
• Link abbreviations can use %h to insert a url-encoded target value
• Improved XHTML compliance

Details

Keybindings in Remember buffers can be configured

The remember buffers created with Org's extensions are in Org-mode, which is nice to prepare snippets that will actually be stored in Org-mode files. However, this makes it hard to configure key bindings without modifying the Org-mode keymap. There is now a minor mode active in these buffers, `org-remember-mode', and its keymap org-remember-mode-map can be used for key bindings. By default, this map only contains the bindings for C-c C-c to store the note, and C-c C-k to abort it. Use `org-remember-mode-hook' to define your own bindings like

(add-hook
 'org-remember-mode-hook
 (lambda ()
   (define-key org-remember-mode-map
     "\C-x\C-s" 'org-remember-finalize)))

If you wish, you can also use this to free the C-c C-c binding (by binding this key to nil in the minor mode map), so that you can use C-c C-c again to set tags.

This modification is based on a request by Tim O'Callaghan.

Support for ido completion

You can now get the completion interface from ido.el for many of Org's internal completion commands by turning on the variable org-completion-use-ido. ido-mode must also be active before you can use this.

This change is based upon a request by Samuel Wales.

New face for date lines in agenda column view

When column view is active in the agenda, and when you have summarizing properties, the date lines become normal column lines and the separation between different days becomes harder to see. If this bothers you, you can now customize the face org-agenda-column-dateline.

This is based on a request by George Pearson.

Invisible targets become now anchors in headlines.

These anchors can be used to jump to a directly with an HTML link, just like the sec-xxx IDs. For example, the following will make a http link //domain/path-to-my-file.html#dummy work:

,# <<dummy>>
*** a headline

This is based on a request by Matt Lundin.

New contributed file org-exp-blocks.el

This new file implements special export behavior of user-defined blocks. The currently supported blocks are

comment

Comment blocks with author-specific markup

ditaa

conversion of ASCII art into pretty png files

using Stathis Sideris' ditaa.jar program

dot

creation of graphs in the dot language

R

Sweave type exporting using the R program

For more details and examples, see the file commentary in org-exp-blocks.el.

Kudos to Eric Schulte for this new functionality, after org-plot.el already his second major contribution. Thanks to Stathis for this excellent program, and for allowing us to bundle it with Org-mode.

New contributed file org-eval-light.el

This module gives control over execution Emacs Lisp code blocks included in a file.

Thanks to Eric Schulte also for this file.

Link translation

You can now configure Org to understand many links created with the Emacs Planner package, so you can cut text from planner pages and paste them into Org-mode files without having to re-write the links. Among other things, this means that the command org-open-at-point-global which follows links not only in Org-mode, but in arbitrary files like source code files etc, will work also with links created by planner. The following customization is needed to make all of this work

(setq org-link-translation-function
      'org-translate-link-from-planner)

I guess an inverse translator could be written and integrated into Planner.

BBDB links may use regular expressions.

This did work all along, but only now I have documented it.

yank-pop works again after yanking an outline tree

Samuel Wales had noticed that org-yank did mess up this functionality. Now you can use yank-pop again, the only restriction is that the so-yanked text will not be pro/demoted or folded.

Link abbreviations can use %h to insert a url-encoded target value

Thanks to Steve Purcell for a patch to this effect.

Improved XHTML compliance

Thanks to Sebastian Rose for pushing this.

Many bug fixes again.

Version 6.12

Overview

• A region of entries can now be refiled with a single command
• Fine-tuning the behavior of `org-yank'
• Formulas for clocktables
• Better implementation of footnotes for HTML export
• More languages for HTML export.

Details

A region of entries can now be refiled with a single command

With transient-make-mode active (zmacs-regions under XEmacs), you can now select a region of entries and refile them all with a single C-c C-w command.

Thanks to Samuel Wales for this useful proposal.

Fine-tuning the behavior of org-yank

The behavior of Org's yanking command has been further fine-tuned in order to avoid some of the small annoyances this command caused.

