Emacs MIME

1.1 Dissection

The mm-dissect-buffer is the function responsible for dissecting a MIME article. If given a multipart message, it will recursively descend the message, following the structure, and return a tree of MIME handles that describes the structure of the message.

1.2 Non-MIME

Gnus also understands some non-MIME attachments, such as postscript, uuencode, binhex, yenc, shar, forward, gnatsweb, pgp, diff. Each of these features can be disabled by add an item into mm-uu-configure-list. For example,

(require 'mm-uu)
(add-to-list 'mm-uu-configure-list '(pgp-signed . disabled))

postscript ¶

PostScript file.

uu ¶

Uuencoded file.

binhex ¶

Binhex encoded file.

yenc ¶

Yenc encoded file.

shar ¶

Shar archive file.

forward ¶

Non-MIME forwarded message.

gnatsweb ¶

Gnatsweb attachment.

pgp-signed ¶

PGP signed clear text.

pgp-encrypted ¶

PGP encrypted clear text.

pgp-key ¶

PGP public keys.

emacs-sources ¶

Emacs source code. This item works only in the groups matching mm-uu-emacs-sources-regexp.

diff ¶

Patches. This is intended for groups where diffs of committed files are automatically sent to. It only works in groups matching mm-uu-diff-groups-regexp.

verbatim-marks ¶

Slrn-style verbatim marks.

LaTeX ¶

LaTeX documents. It only works in groups matching mm-uu-tex-groups-regexp.

Some inlined non-MIME attachments are displayed using the face mm-uu-extract. By default, no MIME button for these parts is displayed. You can force displaying a button using K b (gnus-summary-display-buttonized) or add text/x-verbatim to gnus-buttonized-mime-types, See MIME Commands in Gnus Manual.

1.3 Handles

A MIME handle is a list that fully describes a MIME component.

The following macros can be used to access elements in a handle:

mm-handle-buffer ¶

Return the buffer that holds the contents of the undecoded MIME part.

mm-handle-type ¶

Return the parsed Content-Type of the part.

mm-handle-encoding ¶

Return the Content-Transfer-Encoding of the part.

mm-handle-undisplayer ¶

Return the object that can be used to remove the displayed part (if it has been displayed).

mm-handle-set-undisplayer ¶

Set the undisplayer object.

mm-handle-disposition ¶

Return the parsed Content-Disposition of the part.

mm-get-content-id

Returns the handle(s) referred to by Content-ID.

1.4 Display

Functions for displaying, removing and saving.

mm-display-part ¶

Display the part.

mm-remove-part ¶

Remove the part (if it has been displayed).

mm-inlinable-p ¶

Say whether a MIME type can be displayed inline.

mm-automatic-display-p ¶

Say whether a MIME type should be displayed automatically.

mm-destroy-part ¶

Free all resources occupied by a part.

mm-save-part ¶

Offer to save the part in a file.

mm-pipe-part ¶

Offer to pipe the part to some process.

mm-interactively-view-part ¶

Prompt for a mailcap method to use to view the part.

1.5 Display Customization

mm-inline-media-tests ¶

This is an alist where the key is a MIME type, the second element is a function to display the part inline (i.e., inside Emacs), and the third element is a form to be evaled to say whether the part can be displayed inline.

This variable specifies whether a part can be displayed inline, and, if so, how to do it. It does not say whether parts are actually displayed inline.

mm-inlined-types ¶

This, on the other hand, says what types are to be displayed inline, if they satisfy the conditions set by the variable above. It’s a list of MIME media types.

mm-automatic-display ¶

This is a list of types that are to be displayed “automatically”, but only if the above variable allows it. That is, only inlinable parts can be displayed automatically.

mm-automatic-external-display ¶

This is a list of types that will be displayed automatically in an external viewer.

mm-keep-viewer-alive-types ¶

This is a list of media types for which the external viewer will not be killed when selecting a different article.

mm-attachment-override-types ¶

Some MIME agents create parts that have a content-disposition of ‘attachment’. This variable allows overriding that disposition and displaying the part inline. (Note that the disposition is only overridden if we are able to, and want to, display the part inline.)

mm-discouraged-alternatives ¶

List of MIME types that are discouraged when viewing ‘multipart/alternative’. Viewing agents are supposed to view the last possible part of a message, as that is supposed to be the richest. However, users may prefer other types instead, and this list says what types are most unwanted. If, for instance, ‘text/html’ parts are very unwanted, and ‘text/richtext’ parts are somewhat unwanted, you could say something like:

(setq mm-discouraged-alternatives
      '("text/html" "text/richtext")
      mm-automatic-display
      (remove "text/html" mm-automatic-display))

Adding "image/.*" might also be useful. Spammers use images as the preferred part of ‘multipart/alternative’ messages, so you might not notice there are other parts. See also gnus-buttonized-mime-types, MIME Commands in Gnus Manual. After adding "multipart/alternative" to gnus-buttonized-mime-types you can choose manually which alternative you’d like to view. For example, you can set those variables like:

(setq gnus-buttonized-mime-types
      '("multipart/alternative" "multipart/signed")
      mm-discouraged-alternatives
      '("text/html" "image/.*"))

In this case, Gnus will display radio buttons for such a kind of spam message as follows:

1.  (*) multipart/alternative  ( ) image/gif

2.  (*) text/plain          ( ) text/html

mm-inline-large-images ¶

This variable is resize by default, which means that images that are bigger than the Emacs window are resized so that they fit. If you set this to nil, large images are not displayed in Emacs, but can instead be displayed externally (e.g., with ‘ImageMagick’ or ‘xv’). Setting this variable to t disables this check and makes the library display all inline images as inline, regardless of their size.

mm-inline-large-images-proportion ¶

The proportion used when resizing large images.

mm-inline-override-types ¶

mm-inlined-types may include regular expressions, for example to specify that all ‘text/.*’ parts be displayed inline. If a user prefers to have a type that matches such a regular expression be treated as an attachment, that can be accomplished by setting this variable to a list containing that type. For example assuming mm-inlined-types includes ‘text/.*’, then including ‘text/html’ in this variable will cause ‘text/html’ parts to be treated as attachments.

mm-text-html-renderer ¶

This selects the function used to render HTML. The predefined renderers are selected by the symbols shr, gnus-w3m, w3m¹, links, lynx, or w3m-standalone. You can also specify a function, which will be called with a MIME handle as the argument.

mm-html-inhibit-images ¶

If this is non-nil, inhibit displaying of images inline in the article body. It is effective to images in HTML articles rendered when mm-text-html-renderer (see Display Customization) is shr or w3m. In Gnus, this is overridden by the value of gnus-inhibit-images (see Misc Article in Gnus manual). The default is nil.

mm-html-blocked-images ¶

External images that have URLs that match this regexp won’t be fetched and displayed. For instance, to block all URLs that have the string “ads” in them, do the following:

(setq mm-html-blocked-images "ads")

It is effective when mm-text-html-renderer (see Display Customization) is shr. In Gnus, this is overridden by the value of gnus-blocked-images or the return value of the function that gnus-blocked-images is set to (see HTML in Gnus manual).

Some HTML mails might have the trick of spammers using ‘<img>’ tags. It is likely to be intended to verify whether you have read the mail. You can prevent your personal information from leaking by setting this option to "" (which is the default).

mm-w3m-safe-url-regexp ¶

A regular expression that matches safe URL names, i.e., URLs that are unlikely to leak personal information when rendering HTML email (the default value is ‘\\`cid:’). If nil consider all URLs safe. In Gnus, this will be overridden according to the value of the variable gnus-safe-html-newsgroups, See Various Various in Gnus Manual.

mm-inline-text-html-with-w3m-keymap ¶

You can use emacs-w3m command keys in the inlined text/html part by setting this option to non-nil. The default value is t.

mm-external-terminal-program ¶

This should be a list of strings; typically something like ‘("xterm" "-e")’ or ‘("gnome-terminal" "--")’.

mm-enable-external ¶

Indicate whether external MIME handlers should be used.

If t, all defined external MIME handlers are used. If nil, files are saved to disk (mailcap-save-binary-file). If it is the symbol ask, you are prompted before the external MIME handler is invoked.

When you launch an attachment through mailcap (see mailcap) an attempt is made to use a safe viewer with the safest options—this isn’t the case if you save it to disk and launch it in a different way (command line or double-clicking). Anyhow, if you want to be sure not to launch any external programs, set this variable to nil or ask.

mm-inline-font-lock ¶

If non-nil, inlined parts that support font locking (for instance, patches or code snippets) will be font-locked. This may be overridden by callers that have their own ways of enabling/inhibiting font locking.

