Fix link

[worg.git] / org-devel.org

#+OPTIONS:    H:3 num:nil toc:t \n:nil ::t |:t ^:nil -:t f:t *:t tex:t d:(HIDE) tags:not-in-toc ':t
#+STARTUP:    align fold nodlcheck hidestars oddeven lognotestate
#+SEQ_TODO:   TODO(t) INPROGRESS(i) WAITING(w@) | DONE(d) CANCELED(c@)
#+TAGS:       Write(w) Update(u) Fix(f) Check(c) 
#+TITLE:      Ongoing Development of Org Additions?
#+AUTHOR:     Worg people
#+EMAIL:      mdl AT imapmail DOT org
#+LANGUAGE:   en
#+PRIORITIES: A C B
#+CATEGORY:   worg

#+LINK: fireforgrepofile http://repo.or.cz/w/org-fireforg.git?a=blob_plain;f=%s;hb=HEAD

# This file is the default header for new Org files in Worg.  Feel free
# to tailor it to your needs.

[[file:index.org][{Back to Worg's index}]]

Some org-mode related project currently being developed in worg.

* Org Collector

#+index: Collector

Located in the =contrib/lisp/= directory of org-mode.

A utility for collecting properties from the headers in an org file,
running the properties through arbitrary elisp functions, and
presenting the results in a table.

The functionality of this tool is similar to the functionality of
[[info:org:Capturing%20column%20view][info:org:Capturing column view]], but this has support for processing
prior to the generation of the table.

Here is a simple example application of this utility.

#+begin_comment ems better example
it might be better to put an exercise example here if someone has one.
#+end_comment

#+BEGIN: propview :id "data" :cols (ITEM f d list (apply '* list) (+ f d))

#+END:

*** Example Data
      :PROPERTIES:
      :ID:       data
      :END:

****** run1
      :PROPERTIES:
      :d: 33
      :f: 2
      :list: '(9 2 3 4 5 6 7)
      :END:


****** run2
      :PROPERTIES:
      :d: 34
      :f: 4
      :END:


****** run3
      :PROPERTIES:
      :d: 35
      :f: 4
      :END:


****** run4
      :PROPERTIES:
      :d: 36
      :f: 2
      :END:


* Fireforg, a Firefox extension for org interaction (EXPERIMENTAL)

#+index: Fireforg

Fireforg is a Firefox extension, that interacts with Emacs Org-mode.

For the displayed URL it shows the associated tags in the status bar
and allows to show a list of annotations that can be instantly visited
in Emacs.

Additionally, in every website it marks all links that occur in the agenda files.

For every link in the agenda files the previous heading and its tags
are associated with it.

For *scientific use*, it is possible to *send BibTeX entries* for
sites supported by the Zotero plugin to Emacs. [[*Import%20BibTeX%20using%20Zotero][Additional steps]] are
necessary for this to work.

** Caveats
   :PROPERTIES:
   :ID:       13179b21-d70a-4255-a8f1-9f4e4e729074
   :END:
The following notes should be taken into consideration before using Fireforg:
 - On the Emacs side a registry of all links in the agenda files will
   be generated and saved in plain text onto the harddrive.
   - The generation can take very a long time depending on the size of
     the agenda files and the number of links in them.
   - The generated file can become quite large.
   - When files are *encrypted* having some information from them in
     plain text on the harddrive might not be what you want.
 - Firefox will read and look through the registry file for the
   currently viewed site. Depending on the size of the registry file
   and on the options [[id:e77f15a8-c358-44fd-a207-8c422fee2d1e][Lookup links on page load]] and [[id:d73476f3-6c09-479c-abea-f33d3e0e074a][Prefetch links to extract DOI]] 
   this might slow down the browsing experience.