• Calling org-yank with a prefix arg will stop any special treatment and directly pass through to the normal yank command. Therefore, you can now force a normal yank with C-u C-y.
• Subtrees will only be folded after a yank if doing so will now swallow any non-white characters after the yanked text. This is, I think a really important change to make the command work more sanely.

Formulas for clocktables

You can now add formulas to a clock table, either by hand, or with a :formula parameter. These formulas can be used to create additional columns with further analysis of the measured times.

Thanks to Jurgen Defurne for triggering this addition.

Better implementation of footnotes for HTML export

The footnote export in 6.11 really was not good enough. Now it works fine. If you have customized footnote-section-tag, make sure that your customization is matched by footnote-section-tag-regexp.

Thanks to Sebastian Rose for pushing this change.

More languages for HTML export.

More languages are supported during HTML export. This is only relevant for the few special words Org inserts, like "Table of Contents", or "Footnotes". Also the encoding issues with this feature seem to be solved now.

Thanks to Sebastian Rose for pushing me to fix the encoding problems.

Version 6.11

Overview

• Yanking subtree with C-y now adjusts the tree level
• State changes can now be shown in the log mode in the agenda
• Footnote in HTML export are now collected at the end of the document
• HTML export now validates again as XHTML
• The clock can now be resumed after exiting and re-starting Emacs
• Clock-related data can be saved and resumed across Emacs sessions
• Following file links can now use C-u C-u to force use of an external app
• Inserting absolute files names now abbreviates links with "~"
• Links to attachment files
• Completed repeated tasks listed briefly in agenda
• Remove buffers created during publishing are removed

Details

Yanking subtree with C-y now adjusts the tree level

When yanking a cut/copied subtree or a series of trees, the normal yank key C-y now adjusts the level of the tree to make it fit into the current outline position, without losing its identity, and without swallowing other subtrees.

This uses the command org-past-subtree. An additional change in that command has been implemented: Normally, this command picks the right outline level from the surrounding visible headlines, and uses the smaller one. So if the cursor is between a level 4 and a level 3 headline, the tree will be pasted as level 3. If the cursor is actually at the beginning of a headline, the level of that headline will be used. For example, lets say you have a tree like this:

* Level one
** Level two
,(1)
,(2)* Level one again

with (1) and (2) indicating possible cursor positions for the insertion. When at (1), the tree will be pasted as level 2. When at (2), it will be pasted as level 1.

If you do not want C-y to behave like this, configure the variable org-yank-adjusted-subtrees.

Thanks to Samuel Wales for this idea and a partial implementation.

State changes can now be shown in the log mode in the agenda

If you configure the variable org-agenda-log-mode-items, you can now request that all logged state changes be included in the agenda when log mode is active. If you find this too much for normal applications, you can also temporarily request the inclusion of state changes by pressing C-u l in the agenda.

This was a request by Hsiu-Khuern Tang.

You can also press `C-u C-u l' to get only log items in the agenda, withour any timestamps/deadlines etc.

Footnote in HTML export are now collected at the end of the document

Previously, footnotes would be left in the document where they are defined, now they are all collected and put into a special <div> at the end of the document.

Thanks to Sebastian Rose for this request.

HTML export now validates again as XHTML.

Thanks to Sebastian Rose for pushing this cleanup.

The clock can now be resumed after exiting and re-starting Emacs

If the option org-clock-in-resume is t, and the first clock line in an entry is unclosed, clocking into that task resumes the clock from that time.

Thanks to James TD Smith for a patch to this effect.

Clock-related data can be saved and resumed across Emacs sessions

The data saved include the contents of org-clock-history, and the running clock, if there is one.

To use this, you will need to add to your .emacs

(setq org-clock-persist t)
(setq org-clock-in-resume t)
(org-clock-persistence-insinuate)

Thanks to James TD Smith for a patch to this effect.

Following file links can now use C-u C-u to force use of an external app.

So far you could only bypass your setup in `org-file-apps' and force opening a file link in Emacs by using a C-u prefix arg with C-c C-o. Now you can call C-u C-u C-c C-o to force an external application. Which external application depends on your system. On Mac OS X and Windows, open is used. On a GNU/Linux system, the mailcap settings are used.