1.6 Files and Directories

mm-default-directory ¶

The default directory for saving attachments. If nil use default-directory.

mm-tmp-directory ¶

Directory for storing temporary files.

mm-file-name-rewrite-functions ¶

A list of functions used for rewriting file names of MIME parts. Each function is applied successively to the file name. Ready-made functions include

mm-file-name-delete-control ¶

Delete all control characters.

mm-file-name-delete-gotchas ¶

Delete characters that could have unintended consequences when used with flawed shell scripts, i.e., ‘|’, ‘>’ and ‘<’; and ‘-’, ‘.’ as the first character.

mm-file-name-delete-whitespace ¶

Remove all whitespace.

mm-file-name-trim-whitespace ¶

Remove leading and trailing whitespace.

mm-file-name-collapse-whitespace ¶

Collapse multiple whitespace characters.

mm-file-name-replace-whitespace ¶

Replace whitespace with underscores. Set the variable mm-file-name-replace-whitespace to any other string if you do not like underscores.

The standard Emacs functions capitalize, downcase, upcase and upcase-initials might also prove useful.

mm-path-name-rewrite-functions ¶

List of functions used for rewriting the full file names of MIME parts. This is used when viewing parts externally, and is meant for transforming the absolute name so that non-compliant programs can find the file where it’s saved.

1.7 New Viewers

Here’s an example viewer for displaying text/enriched inline:

(defun mm-display-enriched-inline (handle)
  (let (text)
    (with-temp-buffer
      (mm-insert-part handle)
      (save-window-excursion
        (enriched-decode (point-min) (point-max))
        (setq text (buffer-string))))
    (mm-insert-inline handle text)))

We see that the function takes a MIME handle as its parameter. It then goes to a temporary buffer, inserts the text of the part, does some work on the text, stores the result, goes back to the buffer it was called from and inserts the result.

The two important helper functions here are mm-insert-part and mm-insert-inline. The first function inserts the text of the handle in the current buffer. It handles charset and/or content transfer decoding. The second function just inserts whatever text you tell it to insert, but it also sets things up so that the text can be “undisplayed” in a convenient manner.

2.1 Simple MML Example

Here’s a simple ‘multipart/alternative’:

<#multipart type=alternative>
This is a plain text part.
<#part type=text/enriched>
<center>This is a centered enriched part</center>
<#/multipart>

After running this through mml-generate-mime, we get this:

Content-Type: multipart/alternative; boundary="=-=-="


--=-=-=


This is a plain text part.

--=-=-=
Content-Type: text/enriched


<center>This is a centered enriched part</center>

--=-=-=--

2.2 MML Definition

The MML language is very simple. It looks a bit like an SGML application, but it’s not.

The main concept of MML is the part. Each part can be of a different type or use a different charset. The way to delineate a part is with a ‘<#part ...>’ tag. Multipart parts can be introduced with the ‘<#multipart ...>’ tag. Parts are ended by the ‘<#/part>’ or ‘<#/multipart>’ tags. Parts started with the ‘<#part ...>’ tags are also closed by the next open tag.

There’s also the ‘<#external ...>’ tag. These introduce ‘external/message-body’ parts.

Each tag can contain zero or more parameters on the form ‘parameter=value’. The values may be enclosed in quotation marks, but that’s not necessary unless the value contains white space. So ‘filename=/home/user/#hello$^yes’ is perfectly valid.

If you want to talk about MML in a message, you need a way to “quote” these tags. The way to do that is to include an exclamation point after the opening two characters; i. e. ‘<#!part ...>’.

The following parameters have meaning in MML; parameters that have no meaning are ignored. The MML parameter names are the same as the MIME parameter names; the things in the parentheses say which header it will be used in.

‘type’

The MIME type of the part (Content-Type).

‘filename’

Use the contents of the file in the body of the part (Content-Disposition).

‘recipient-filename’

Use this as the file name in the generated MIME message for the recipient. That is, even if the file is called foo.txt locally, use this name instead in the Content-Disposition in the sent message.

‘charset’

The contents of the body of the part are to be encoded in the character set specified (Content-Type). See Charset Translation.

‘name’

Might be used to suggest a file name if the part is to be saved to a file (Content-Type).

‘disposition’

Valid values are ‘inline’ and ‘attachment’ (Content-Disposition).

‘encoding’

Valid values are ‘7bit’, ‘8bit’, ‘quoted-printable’ and ‘base64’. See Charset Translation. This parameter says what Content-Transfer-Encoding to use when sending the part, and is normally computed automatically.

‘data-encoding’

This parameter says what encoding has been used on the data, and the data will be decoded before use. Valid values are ‘quoted-printable’ and ‘base64’. This is useful when you have a part with binary data (for instance an image) inserted directly into the Message buffer inside the ‘"<#part>...<#/part>"’ tags.