** Installation

   - Install org-protocol according to its [[file:org-contrib/org-protocol.org][manual]] and *please verify
     that it's working* using either the bookmarks or by invoking it directly in the shell:
     : emacsclient 'org-protocol://remember://w/http%3A%2F%2Forgmode.org%2F/Org-Mode%3A%20Your%20Life%20in%20Plain%20Text%20-%20Mozilla%20Firefox/'
   - Get org-fireforg.el from [[fireforgrepofile:lisp/org-fireforg.el][here]] and put the following into your
     =.emacs= *after* the section that initializes org-protocol:
     : (add-to-list 'load-path "~/path/to/org/fireforg/")
     : (require 'org-fireforg)
     : (org-fireforg-registry-insinuate)
     
     and as the last line of the file:
     : (org-fireforg-registry-initialize t)

     Note that creating the registry can take a long time depending on
     the size of the agenda files and the number of links in them.

   - Visit [[fireforgrepofile:build/fireforg.xpi][this link]] with Firefox, confirm installation of the extension and restart.

     (Windows users will have to adjust: org-fireforg-registry-file-xml  in Emacs
     and "registry-file" in Fireforgs preferences to be the same file.)

*** Installation from a git clone

    It is also possible to clone the git repository first, using:
    : git clone http://repo.or.cz/r/org-fireforg.git
    , install lisp/org-fireforg.el according to the instructions above
    and add a file named =fireforg@burtzlaff.de= containing the path
    to the "fireforg" subdirectory into =~/.mozilla/firefox/<your
    profile>/extensions/=.

    In my setup for example there is a file
    =~/.mozilla/firefox/4xyx9l73.default/extensions/fireforg@burtzlaff.de=
    containing
    : /home/andy/projects/org-fireforg/fireforg
 
    A restart of Firefox is required for this to work.

** Basic usage
*** The status bar entry

    If the currently viewed url is found in your agenda files, the number
    of occurrences together with all associated tags will be shown in the
    status bar. A left click on the status bar entry will show a list of
    all headings associated with the currently viewed url. Selecting one
    of them lets Emacs visit that heading.

    For example, this heading in one of your agenda files:
    : * Greatest tool in the world [[http://orgmode.org/][Org mode]]    :Org:
    will yield the following when visiting http://orgmode.org/ and left clicking:

    [[file:images/screenshots/org-fireforg-screenshot.png]]

*** Triggering store-link and remember
    Right clicking on the status bar entry shows a menu that let's you
    trigger org-protocol's "store-link" and "remember".

    There is an entry for every remember template listed in the [[*List%20of%20characters%20specifying%20available%20remember%20templates][preferences]].

    The entries in the submenu "All tabs" will call remember for every tab
    in the current window. This option only makes sense if a remember
    template is used, that stores the note automatically ("%!" in the template string), e.g.:
    : * %:description \n %:link %!

*** Mark links that occur in the agenda files in websites

    Whenever a site is loaded, Fireforg will alter the style of all links
    in it, that occur in the agenda files. The tooltip of those links is
    set to contain the annotations.

    This feature can be turned off using an [[*Lookup%20links%20on%20page%20load][option]] in the [[*Preferences][preferences
    dialog]].  [[*CSS%20style%20string%20for%20links%20with%20annotations][The CSS-style]] used for marking the link and [[*Overwrite%20tooltip%20for%20links%20with%20annotations][whether or
    not the tooltip is set]] is also customizable.

*** Context menu for links

    The context menu (accessible by right clicking on a link) has a
    submenu "Fireforg", where all annotations for that link are listed:

    [[file:images/screenshots/org-fireforg-screenshot-context-menu.png]]

** Import BibTeX using Zotero
   :PROPERTIES:
   :ID:       e6fc94c6-7fef-4348-b998-f6a6f58eded8
   :END:
   Fireforg is able to retrieve BibTex entries for the all sites
   supported by [[http://www.zotero.org/][Zotero]]. To achieve this the following additional steps
   are necessary:
   - Install Zotero 1.0.10 from the [[http://www.zotero.org/][Zotero Website]]
   - Set "Inject Zotero" in Fireforg's preference dialog
     [fn:ffprefdiag: Fireforg's preference dialogue is accessible in
     Firefox's menu under Tools->Add-ons->Fireforg->Preferences].
     
