10. Appendices

10.1 XEmacs

XEmacs is distributed as a collection of packages. You should install whatever packages the Gnus XEmacs package requires. The current requirements are `gnus', `mail-lib', `xemacs-base', `eterm', `sh-script', `net-utils', `os-utils', `dired', `mh-e', `sieve', `ps-print', `W3', `pgg', `mailcrypt', `ecrypto', and `sasl'.

10.2 History

GNUS was written by Masanobu UMEDA. When autumn crept up in '94, Lars Magne Ingebrigtsen grew bored and decided to rewrite Gnus.

If you want to investigate the person responsible for this outrage, you can point your (feh!) web browser to http://quimby.gnus.org/. This is also the primary distribution point for the new and spiffy versions of Gnus, and is known as The Site That Destroys Newsrcs And Drives People Mad.

During the first extended alpha period of development, the new Gnus was called "(ding) Gnus". (ding) is, of course, short for ding is not Gnus, which is a total and utter lie, but who cares? (Besides, the "Gnus" in this abbreviation should probably be pronounced "news" as UMEDA intended, which makes it a more appropriate name, don't you think?)

In any case, after spending all that energy on coming up with a new and spunky name, we decided that the name was too spunky, so we renamed it back again to "Gnus". But in mixed case. "Gnus" vs. "GNUS". New vs. old.

10.2.1 Gnus Versions

The first "proper" release of Gnus 5 was done in November 1995 when it was included in the Emacs 19.30 distribution (132 (ding) Gnus releases plus 15 Gnus 5.0 releases).

In May 1996 the next Gnus generation (aka. "September Gnus" (after 99 releases)) was released under the name "Gnus 5.2" (40 releases).

On July 28th 1996 work on Red Gnus was begun, and it was released on January 25th 1997 (after 84 releases) as "Gnus 5.4" (67 releases).

On September 13th 1997, Quassia Gnus was started and lasted 37 releases. It was released as "Gnus 5.6" on March 8th 1998 (46 releases).

Gnus 5.6 begat Pterodactyl Gnus on August 29th 1998 and was released as "Gnus 5.8" (after 99 releases and a CVS repository) on December 3rd 1999.

On the 26th of October 2000, Oort Gnus was begun and was released as Gnus 5.10 on May 1st 2003 (24 releases).

On the January 4th 2004, No Gnus was begun.

If you happen upon a version of Gnus that has a prefixed name -- "(ding) Gnus", "September Gnus", "Red Gnus", "Quassia Gnus", "Pterodactyl Gnus", "Oort Gnus", "No Gnus" -- don't panic. Don't let it know that you're frightened. Back away. Slowly. Whatever you do, don't run. Walk away, calmly, until you're out of its reach. Find a proper released version of Gnus and snuggle up to that instead.

10.2.2 Other Gnus Versions

In addition to the versions of Gnus which have had their releases coordinated by Lars, one major development has been Semi-gnus from Japan. It's based on a library called SEMI, which provides MIME capabilities.

These Gnusae are based mainly on Gnus 5.6 and Pterodactyl Gnus. Collectively, they are called "Semi-gnus", and different strains are called T-gnus, ET-gnus, Nana-gnus and Chaos. These provide powerful MIME and multilingualization things, especially important for Japanese users.

10.2.3 Why?

What's the point of Gnus?

I want to provide a "rad", "happening", "way cool" and "hep" newsreader, that lets you do anything you can think of. That was my original motivation, but while working on Gnus, it has become clear to me that this generation of newsreaders really belong in the stone age. Newsreaders haven't developed much since the infancy of the net. If the volume continues to rise with the current rate of increase, all current newsreaders will be pretty much useless. How do you deal with newsgroups that have thousands of new articles each day? How do you keep track of millions of people who post?

Gnus offers no real solutions to these questions, but I would very much like to see Gnus being used as a testing ground for new methods of reading and fetching news. Expanding on UMEDA-san's wise decision to separate the newsreader from the back ends, Gnus now offers a simple interface for anybody who wants to write new back ends for fetching mail and news from different sources. I have added hooks for customizations everywhere I could imagine it being useful. By doing so, I'm inviting every one of you to explore and invent.

May Gnus never be complete. C-u 100 M-x all-hail-emacs and C-u 100 M-x all-hail-xemacs.

10.2.4 Compatibility

Gnus was designed to be fully compatible with GNUS. Almost all key bindings have been kept. More key bindings have been added, of course, but only in one or two obscure cases have old bindings been changed.

Our motto is:

In a cloud bones of steel.