‘description’

A description of the part (Content-Description).

‘creation-date’

Date when the part was created (Content-Disposition). This uses the format of RFC 822 or its successors.

‘modification-date’

RFC 822 (or later) date when the part was modified (Content-Disposition).

‘read-date’

RFC 822 (or later) date when the part was read (Content-Disposition).

‘recipients’

Who to encrypt/sign the part to. This field is used to override any auto-detection based on the To/Cc headers.

‘sender’

Identity used to sign the part. This field is used to override the default key used.

‘size’

The size (in octets) of the part (Content-Disposition).

‘sign’

What technology to sign this MML part with (smime, pgp or pgpmime)

‘encrypt’

What technology to encrypt this MML part with (smime, pgp or pgpmime)

Parameters for ‘text/plain’:

‘format’

Formatting parameter for the text, valid values include ‘fixed’ (the default) and ‘flowed’. Normally you do not specify this manually, since it requires the textual body to be formatted in a special way described in RFC 2646. See Flowed text.

Parameters for ‘application/octet-stream’:

‘type’

Type of the part; informal—meant for human readers (Content-Type).

Parameters for ‘message/external-body’:

‘access-type’

A word indicating the supported access mechanism by which the file may be obtained. Values include ‘ftp’, ‘anon-ftp’, ‘tftp’, ‘localfile’, and ‘mailserver’. (Content-Type.)

‘expiration’

RFC 822 (or later) date after which the file may no longer be fetched. (Content-Type.)

‘size’

The size (in octets) of the file. (Content-Type.)

‘permission’

Valid values are ‘read’ and ‘read-write’ (Content-Type).

Parameters for ‘sign=smime’:

‘keyfile’

File containing key and certificate for signer.

Parameters for ‘encrypt=smime’:

‘certfile’

File containing certificate for recipient.

2.3 Advanced MML Example

Here’s a complex multipart message. It’s a ‘multipart/mixed’ that contains many parts, one of which is a ‘multipart/alternative’.

<#multipart type=mixed>
<#part type=image/jpeg filename=~/rms.jpg disposition=inline>
<#multipart type=alternative>
This is a plain text part.
<#part type=text/enriched name=enriched.txt>
<center>This is a centered enriched part</center>
<#/multipart>
This is a new plain text part.
<#part disposition=attachment>
This plain text part is an attachment.
<#/multipart>

And this is the resulting MIME message:

Content-Type: multipart/mixed; boundary="=-=-="


--=-=-=



--=-=-=
Content-Type: image/jpeg;
 filename="~/rms.jpg"
Content-Disposition: inline;
 filename="~/rms.jpg"
Content-Transfer-Encoding: base64
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--=-=-=
Content-Type: multipart/alternative; boundary="==-=-="


--==-=-=


This is a plain text part.

--==-=-=
Content-Type: text/enriched;
 name="enriched.txt"


<center>This is a centered enriched part</center>

--==-=-=--

--=-=-=

This is a new plain text part.

--=-=-=
Content-Disposition: attachment


This plain text part is an attachment.

--=-=-=--

2.4 Encoding Customization

mm-body-charset-encoding-alist ¶

Mapping from MIME charset to encoding to use. This variable is usually used except, e.g., when other requirements force a specific encoding (digitally signed messages require 7bit encodings). The default is

((iso-2022-jp . 7bit)
 (iso-2022-jp-2 . 7bit)
 (utf-16 . base64)
 (utf-16be . base64)
 (utf-16le . base64))

As an example, if you do not want to have ISO-8859-1 characters quoted-printable encoded, you may add (iso-8859-1 . 8bit) to this variable. You can override this setting on a per-message basis by using the encoding MML tag (see MML Definition).

mm-coding-system-priorities ¶

Prioritize coding systems to use for outgoing messages. The default is nil, which means to use the defaults in Emacs, but is (iso-8859-1 iso-2022-jp utf-8) when running Emacs in the Japanese language environment. It is a list of coding system symbols (aliases of coding systems are also allowed, use M-x describe-coding-system to make sure you are specifying correct coding system names). For example, if you have configured Emacs to prefer UTF-8, but wish that outgoing messages should be sent in ISO-8859-1 if possible, you can set this variable to (iso-8859-1). You can override this setting on a per-message basis by using the charset MML tag (see MML Definition).