     *Warning: On restart a small function is injected into Zotero to
     catch imported entries. The change to the code is minimal and
     non-permanent. In a worst case scenario Zotero's database might
     get corrupted, though that hasn't happened yet.*

   - Restart Firefox

     Whenever a single entry (*not* a collection) is imported into Zotero - 
     e.g. by clicking the white "document" symbol that appears to the right
     in the url bar if Zotero supports importing the current site - it is
     automatically exported to BibTeX and sent to org mode using
     org-protocol. In Emacs it is put into the kill ring in a format
     depending on the variable =org-fireforg-received-bibtex-format=:

   - =nil=: The BibTeX entry is passed directly into the kill ring.
   - =heading= (Default): A heading is generated with the BibTeX
     fields as properties with prefix "BIB_":
     : * [[<link to site>][<Title>]]
     :  :PROPERTIES:
     :   :CUSTOM_ID: <BibTeX key>
     :   :BIB_entryType: <article, ...>
     :   :BIB_author: ...
     :   :BIB_title: ...
     :   ...
     :   :END:
   - =headingWithPropsAndBibTeXContent=: a heading with properties as
     described above is generated and the entry in BibTeX format is
     appended.
   - =headingWithBibTeXContent=: same as the previous one but without the properties

     If the 'url' field is a *static URL*, the link to it will be highlighted
     in search results in every search engine. Otherwise the option
     [[id:3ab02821-03c4-4fa7-9a3a-e9701245c5d8][Match sites by comparing their DOI with saved bibliography entries]] might help.

     To export BibTeX information from all headings in the current buffer
     with at least the "BIB_entryType" property call:
   - =org-fireforg-export-bibtex-to-file= to export to a file
   - =org-fireforg-export-bibtex-to-new-buffer= to export into a new buffer

*** Digital Object Identifiers (DOI)
    :PROPERTIES:
    :ID:       fbd09ba5-6632-40df-bcc0-8e5a7b9eb7a7
    :END:
    If a BibTeX entry contains a field 'doi', a URL will be generated by
    prepending "http://dx.doi.org/" to the corresponding property
    'BIB_doi'. It will be handled as any other URL. There is a
    [[id:3ab02821-03c4-4fa7-9a3a-e9701245c5d8][corresponding functionality in the plugin]] that allows to associated
    pages despite non-static URLs.
** Preferences
   Fireforg's preference dialogue is accessible in Firefox's menu
   under Tools->Add-ons->Fireforg->Preferences
***** Registry file
      The file containing an xml tree with all link-headline associations
      generated from the agenda files. You should not need to change
      this. It has to be same as the customizable variable
      =org-fireforg-registry-file-xml= in Emacs.
***** Lookup links on page load
      :PROPERTIES:
      :ID:       e77f15a8-c358-44fd-a207-8c422fee2d1e
      :END:
      When enabled, all links in a web page that have headlines
      associated with them are marked. This is achieve by adding the
      following CSS style string to the element:
      Depending on the size of the registry this *might slow Firefox down*.
***** CSS style string for links with annotations
      The string that is appended to the CSS =style= string of a link
      element, if annotations for it exist.
***** Overwrite tooltip for links with annotations
      If enabled, sets the tooltip for the links, for which
      annotations exist to contain those annotations.
***** Inject Zotero
      After restarting Firefox, a function in the Zotero code gets altered
      so that all BibtTeX entries (*not* collections) that are imported are
      sent using org-protocol and are handled in Emacs according to the
      variable =org-fireforg-received-bibtex-format= as described [[* Import BibTeX using Zotero][here]].  Due
      to [[* Technical note][design choices in Zotero]] this is a bit fragile and can yield errors
      and *might possibly even break Zotero's database*. It is not advisable
      to use Zotero for production when enabling this option in Fireforg.
***** Match sites by comparing their DOI with saved bibliography entries
      :PROPERTIES:
      :ID:       3ab02821-03c4-4fa7-9a3a-e9701245c5d8
      :END:
      Extract the Digital Object Identifier (DOI) from a page, prepend
      "http://dx.doi.org/" to it and look up the resulting URL.