All commands have kept their names. Some internal functions have changed their names.

The gnus-uu package has changed drastically. See section 3.16 Decoding Articles.

One major compatibility question is the presence of several summary buffers. All variables relevant while reading a group are buffer-local to the summary buffer they belong in. Although many important variables have their values copied into their global counterparts whenever a command is executed in the summary buffer, this change might lead to incorrect values being used unless you are careful.

All code that relies on knowledge of GNUS internals will probably fail. To take two examples: Sorting gnus-newsrc-alist (or changing it in any way, as a matter of fact) is strictly verboten. Gnus maintains a hash table that points to the entries in this alist (which speeds up many functions), and changing the alist directly will lead to peculiar results.

Old hilit19 code does not work at all. In fact, you should probably remove all hilit code from all Gnus hooks (gnus-group-prepare-hook and gnus-summary-prepare-hook). Gnus provides various integrated functions for highlighting. These are faster and more accurate. To make life easier for everybody, Gnus will by default remove all hilit calls from all hilit hooks. Uncleanliness! Away!

Packages like expire-kill will no longer work. As a matter of fact, you should probably remove all old GNUS packages (and other code) when you start using Gnus. More likely than not, Gnus already does what you have written code to make GNUS do. (Snicker.)

Even though old methods of doing things are still supported, only the new methods are documented in this manual. If you detect a new method of doing something while reading this manual, that does not mean you have to stop doing it the old way.

Gnus understands all GNUS startup files.

Overall, a casual user who hasn't written much code that depends on GNUS internals should suffer no problems. If problems occur, please let me know by issuing that magic command M-x gnus-bug.

If you are in the habit of sending bug reports very often, you may find the helpful help buffer annoying after a while. If so, set gnus-bug-create-help-buffer to nil to avoid having it pop up at you.

10.2.5 Conformity

No rebels without a clue here, ma'am. We conform to all standards known to (wo)man. Except for those standards and/or conventions we disagree with, of course.

RFC (2)822

There are no known breaches of this standard.

RFC 1036

There are no known breaches of this standard, either.

Son-of-RFC 1036

We do have some breaches to this one.

X-Newsreader

User-Agent

These are considered to be "vanity headers", while I consider them to be consumer information. After seeing so many badly formatted articles coming from tin and Netscape I know not to use either of those for posting articles. I would not have known that if it wasn't for the X-Newsreader header.

USEFOR

USEFOR is an IETF working group writing a successor to RFC 1036, based on Son-of-RFC 1036. They have produced a number of drafts proposing various changes to the format of news articles. The Gnus towers will look into implementing the changes when the draft is accepted as an RFC.

MIME - RFC 2045-2049 etc

All the various MIME RFCs are supported.

Disposition Notifications - RFC 2298

Message Mode is able to request notifications from the receiver.

PGP - RFC 1991 and RFC 2440

RFC 1991 is the original PGP message specification, published as an informational RFC. RFC 2440 was the follow-up, now called Open PGP, and put on the Standards Track. Both document a non-MIME aware PGP format. Gnus supports both encoding (signing and encryption) and decoding (verification and decryption).

PGP/MIME - RFC 2015/3156

RFC 2015 (superseded by 3156 which references RFC 2440 instead of RFC 1991) describes the MIME-wrapping around the RF 1991/2440 format. Gnus supports both encoding and decoding.

S/MIME - RFC 2633

RFC 2633 describes the S/MIME format.

IMAP - RFC 1730/2060, RFC 2195, RFC 2086, RFC 2359, RFC 2595, RFC 1731

RFC 1730 is IMAP version 4, updated somewhat by RFC 2060 (IMAP 4 revision 1). RFC 2195 describes CRAM-MD5 authentication for IMAP. RFC 2086 describes access control lists (ACLs) for IMAP. RFC 2359 describes a IMAP protocol enhancement. RFC 2595 describes the proper TLS integration (STARTTLS) with IMAP. RFC 1731 describes the GSSAPI/Kerberos4 mechanisms for IMAP.

If you ever notice Gnus acting non-compliant with regards to the texts mentioned above, don't hesitate to drop a note to Gnus Towers and let us know.

10.2.6 Emacsen

Gnus should work on:

• Emacs 20.7 and up.

• XEmacs 21.1 and up.

This Gnus version will absolutely not work on any Emacsen older than that. Not reliably, at least. Older versions of Gnus may work on older Emacs versions.

There are some vague differences between Gnus on the various platforms--XEmacs features more graphics (a logo and a toolbar)---but other than that, things should look pretty much the same under all Emacsen.