As different hierarchies prefer different charsets, you may want to set mm-coding-system-priorities according to the hierarchy in Gnus. Here’s an example:

(add-to-list 'gnus-newsgroup-variables 'mm-coding-system-priorities)
(setq gnus-parameters
      (append
       ;; Some charsets are just examples!
       '(("^cn\\." ;; Chinese
          (mm-coding-system-priorities
           '(iso-8859-1 cn-big5 chinese-iso-7bit utf-8)))
         ("^cz\\.\\|^pl\\." ;; Central and Eastern European
          (mm-coding-system-priorities '(iso-8859-2 utf-8)))
         ("^de\\." ;; German language
          (mm-coding-system-priorities '(iso-8859-1 iso-8859-15 utf-8)))
         ("^fr\\." ;; French
          (mm-coding-system-priorities '(iso-8859-15 iso-8859-1 utf-8)))
         ("^fj\\." ;; Japanese
          (mm-coding-system-priorities
           '(iso-8859-1 iso-2022-jp utf-8)))
         ("^ru\\." ;; Cyrillic
          (mm-coding-system-priorities
           '(koi8-r iso-8859-5 iso-8859-1 utf-8))))
       gnus-parameters))

mm-content-transfer-encoding-defaults ¶

Mapping from MIME types to encoding to use. This variable is usually used except, e.g., when other requirements force a safer encoding (digitally signed messages require 7bit encoding). Besides the normal MIME encodings, qp-or-base64 may be used to indicate that for each case the most efficient of quoted-printable and base64 should be used.

qp-or-base64 has another effect. It will fold long lines so that MIME parts may not be broken by MTA. So do quoted-printable and base64.

Note that it affects body encoding only when a part is a raw forwarded message (which will be made by gnus-summary-mail-forward with the arg 2 for example) or is neither the ‘text/*’ type nor the ‘message/*’ type. Even though in those cases, you can override this setting on a per-message basis by using the encoding MML tag (see MML Definition).

mm-use-ultra-safe-encoding ¶

When this is non-nil, it means that textual parts are encoded as quoted-printable if they contain lines longer than 76 characters or starting with "From " in the body. Non-7bit encodings (8bit, binary) are generally disallowed. This reduce the probability that a non-8bit clean MTA or MDA changes the message. This should never be set directly, but bound by other functions when necessary (e.g., when encoding messages that are to be digitally signed).

2.5 Charset Translation

During translation from MML to MIME, for each MIME part which has been composed inside Emacs, an appropriate charset has to be chosen.

If you are running a non-MULE Emacs, this process is simple: If the part contains any non-ASCII (8-bit) characters, the MIME charset given by mail-parse-charset (a symbol) is used. (Never set this variable directly, though. If you want to change the default charset, please consult the documentation of the package which you use to process MIME messages. See Various Message Variables in Message Manual, for example.) If there are only ASCII characters, the MIME charset US-ASCII is used, of course.

Things are slightly more complicated when running Emacs with MULE support. In this case, a list of the MULE charsets used in the part is obtained, and the MULE charsets are translated to MIME charsets by consulting the table provided by Emacs itself. If this results in a single MIME charset, this is used to encode the part. But if the resulting list of MIME charsets contains more than one element, two things can happen: If it is possible to encode the part via UTF-8, this charset is used. (For this, Emacs must support the utf-8 coding system, and the part must consist entirely of characters which have Unicode counterparts.) If UTF-8 is not available for some reason, the part is split into several ones, so that each one can be encoded with a single MIME charset. The part can only be split at line boundaries, though—if more than one MIME charset is required to encode a single line, it is not possible to encode the part.

When running Emacs with MULE support, the preferences for which coding system to use is inherited from Emacs itself. This means that if Emacs is set up to prefer UTF-8, it will be used when encoding messages. You can modify this by altering the mm-coding-system-priorities variable though (see Encoding Customization).

The charset to be used can be overridden by setting the charset MML tag (see MML Definition) when composing the message.

The encoding of characters (quoted-printable, 8bit, etc.) is orthogonal to the discussion here, and is controlled by the variables mm-body-charset-encoding-alist and mm-content-transfer-encoding-defaults (see Encoding Customization).

2.6 Conversion

A (multipart) MIME message can be converted to MML with the mime-to-mml function. It works on the message in the current buffer, and substitutes MML markup for MIME boundaries. Non-textual parts do not have their contents in the buffer, but instead have the contents in separate buffers that are referred to from the MML tags.

An MML message can be converted back to MIME by the mml-to-mime function.

These functions are in certain senses “lossy”—you will not get back an identical message if you run mime-to-mml and then mml-to-mime. Not only will trivial things like the order of the headers differ, but the contents of the headers may also be different. For instance, the original message may use base64 encoding on text, while mml-to-mime may decide to use quoted-printable encoding, and so on.