This was a proposal by Samuel Wales.

Inserting absolute files names now abbreviates links with "~".

Inserting file links with C-u C-c C-l was buggy if the setting of `org-link-file-path-type' was `adaptive' (the default). Absolute file paths were not abbreviated relative to the users home directory. This bug has been fixed.

Thanks to Matt Lundin for the report.

Links to attachment files

Even though one of the purposes of entry attachments was to reduce the number of links in an entry, one might still want to have the occasional link to one of those files. You can now use link abbreviations to set up a special link type that points to attachments in the current entry. Note that such links will only work from within the same entry that has the attachment, because the directory path is entry specific. Here is the setup you need:

(setq org-link-abbrev-alist '(("att" . org-attach-expand-link)))

After this, a link like this will work

     [[att:some-attached-file.txt]]

This was a proposal by Lindsay Todd.

Completed repeated tasks listed briefly in agenda

When a repeating task, listed in the daily/weekly agenda under today's date, is completed from the agenda, it is listed as DONE in the agenda until the next update happens. After the next update, the task will have disappeared, of course, because the new date is no longer today.

Remove buffers created during publishing are removed

Buffers that are created during publishing are now deleted when the publishing is over. At least I hope it works like this.

Version 6.10

Overview

• Secondary agenda filtering is becoming a killer feature
• Setting tags has now its own binding, C-c C-q
• Todo state changes can trigger tag changes
• C-RET will now always insert a new headline, never an item.
• Customize org-mouse.el feature set to free up mouse events
• New commands for export all the way to PDF (through LaTeX)
• Some bug fixed for LaTeX export, more bugs remain.

Details

Enhancements to secondary agenda filtering

This is, I believe, becoming a killer feature. It allows you to define fewer and more general custom agenda commands, and then to do the final narrowing to specific tasks you are looking for very quickly, much faster than calling a new agenda command.

If you have not tries this yet, you should!

You can now refining the current filter by an additional criterion

When filtering an existing agenda view with /, you can now narrow down the existing selection by an additional condition. Do do this, use \ instead of / to add the additional criterion. You can also press + or - after / to add a positive or negative condition. A condition can be a TAG, or an effort estimate limit, see below.

It is now possible to filter for effort estimates

This means to filter the agenda for the value of the Effort property. For this you should best set up global allowed values for effort estimates, with

(setq org-global-properties
      '(("Effort_ALL" . "0 0:10 0:30 1:00 2:00 3:00 4:00")))

You may then select effort limits with single keys in the filter. It works like this: After / or \, first select the operator which you want to use to compare effort estimates:

<   Select entries with effort smaller than or equal to the limit
>   Select entries with effort larger than or equal to the limit
=   Select entries with effort equal to the limit

After that, you can press a single digit number which is used as an index to the allowed effort estimates.

If you do not use digits to fast-select tags, you can even skip the operator, which will then default to `org-agenda-filter-effort-default-operator', which is by default <.

Thanks to Manish for the great idea to include fast effort filtering into the agenda filtering process.

The mode line will show the active filter

For example, if there is a filter in place that does select for HOME tags, against EMAIL tags, and for tasks with an estimated effort smaller than 30 minutes, the mode-line with show +HOME-EMAIL+<0:30

The filter now persists when the agenda view is refreshed

All normal refresh commands, including those that move the weekly agenda from one week to the next, now keep the current filter in place.

You need to press / / to turn off the filter. However, when you run a new agenda command, for example going from the weekly agenda to the TODO list, the filter will be switched off.

Setting tags has now its own binding, C-c C-q

You can still use C-c C-c on a headline, but the new binding should be considered as the main binding for this command. The reasons for this change are:

• Using C-c C-c for tags is really out of line with other uses of C-c C-c.
• I hate it in Remember buffers when I try to set tags and I cannot, because C-c C-c exits the buffer :-(
• C-c C-q will also work when the cursor is somewhere down in the entry, it does not have to be on the headline.