10.2.7 Gnus Development

Gnus is developed in a two-phased cycle. The first phase involves much discussion on the `ding@gnus.org' mailing list, where people propose changes and new features, post patches and new back ends. This phase is called the alpha phase, since the Gnusae released in this phase are alpha releases, or (perhaps more commonly in other circles) snapshots. During this phase, Gnus is assumed to be unstable and should not be used by casual users. Gnus alpha releases have names like "Red Gnus" and "Quassia Gnus".

After futzing around for 50-100 alpha releases, Gnus is declared frozen, and only bug fixes are applied. Gnus loses the prefix, and is called things like "Gnus 5.6.32" instead. Normal people are supposed to be able to use these, and these are mostly discussed on the `gnu.emacs.gnus' newsgroup.

Some variable defaults differ between alpha Gnusae and released Gnusae. In particular, mail-source-delete-incoming defaults to nil in alpha Gnusae and t in released Gnusae. This is to prevent lossage of mail if an alpha release hiccups while handling the mail.

The division of discussion between the ding mailing list and the Gnus newsgroup is not purely based on publicity concerns. It's true that having people write about the horrible things that an alpha Gnus release can do (sometimes) in a public forum may scare people off, but more importantly, talking about new experimental features that have been introduced may confuse casual users. New features are frequently introduced, fiddled with, and judged to be found wanting, and then either discarded or totally rewritten. People reading the mailing list usually keep up with these rapid changes, while people on the newsgroup can't be assumed to do so.

10.2.8 Contributors

The new Gnus version couldn't have been done without the help of all the people on the (ding) mailing list. Every day for over a year I have gotten billions of nice bug reports from them, filling me with joy, every single one of them. Smooches. The people on the list have been tried beyond endurance, what with my "oh, that's a neat idea <type type>, yup, I'll release it right away <ship off> no wait, that doesn't work at all <type type>, yup, I'll ship that one off right away <ship off> no, wait, that absolutely does not work" policy for releases. Micro$oft--bah. Amateurs. I'm much worse. (Or is that "worser"? "much worser"? "worsest"?)

I would like to take this opportunity to thank the Academy for... oops, wrong show.

• Masanobu UMEDA---the writer of the original GNUS.

• Shenghuo Zhu--uudecode.el, mm-uu.el, rfc1843.el, webmail.el, nnwarchive and many, many other things connected with MIME and other types of en/decoding, as well as general bug fixing, new functionality and stuff.

• Per Abrahamsen--custom, scoring, highlighting and SOUP code (as well as numerous other things).

• Luis Fernandes--design and graphics.

• Joe Reiss--creator of the smiley faces.

• Justin Sheehy--the FAQ maintainer.

• Erik Naggum--help, ideas, support, code and stuff.

• Wes Hardaker---`gnus-picon.el' and the manual section on picons (see section 8.17.4 Picons).

• Kim-Minh Kaplan--further work on the picon code.

• Brad Miller---`gnus-gl.el' and the GroupLens manual section (see section 7.15 GroupLens).

• Sudish Joseph--innumerable bug fixes.

• Ilja Weis---`gnus-topic.el'.

• Steven L. Baur--lots and lots and lots of bugs detections and fixes.

• Vladimir Alexiev--the refcard and reference booklets.

• Felix Lee & Jamie Zawinski--I stole some pieces from the XGnus distribution by Felix Lee and JWZ.

• Scott Byer---`nnfolder.el' enhancements & rewrite.

• Peter Mutsaers--orphan article scoring code.

• Ken Raeburn--POP mail support.

• Hallvard B Furuseth--various bits and pieces, especially dealing with .newsrc files.

• Brian Edmonds---`gnus-bbdb.el'.

• David Moore--rewrite of `nnvirtual.el' and many other things.

• Kevin Davidson--came up with the name ding, so blame him.

• François Pinard--many, many interesting and thorough bug reports, as well as autoconf support.

This manual was proof-read by Adrian Aichner, with Ricardo Nassif, Mark Borges, and Jost Krieger proof-reading parts of the manual.

The following people have contributed many patches and suggestions:

Christopher Davis, Andrew Eskilsson, Kai Grossjohann, Kevin Greiner, Jesper Harder, Paul Jarc, Simon Josefsson, David Kċgedal, Richard Pieri, Fabrice Popineau, Daniel Quinlan, Michael Shields, Reiner Steib, Jason L. Tibbitts, III, Jack Vinson, Katsumi Yamaoka, and Teodor Zlatanov.