      If using a bibliography format [[id:e6fc94c6-7fef-4348-b998-f6a6f58eded8][with properties]], a [[id:fbd09ba5-6632-40df-bcc0-8e5a7b9eb7a7][URL is
      generated in the same way from the value of the field "BIB_doi"
      if it exists]]. Thus documents can be matched regardless of the
      possibly non-static URL.
***** Prefetch links to extract DOI
      :PROPERTIES:
      :ID:       d73476f3-6c09-479c-abea-f33d3e0e074a
      :END:
      *Prefetch all links in a page* after it is loaded, extract the
      DOIs - if any - and [[id:fbd09ba5-6632-40df-bcc0-8e5a7b9eb7a7][use it to find annotations]]. This requires
      the option [[id:3ab02821-03c4-4fa7-9a3a-e9701245c5d8][Match sites by comparing their DOI with saved
      bibliography entries]] to be set.

      A site is only prefetched once in every Firefox session, because the
      associated URL mapping is saved until Firefox is restarted.

      *All links starting with "http" will be prefetched (except for
      files with extensions: PDF, GIF, PNG or SWF).* This option can
      also be toggled in the status bar menu.

      *This option will generate additional network traffic and might
      slow the browsing experience*
***** List of characters specifying available remember templates
      For every character in this list an entry in the [[*Triggering%20store%20link%20and%20remember][popup menu]] will
      be generated, that triggers remember with the template
      associated with the character.
***** Enable workaround for Mac
      see [[* Workaround for the inability to register a protocol in Firefox on the Mac][here]]
***** Temporary file for Mac workaround
      see [[* Workaround for the inability to register a protocol in Firefox on the Mac][here]]

** Workaround for the inability to register a protocol in Firefox on the Mac

   A long known bug in Firefox on the Mac is reported to stop protocol
   registration from working. To work around this Fireforg is able to
   write the org-protocol urls to a temporary file, that is read every
   second and, if non empty, passed to emacsclient and emptied.

   To enable this:
   - check "Enable workaround for Mac" in Fireforg's preference dialogue [fn:ffprefdiag]
   - get pull.sh from [[http://repo.or.cz/w/org-fireforg.git?a=blob_plain;f=ff_mac_workaround/pull.sh;hb=HEAD][the repository]] and run it.

** Updating 

   To avoid confusion, always update both org-fireforg.el and the plugin.

   The plugin has to be uninstalled and then reinstalled as described
   above. Automatic updating will be used when the testing phase is over.

** Bugreporting and discussion

   - Discussions go to the org-mode list.
   - Bugreports are better not sent to the list, but rather directly to
     the [[mailto:andreas%20AT%20burtzlaff%20DOT%20de][author]] (Please add "[Fireforg]" to the subject.).
     
*** A checklist for bug tracing

    To create a test case put:
    : * Greatest tool in the world [[http://orgmode.org/][Org mode]]    :Org:
    into one of your agenda files and save it.

    If problems arise please go through this checklist to locate the problem:

    - Does the file "~/.org-fireforg-registry.xml" exist and does it contain "orgmode.org"?
    - *No on either*: Send me the last content of the Messages buffer in Emacs
    - *Yes*: In the Firefox menu: "Tools"->"Error console" look for
      errors containing: "chrome://fireforg/" and send them to me.

** Technical note

   Different instances of Zotero's Translator object seem to share state
   in a non-obvious way. This makes coding very fragile and even lets
   some imports fail (silently) after Fireforg has injected its code. The
   failure when importing collections is somehow related to this.  While
   I find it a strange design choice, it is not in my power to change it.
* MEMO org-mail-htmlize: Create MIME messages based on Org

** Representing a MIME internet message

   A MIME internet message consists of one or more MIME entities. Each
   MIME entity is of a distinct type and subtype, has a body and
   optional MIME headers related to its content.

   A MIME entity is represented as a list:

   (TYPE SUBTYPE BODY CONT-HEAD)

   - TYPE :: Symbol of MIME media type (e.g. text, video, audio).