Todo state changes can trigger tag changes

The new option org-todo-state-tags-triggers can be used to define automatic changes to tags when a TODO state changes. For example, the setting

(setq org-todo-state-tags-triggers
      '((done ("Today" . nil) ("NEXT" . nil))
        ("WAITING" ("Today" . t))))    

will make sure that any change to any of the DONE states will remove tags "Today" and "NEXT", while switching to the "WAITING" state will trigger the tag "Today" to be added.

I use this mostly to get rid of TODAY and NEXT tags which I apply to select an entry for execution in the near future, which I often prefer to specific time scheduling.

C-RET will now always insert a new headline, never an item.

The new headline is inserted after the current subtree.

Thanks to Peter Jones for patches to fine-tune this behavior.

Customize org-mouse.el feature set

There is a new variable org-mouse-features which gives you some control about what features of org-mouse you want to use. Turning off some of the feature will free up the corresponding mouse events, or will avoid activating special regions for mouse clicks. By default I have urned off the feature to use drag mouse events to move or promote/demote entries. You can of course turn them back on if you wish.

This variable may still change in the future, allowing more fine-grained control.

New commands for export to PDF

This is using LaTeX export, and then processes it to PDF using pdflatex.

C-c C-e p     process to PDF.
C-c C-e d     process to PDF, and open the file.

LaTeX export

• \usepackage{graphicx} is now part of the standard class definitions.
• Several bugs fixed, but definitely not all of them :-(

New option `org-log-state-notes-insert-after-drawers'

Set this to t if you want state change notes to be inserted after any initial drawers, i.e drawers the immediately follow the headline and the planning line (the one with DEADLINE/SCHEDULED/CLOSED information).

Version 6.09

Incompatible

org-file-apps now uses regular expressions, see below

Details

org-file-apps now uses regular repressions instead of extensions

Just like in auto-mode-alist, car's in the variable org-file-apps that are strings are now interpreted as regular expressions that are matched against a file name. So instead of "txt", you should now write "\\.txt\\'" to make sure the matching is done correctly (even though "txt" will be recognized and still be interpreted as an extension).

There is now a shortcut to get many file types visited by Emacs. If org-file-apps contains `(auto-mode . emacs)', then any files that are matched by `auto-mode-alist' will be visited in emacs.

Changes to the attachment system

• The default method to attach a file is now to copy it instead of moving it.
• You can modify the default method using the variable `org-attach-method'. I believe that most Unix people want to set it to `ln' to create hard links.
• The keys c, m, and l specifically select copy, move, or link, respectively, as the attachment method for a file, overruling `org-attach-method'.
• To create a new attachment as an Emacs buffer, you have not now use n instead of c.
• The file list is now always retrieved from the directory itself, not from the "Attachments" property. We still keep this property by default, but you can turn it off, by customizing the variable org-attach-file-list-property.

Version 6.08

Incompatible changes

• Changes in the structure of IDs, see here for details.
• C-c C-a has been redefined, see here for details.

Details

The default structure of IDs has changed

IDs created by Org have changed a bit:

• By default, there is no prefix on the ID. There used to be an "Org" prefix, but I now think this is not necessary.
• IDs use only lower-case letters, no upper-case letters anymore. The reason for this is that IDs are now also used as directory names for org-attach, and some systems do not distinguish upper and lower case in the file system.
• The ID string derived from the current time is now reversed to become an ID. This assures that the first two letters of the ID change fast, so hat it makes sense to split them off to create subdirectories to balance load.
• You can now set the `org-id-method' to `uuidgen' on systems which support it.

C-c C-a no longer calls `show-all'

The reason for this is that C-c C-a is now used for the attachment system. On the rare occasions that this command is needed, use M-x show-all, or C-u C-u C-u TAB.

New attachment system

You can now attach files to each node in the outline tree. This works by creating special directories based on the ID of an entry, and storing files in these directories. Org can keep track of changes to the attachments by automatically committing changes to git. See the manual for more information.

Thanks to John Wiegley who contributed this fantastic new concept and wrote org-attach.el to implement it.