In essence, however, these two functions should be the inverse of each other. The resulting contents of the message should remain equivalent, if not identical.

2.7 Flowed text

The Emacs MIME library will respect the use-hard-newlines variable (see Hard and Soft Newlines in Emacs Manual) when encoding a message, and the “format=flowed” Content-Type parameter when decoding a message.

On encoding text, regardless of use-hard-newlines, lines terminated by soft newline characters are filled together and wrapped after the column decided by fill-flowed-encode-column. Quotation marks (matching ‘^>* ?’) are respected. The variable controls how the text will look in a client that does not support flowed text, the default is to wrap after 66 characters. If hard newline characters are not present in the buffer, no flow encoding occurs.

You can customize the value of the mml-enable-flowed variable to enable or disable the flowed encoding usage when newline characters are present in the buffer.

On decoding flowed text, lines with soft newline characters are filled together and wrapped after the column decided by fill-flowed-display-column. The default is to wrap after fill-column.

mm-fill-flowed ¶

If non-nil a format=flowed article will be displayed flowed.

4.1 rfc2045

RFC2045 is the “main” MIME document, and as such, one would imagine that there would be a lot to implement. But there isn’t, since most of the implementation details are delegated to the subsequent RFCs.

So rfc2045.el has only a single function:

rfc2045-encode-string ¶

Takes a parameter and a value and returns a ‘PARAM=VALUE’ string. value will be quoted if there are non-safe characters in it.

4.2 rfc2231

RFC2231 defines a syntax for the Content-Type and Content-Disposition headers. Its snappy name is MIME Parameter Value and Encoded Word Extensions: Character Sets, Languages, and Continuations.

In short, these headers look something like this:

Content-Type: application/x-stuff;
 title*0*=us-ascii'en'This%20is%20even%20more%20;
 title*1*=%2A%2A%2Afun%2A%2A%2A%20;
 title*2="isn't it!"

They usually aren’t this bad, though.

The following functions are defined by this library:

rfc2231-parse-string ¶

Parse a Content-Type header and return a list describing its elements.

(rfc2231-parse-string
 "application/x-stuff;
 title*0*=us-ascii'en'This%20is%20even%20more%20;
 title*1*=%2A%2A%2Afun%2A%2A%2A%20;
 title*2=\"isn't it!\"")
⇒ ("application/x-stuff"
    (title . "This is even more ***fun*** isn't it!"))

rfc2231-get-value ¶

Takes one of the lists on the format above and returns the value of the specified attribute.

rfc2231-encode-string ¶

Encode a parameter in headers likes Content-Type and Content-Disposition.

4.3 ietf-drums

drums was an IETF working group that worked on Internet RFC 2822, the first successor to RFC 822 and a predecessor of the current email standard.

The functions provided by this library include:

ietf-drums-remove-comments ¶

Remove the comments from the argument and return the results.

ietf-drums-remove-whitespace ¶

Remove linear white space from the string and return the results. Spaces inside quoted strings and comments are left untouched.

ietf-drums-get-comment ¶

Return the last most comment from the string.

ietf-drums-parse-address ¶

Parse an address string and return a list that contains the mailbox and the plain text name.

ietf-drums-parse-addresses ¶

Parse a string that contains any number of comma-separated addresses and return a list that contains mailbox/plain text pairs.

ietf-drums-parse-date ¶

Parse a date string and return an Emacs time structure.

ietf-drums-narrow-to-header ¶

Narrow the buffer to the header section of the current buffer.

4.4 rfc2047

RFC2047 (Message Header Extensions for Non-ASCII Text) specifies how non-ASCII text in headers are to be encoded. This is actually rather complicated, so a number of variables are necessary to tweak what this library does.

The following variables are tweakable:

rfc2047-header-encoding-alist ¶

This is an alist of header / encoding-type pairs. Its main purpose is to prevent encoding of certain headers.

The keys can either be header regexps, or t.