   - SUBTYPE :: Symbol of MIME media subtype (e.g. plain, html).

   - BODY :: String with entity body -or- list of other MIME entities.

   - CONT-HEAD :: List of cons with content related MIME header
                  fields.  The name of the header field without the
                  prefix "Content-" is car, the value cdr.

   Example:

   #+begin_src emacs-lisp
   '(text html "<h1>Headline</h1>" ((disposition . inline)))
   #+end_src

   For messages of type multipart the body consists of a list of one
   or more MIME entities.

   #+begin_src emacs-lisp
     '(multipart alternative
                 '((text plain "* Headline")
                   (text html "<h1>headline</h1>")))
   #+end_src

** MIME delimiters of SEMI and mml

   The MIME delimiters are defined in an association list with a
   symbol of the library's name as key and delimiter format strings as
   values.  For each library there are three formatstrings.

   (SYMBOL DELIM-SINGLE DELIM-SINGLE-CONT DELIM-MULTI)

   - DELIM-SINGLE :: Denoting a single MIME entity.

                     Strings are passed in this order:

                     1. type

                     2. subtype

                     3. content header

                     4. body

   - DELIM-SINGLE-CONT :: Format of content header strings.

        Strings are passed in this order:

        1. header field name

        2. header field value

   - DELIM-MULTI :: Enclosing parts of a multipart entity.

                    Strings are passed in this order:

                    1. subtype

                    2. body

                    3. subtype

   #+begin_src emacs-lisp
     (setq org-mail-htmlize-mime-delimiter-alist
           '((semi "\n--[[%s/%s%s]]\n%s" "\ncontent-%s: %s" "\n--<<%s>>-{\n%s\n--}-<<%s>>")
             (mml "\n<#part type=\"%s/%s\"%s>\n%s" " %s=\"%s\"" "\n<#multipart type=\"%s\">\n%s\n<#/multipart>")))
   #+end_src

** Generic function

   This generic function returns a string representation with MIME
   delimiters depending on the variable =org-mail-htmlize-mime-lib=.

   #+begin_src emacs-lisp
     (setq org-mail-htmlize-mime-lib 'semi)
   #+end_src

   #+begin_src emacs-lisp
     (defun org-mail-htmlize-mime-entity (type subtype body
                                               &optional cont-head)
       "Return string representation of MIME entity.
     
     TYPE is the type of entity body.
     SUBTYPE is the subtype of body.
     BODY is the body of the entity.  Either a string with the body
     content or a list with one ore more MIME entities.
     Optional argument CONT-HEAD is a list of cons with content
     specific headers, name in car and value in cdr."
       (let ((delimlst (assoc org-mail-htmlize-mime-lib
                              org-mail-htmlize-mime-delimiter-alist)))
         (if (eq type 'multipart)
             (format (nth 3 delimlst) subtype
                     (mapconcat (lambda (b)
                                   (apply 'org-mail-htmlize-mime-entity
                                          (car b) (cadr b) (cddr b)))
                                body "")
                     subtype)
           (format (nth 1 delimlst)
                   type subtype
                   (mapconcat (lambda (h)
                                 (format (nth 2 delimlst) (car h) (cdr h)))
                              cont-head "")
                   body))))
   #+end_src

** Open questions

*** How to handle charset information?

*** How to attach files?

    The generic function expects BODY either be a string or a list.
    Attaching binary file (image, etc.) requires to encode it so the
    message will pass the message system.  So we /might/ use kind of a
    encoder (e.g. base64) on our own.

    Or, what seems a cleaner solution: Use attachment function of the
    respective MIME mode.  To achive this: Introduce third possibility
    for BODY: A cons with the filename in car and symbol of the
    function in cdr.