New remember template escapes

%^{prop}p   to insert a property
%k          the heading of the item currently being clocked
%K          a link to the heading of the item currently being clocked

Also, when you exit remember with C-2 C-c C-c, the item will be filed as a child of the item currently being clocked. So the idea is, if you are working on something and think of a new task related to this or a new note to be added, you can use this to quickly add information to that task.

Thanks to James TD Smith for a patch to this effect.

Clicking with mouse-2 on clock info in mode-line visits the clock.

Thanks to James TD Smith for a patch to this effect.

New file in contrib: lisp/org-checklist.el

This module deals with repeated tasks that have checkbox lists below them.

Thanks to James TD Smith for this contribution.

New in-buffer setting #+STYLE

It can be used to locally set the variable `org-export-html-style-extra'. Several such lines are allowed-, they will all be concatenated. For an example on how to use it, see the publishing tutorial.

Version 6.07

Overview

• Filtering existing agenda views with respect to a tag
• Editing fixed-width regions with picture or artist mode
• org-plot.el is now part of Org
• Tags can be used to select the export part of a document
• Prefix interpretation when storing remember notes
• Yanking inserts folded subtrees
• Column view capture tables can have formulas, plotting info
• In column view, date stamps can be changed with S-cursor keys
• The note buffer for clocking out now mentions the task
• Sorting entries alphabetically ignores TODO keyword and priority
• Agenda views can sort entries by TODO state
• New face org-scheduled for entries scheduled in the future.
• Remember templates for gnus links can use the :to escape.
• The file specification in a remember template may be a function
• Categories in iCalendar export include local tags
• It is possible to define filters for column view
• Disabling integer increment during table Field copy
• Capturing column view is on `C-c C-x i'
• And tons of bugs fixed.

Incompatible changes

Prefix interpretation when storing remember notes has changed

The prefix argument to the `C-c C-c' command that finishes a remember process is now interpreted differently:

C-c C-c       Store the note to predefined file and headline
C-u C-c C-c   Like C-c C-c, but immediately visit the note
              in its new location.
C-1 C-c C-c   Select the storage location interactively
C-0 C-c C-c   Re-use the last used location

This was requested by John Wiegley.

Capturing column view is now on `C-c C-x i'

The reason for this change was that `C-c C-x r' is also used as a tty key replacement.

Categories in iCalendar export now include local tags

The locally defined tags are now listed as categories when exporting to iCalendar format. Org's traditional file/tree category is now the last category in this list. Configure the variable org-icalendar-categories to modify or revert this behavior.

This was a request by Charles Philip Chan.

Details

Secondary filtering of agenda views.

You can now easily and interactively filter an existing agenda view with respect to a tag. This command is executed with the / key in the agenda. You will be prompted for a tag selection key, and all entries that do not contain or inherit the corresponding tag will be hidden. With a prefix argument, the opposite filter is applied: entries that do have the tag will be hidden.

This operation only hides lines in the agenda buffer, it does not remove them. Changing the secondary filtering does not require a new search and is very fast.

If you press TAB at the tag selection prompt, you will be switched to a completion interface to select a tag. This is useful when you want to select a tag that does not have a direct access character.

A double / / will restore the original agenda view by unhiding any hidden lines.

This functionality was John Wiegley's idea. It is a simpler implementation of some of the query-editing features proposed and implemented some time ago by Christopher League (see the file contrib/lisp/org-interactive-query.el).

Editing fixed-width regions with picture or artist mode

The command @<code>C-c '@</code> (that is C-c followed by a single quote) can now also be used to switch to a special editing mode for fixed-width sections. The default mode is artist-mode which allows you to create ASCII drawings.

It works like this: Enter the editing mode with @<code>C-c '@</code>. An indirect buffer will be created and narrowed to the fixed-width region. Edit the drawing, and press @<code>C-c '@</code> again to exit.

Lines in a fixed-width region should be preceded by a colon followed by at least one space. These will be removed during editing, and then added back when you exit the editing mode.

Using the command in an empty line will create a new fixed-width region.

This new feature arose from a discussion involving Scott Otterson, Sebastian Rose and Will Henney.

org-plot.el is now part of Org.