The values can be nil, in which case the header(s) in question won’t be encoded, mime, which means that they will be encoded, or address-mime, which means the header(s) will be encoded carefully assuming they contain addresses.

rfc2047-charset-encoding-alist ¶

RFC2047 specifies two forms of encoding—Q (a Quoted-Printable-like encoding) and B (base64). This alist specifies which charset should use which encoding.

rfc2047-encode-function-alist ¶

This is an alist of encoding / function pairs. The encodings are Q, B and nil.

rfc2047-encoded-word-regexp ¶

When decoding words, this library looks for matches to this regexp.

rfc2047-encoded-word-regexp-loose ¶

This is a version from which the regexp for the Q encoding pattern of rfc2047-encoded-word-regexp is made loose.

rfc2047-encode-encoded-words ¶

The boolean variable specifies whether encoded words (e.g., ‘=?us-ascii?q?hello?=’) should be encoded again. rfc2047-encoded-word-regexp is used to look for such words.

rfc2047-allow-irregular-q-encoded-words ¶

The boolean variable specifies whether irregular Q encoded words (e.g., ‘=?us-ascii?q?hello??=’) should be decoded. If it is non-nil, rfc2047-encoded-word-regexp-loose is used instead of rfc2047-encoded-word-regexp to look for encoded words.

Those were the variables, and these are this functions:

rfc2047-narrow-to-field ¶

Narrow the buffer to the header on the current line.

rfc2047-encode-message-header ¶

Should be called narrowed to the header of a message. Encodes according to rfc2047-header-encoding-alist.

rfc2047-encode-region ¶

Encodes all encodable words in the region specified.

rfc2047-encode-string ¶

Encode a string and return the results.

rfc2047-decode-region ¶

Decode the encoded words in the region.

rfc2047-decode-string ¶

Decode a string and return the results.

rfc2047-encode-parameter ¶

Encode a parameter in the RFC2047-like style. This is a substitution for the rfc2231-encode-string function, that is the standard but many mailers don’t support it. See rfc2231.

4.5 time-date

While not really a part of the MIME library, it is convenient to document time conversion functions often used when parsing Date headers and manipulating time. (Not by using tesseracts, though, I’m sorry to say.)

These functions convert between five formats: A date string, a Lisp timestamp, a decoded time list, a second number, and a day number.

Here’s a bunch of time/date/second/day examples:

(parse-time-string "Sat Sep 12 12:21:54 1998 +0200")
⇒ (54 21 12 12 9 1998 6 -1 7200)

(time-convert
  (date-to-time "Sat Sep 12 12:21:54 1998 +0200")
  'integer)
⇒ 905595714

(float-time '(905595714000000 . 1000000))
⇒ 905595714.0

(time-convert 905595714.0 1000000)
⇒ (905595714000000 . 1000000)

(time-to-days '(905595714000000 . 1000000))
⇒ 729644

(time-convert (days-to-time 729644) 'integer)
⇒ 63041241600

(time-convert (time-since '(905595714000000 . 1000000))
              1000000)
⇒ (631963244775642171 . 1000000000)

(time-less-p '(905595714000000 . 1000000)
             '(905595593000000000 . 1000000000))
⇒ nil

(time-equal-p '(905595593000000000 . 1000000000)
              '(905595593000000    . 1000000   ))
⇒ t

(time-subtract '(905595714000000 . 1000000)
               '(905595593000000000 . 1000000000))
⇒ (121000000 . 1000000)

(days-between "Sat Sep 12 12:21:54 1998 +0200"
              "Sat Sep 07 12:21:54 1998 +0200")
⇒ 5

(date-leap-year-p 2000)
⇒ t

(time-to-day-in-year '(905595714000000 . 1000000))
⇒ 255

(time-to-number-of-days
 (time-since
  (date-to-time "Mon, 01 Jan 2001 02:22:26 GMT")))
⇒ 6472.722661506652

And finally, we have safe-date-to-time, which does the same as date-to-time, but returns a zero time if the date is syntactically malformed.

The five data representations used are the following:

date

An RFC 822 (or similar) date string. For instance: "Sat Sep 12 12:21:54 1998 +0200".

time

A Lisp timestamp. For instance: (905595714000000 . 1000000).

seconds

An integer or floating point count of seconds. For instance: 905595714.0, 905595714.

days

An integer number representing the number of days since Sunday, December 31, 1 BC (Gregorian). For instance: 729644.

decoded time

A list of decoded time. For instance: (54 21 12 12 9 1998 6 nil 7200).

All the examples above represent the same moment, except that days represents the day containing the moment.