    (FILENAME . FUNCTION)

    #+begin_src emacs-lisp
      '(image jpeg ("/path/to/image" . org-mail-htmlize-add-attachment))
    #+end_src

    The function =org-mail-htmlize-add-attachment= is called with file
    name as argument and calls the appropriate function depending on
    =org-mail-htmlize-mime-lib= and returns a string

       - with the encoded body

         -or-

       - the complete MIME entity

    Side effect: The user might be prompted for attachment settings
    (e.g. encoding).  But, on the other hand: We delegate the job of
    creating the attachment to the library that is responsible for
    mime.

** Quotes from the specs

*** MIME multipart: Notion of structured, related body parts
  :PROPERTIES:
  :Created: [2010-03-25 Do]
  :END:

  - [[http://tools.ietf.org/html/rfc2046.html#section-5.1.1][RFC2046, 5.1.1]]

    #+BEGIN_QUOTE
       NOTE:  Conspicuously missing from the "multipart" type is a notion of
       structured, related body parts. It is recommended that those wishing
       to provide more structured or integrated multipart messaging
       facilities should define subtypes of multipart that are syntactically
       identical but define relationships between the various parts. For
       example, subtypes of multipart could be defined that include a
       distinguished part which in turn is used to specify the relationships
       between the other parts, probably referring to them by their
       Content-ID field.  Old implementations will not recognize the new
       subtype if this approach is used, but will treat it as
       multipart/mixed and will thus be able to show the user the parts that
       are recognized.
    #+END_QUOTE
*** MIME multipart: order of entities inside a multipart
  :PROPERTIES:
  :Created: [2010-03-25 Do]
  :END:

  - [[http://tools.ietf.org/html/rfc2046.html#section-5.1.3][RFC2046, 5.1.3]]

    #+BEGIN_QUOTE
    5.1.3.  Mixed Subtype

       The "mixed" subtype of "multipart" is intended for use when the body
       parts are independent and need to be bundled in a particular order.
       Any "multipart" subtypes that an implementation does not recognize
       must be treated as being of subtype "mixed".

    #+END_QUOTE

  - [[http://tools.ietf.org/html/rfc2046.html#section-5.1.4][RFC2046, 5.1.4]]

    #+BEGIN_QUOTE
    5.1.4.  Alternative Subtype

       The "multipart/alternative" type is syntactically identical to
       "multipart/mixed", but the semantics are different.  In particular,
       each of the body parts is an "alternative" version of the same
       information.

       Systems should recognize that the content of the various parts are
       interchangeable.  Systems should choose the "best" type based on the
       local environment and references, in some cases even through user
       interaction.  As with "multipart/mixed", the order of body parts is
       significant.  In this case, the alternatives appear in an order of
       increasing faithfulness to the original content.  In general, the
       best choice is the LAST part of a type supported by the recipient
       system's local environment.
    #+END_QUOTE

    #+BEGIN_QUOTE
       In general, user agents that compose "multipart/alternative" entities
       must place the body parts in increasing order of preference, that is,
       with the preferred format last.  For fancy text, the sending user
       agent should put the plainest format first and the richest format
       last.  Receiving user agents should pick and display the last format
       they are capable of displaying.  In the case where one of the
       alternatives is itself of type "multipart" and contains unrecognized
       sub-parts, the user agent may choose either to show that alternative,
       an earlier alternative, or both.
    #+END_QUOTE
* Org mode issue tracking library

A collection of helper functions to maintain the [[file:org-issues.org][Issue file]] from within
Wanderlust and (partly) Gnus.

You can download a current version of this file [[file:code/elisp/org-issue.el][here]].

Currently following commands are provided:

  - 'org-issue-new' :: File a new issue for current message

       Creates a new TODO in 'org-issue-issue-file' below the headline
       "New Issues" with keyword NEW.  If customization variable
       'org-issue-message-flag' is non-nil and flagging messages is
       supported, the message of this issue is flagged.

  - 'org-issue-close' :: Close issue of current message.

  - 'org-issue-tag'  :: Tag issue of current message.

  - 'org-issue-update-message-flag' :: Update message flag according
       to issue file.

       If the issue for current message is closed, the message flag is
       removed.

  - 'org-issue-link-gmane' :: An Org mode web link pointing to current
       message on gmane is pushed to killring and clipboard.