You can run it by simple calling org-plot/gnuplot. Documentation is not yet included with Org, please refer to http://github.com/eschulte/org-plot/tree/master until we have moved the docs into Org or Worg.

Thanks to Eric Schulte for this great contribution.

Tags can be used to select the export part of a document

You may now use tags to select parts of a document for inclusion into the export, and to exclude other parts. This behavior is governed by two new variables: org-export-select-tags and org-export-exclude-tags. These default to ("export") and ("noexport"), but can be changed, even to include a list of several tags.

Org first checks if any of the select tags is present in the buffer. If yes, all trees that do not carry one of these tags will be excluded. If a selected tree is a subtree, the heading hierarchy above it will also be selected for export, but not the text below those headings. If none of the select tags is found anywhere in the buffer, the whole buffer will be selected for export. Finally, all subtrees that are marked by any of the exclude tags will be removed from the export buffer.

You may set these tags with in-buffer options EXPORT_SELECT_TAGS and EXPORT_EXCLUDE_TAGS.

I love this feature. Thanks to Richard G Riley for coming up with the idea.

Prefix interpretation when storing remember notes

The prefix argument to the `C-c C-c' command that finishes a remember process is now interpreted differently:

C-c C-c       Store the note to predefined file and headline
C-u C-c C-c   Like C-c C-c, but immediately visit the note
              in its new location.
C-1 C-c C-c   Select the storage location interactively
C-0 C-c C-c   Re-use the last used location

This was requested by John Wiegley.

Yanking inserts folded subtrees

If the kill is a subtree or a sequence of subtrees, yanking them with C-y will leave all the subtrees in a folded state. This basically means, that kill and yank are now much more useful in moving stuff around in your outline. If you do not like this, customize the variable org-yank-folded-subtrees.

Right now, I am only binding C-y to this new function, should I modify all bindings of yank? Do we need to amend yank-pop as well?

This feature was requested by John Wiegley.

Column view capture tables can have formulas, plotting info

If you attach formulas and plotting instructions to a table capturing column view, these extra lines will now survive an update of the column view capture, and any formulas will be re-applied to the captured table. This works by keeping any continuous block of comments before and after the actual table.

In column view, date stamps can be changed with S-cursor keys

If a property value is a time stamp, S-left and S-right can now be used to shift this date around while in column view.

This was a request by Chris Randle.

The note buffer for clocking out now mentions the task

This was a request by Peter Frings.

Sorting entries alphabetically ignores TODO keyword and priority

Numerical and alphanumerical sorting now skips any TODO keyword or priority cookie when constructing the comparison string. This was a request by Wanrong Lin.

Agenda views can sort entries by TODO state

You can now define a sorting strategy for agenda entries that does look at the TODO state of the entries. Sorting by TODO entry does first separate the non-done from the done states. Within each class, the entries are sorted not alphabetically, but in definition order. So if you have a sequence of TODO entries defined, the entries will be sorted according to the position of the keyword in this sequence.

This follows an idea and sample implementation by Christian Egli.

New face org-scheduled for entries scheduled in the future.

This was a request by Richard G Riley.

Remember templates for gnus links can now use the :to escape.

Thanks to Tommy Lindgren for a patch to this effect.

The file specification in a remember template may now be a function

Thanks to Gregory Sullivan for a patch to this effect.

Categories in iCalendar export now include local tags

The locally defined tags are now listed as categories when exporting to iCalendar format. Org's traditional file/tree category is now the last category in this list. Configure the variable org-icalendar-categories to modify or revert this behavior.

This was a request by Charles Philip Chan.

It is now possible to define filters for column view

The filter can modify the value that will be displayed in a column, for example it can cut out a part of a time stamp. For more information, look at the variable org-columns-modify-value-for-display-function.

Disabling integer increment during table field copy

Prefix arg 0 to S-RET does the trick.

This was a request by Chris Randle.

Older changes

For older Changes, see /tec/org-mode/src/commit/ef2ee3df5d9eb47a171d6bdc86c67708711aa0be/ORGWEBPAGE/Changes_old.org