These are the functions available:

date-to-time

Take a date and return a time.

time-convert

Take a time and return a timestamp in a specified form.

float-time

Take a time and return seconds.

encode-time

Take a decoded time and return a timestamp.

time-to-days

Take a time and return days.

days-to-time

Take days and return a time.

date-to-day

Take a date and return days.

time-to-number-of-days

Take a time and return the number of days that represents.

safe-date-to-time

Take a date and return a time. If the date is not syntactically valid, return a “zero” time.

time-less-p

Take two times and say whether the first time is less (i.e., earlier) than the second time. (This is a built-in function.)

time-equal-p

Check whether two time values are equal. The time values need not be in the same format. (This is a built-in function.)

time-since

Take a time and return a time saying how long it was since that time.

time-subtract

Take two times and subtract the second from the first. I.e., return the time between the two times. (This is a built-in function.)

days-between

Take two days and return the number of days between those two days.

date-leap-year-p

Take a year number and say whether it’s a leap year.

time-to-day-in-year

Take a time and return the day number within the year that the time is in.

4.6 qp

This library deals with decoding and encoding Quoted-Printable text.

Very briefly explained, qp encoding means translating all 8-bit characters (and lots of control characters) into things that look like ‘=EF’; that is, an equal sign followed by the byte encoded as a hex string.

The following functions are defined by the library:

quoted-printable-decode-region ¶

QP-decode all the encoded text in the specified region.

quoted-printable-decode-string ¶

Decode the QP-encoded text in a string and return the results.

quoted-printable-encode-region ¶

QP-encode all the encodable characters in the specified region. The third optional parameter fold specifies whether to fold long lines. (Long here means 72.)

quoted-printable-encode-string ¶

QP-encode all the encodable characters in a string and return the results.

4.7 base64

Base64 is an encoding that encodes three bytes into four characters, thereby increasing the size by about 33%. The alphabet used for encoding is very resistant to mangling during transit.

The following functions are defined by this library:

base64-encode-region ¶

base64 encode the selected region. Return the length of the encoded text. Optional third argument no-line-break means do not break long lines into shorter lines.

base64-encode-string ¶

base64 encode a string and return the result.

base64-decode-region ¶

base64 decode the selected region. Return the length of the decoded text. If the region can’t be decoded, return nil and don’t modify the buffer.

base64-decode-string ¶

base64 decode a string and return the result. If the string can’t be decoded, nil is returned.

4.8 binhex

binhex is an encoding that originated in Macintosh environments. The following function is supplied to deal with these:

binhex-decode-region ¶

Decode the encoded text in the region. If given a third parameter, only decode the binhex header and return the filename.

4.9 uudecode

uuencode is probably still the most popular encoding of binaries used on Usenet, although base64 rules the mail world.

The following function is supplied by this package:

uudecode-decode-region ¶

Decode the text in the region.

4.10 yenc

yenc is used for encoding binaries on Usenet. The following function is supplied by this package:

yenc-decode-region ¶

Decode the encoded text in the region.

4.11 rfc1843

RFC1843 deals with mixing Chinese and ASCII characters in messages. In essence, RFC1843 switches between ASCII and Chinese by doing this:

This sentence is in ASCII.
The next sentence is in GB.~{<:Ky2;S{#,NpJ)l6HK!#~}Bye.

Simple enough, and widely used in China.

The following functions are available to handle this encoding:

rfc1843-decode-region

Decode HZ-encoded text in the region.

rfc1843-decode-string

Decode a HZ-encoded string and return the result.

4.12 mailcap

The ~/.mailcap file is parsed by most MIME-aware message handlers and describes how elements are supposed to be displayed. Here’s an example file:

image/*; gimp -8 %s
audio/wav; wavplayer %s
application/msword; catdoc %s ; copiousoutput ; nametemplate=%s.doc

This says that all image files should be displayed with gimp, that WAVE audio files should be played by wavplayer, and that MS-WORD files should be inlined by catdoc.

The mailcap library parses this file, and provides functions for matching types.

mailcap-mime-data ¶

This variable is an alist of alists containing backup viewing rules.

mailcap-user-mime-data ¶

A customizable list of viewers that take preference over mailcap-mime-data.

Interface functions:

mailcap-view-file ¶

Prompt for a file name, and start a viewer applicable for the file type in question.

mailcap-parse-mailcaps ¶

Parse the ~/.mailcap file.

mailcap-mime-info

Takes a MIME type as its argument and returns the matching viewer.

The mailcap-prefer-mailcap-viewers variable controls which viewer is chosen. The default non-nil value means that settings from ~/.mailcap is preferred over system-wide or Emacs-provided viewer settings.

If nil, Emacs-provided viewer settings have precedence. Next, the most specific viewer has precedence over less specific settings, no matter if they’re system-provided or private, so ‘image/gif’ in /etc/mailcap will “win” over an ‘image/*’ setting in ~/.mailcap.