Also thanks to the following for patches and stuff:

Jari Aalto, Adrian Aichner, Vladimir Alexiev, Russ Allbery, Peter Arius, Matt Armstrong, Marc Auslander, Miles Bader, Alexei V. Barantsev, Frank Bennett, Robert Bihlmeyer, Chris Bone, Mark Borges, Mark Boyns, Lance A. Brown, Rob Browning, Kees de Bruin, Martin Buchholz, Joe Buehler, Kevin Buhr, Alastair Burt, Joao Cachopo, Zlatko Calusic, Massimo Campostrini, Castor, David Charlap, Dan Christensen, Kevin Christian, Jae-you Chung, James H. Cloos, Jr., Laura Conrad, Michael R. Cook, Glenn Coombs, Andrew J. Cosgriff, Neil Crellin, Frank D. Cringle, Geoffrey T. Dairiki, Andre Deparade, Ulrik Dickow, Dave Disser, Rui-Tao Dong, Joev Dubach, Michael Welsh Duggan, Dave Edmondson, Paul Eggert, Mark W. Eichin, Karl Eichwalder, Enami Tsugutomo, Michael Ernst, Luc Van Eycken, Sam Falkner, Nelson Jose dos Santos Ferreira, Sigbjorn Finne, Sven Fischer, Paul Fisher, Decklin Foster, Gary D. Foster, Paul Franklin, Guy Geens, Arne Georg Gleditsch, David S. Goldberg, Michelangelo Grigni, Dale Hagglund, D. Hall, Magnus Hammerin, Kenichi Handa, Raja R. Harinath, Yoshiki Hayashi, P. E. Jareth Hein, Hisashige Kenji, Scott Hofmann, Marc Horowitz, Gunnar Horrigmo, Richard Hoskins, Brad Howes, Miguel de Icaza, François Felix Ingrand, Tatsuya Ichikawa, Ishikawa Ichiro, Lee Iverson, Iwamuro Motonori, Rajappa Iyer, Andreas Jaeger, Adam P. Jenkins, Randell Jesup, Fred Johansen, Gareth Jones, Greg Klanderman, Karl Kleinpaste, Michael Klingbeil, Peter Skov Knudsen, Shuhei Kobayashi, Petr Konecny, Koseki Yoshinori, Thor Kristoffersen, Jens Lautenbacher, Martin Larose, Seokchan Lee, Joerg Lenneis, Carsten Leonhardt, James LewisMoss, Christian Limpach, Markus Linnala, Dave Love, Mike McEwan, Tonny Madsen, Shlomo Mahlab, Nat Makarevitch, Istvan Marko, David Martin, Jason R. Mastaler, Gordon Matzigkeit, Timo Metzemakers, Richard Mlynarik, Lantz Moore, Morioka Tomohiko, Erik Toubro Nielsen, Hrvoje Niksic, Andy Norman, Fred Oberhauser, C. R. Oldham, Alexandre Oliva, Ken Olstad, Masaharu Onishi, Hideki Ono, Ettore Perazzoli, William Perry, Stephen Peters, Jens-Ulrik Holger Petersen, Ulrich Pfeifer, Matt Pharr, Andy Piper, John McClary Prevost, Bill Pringlemeir, Mike Pullen, Jim Radford, Colin Rafferty, Lasse Rasinen, Lars Balker Rasmussen, Joe Reiss, Renaud Rioboo, Roland B. Roberts, Bart Robinson, Christian von Roques, Markus Rost, Jason Rumney, Wolfgang Rupprecht, Jay Sachs, Dewey M. Sasser, Conrad Sauerwald, Loren Schall, Dan Schmidt, Ralph Schleicher, Philippe Schnoebelen, Andreas Schwab, Randal L. Schwartz, Danny Siu, Matt Simmons, Paul D. Smith, Jeff Sparkes, Toby Speight, Michael Sperber, Darren Stalder, Richard Stallman, Greg Stark, Sam Steingold, Paul Stevenson, Jonas Steverud, Paul Stodghill, Kiyokazu Suto, Kurt Swanson, Samuel Tardieu, Teddy, Chuck Thompson, Tozawa Akihiko, Philippe Troin, James Troup, Trung Tran-Duc, Jack Twilley, Aaron M. Ucko, Aki Vehtari, Didier Verna, Vladimir Volovich, Jan Vroonhof, Stefan Waldherr, Pete Ware, Barry A. Warsaw, Christoph Wedler, Joe Wells, Lee Willis, and Lloyd Zusman.

For a full overview of what each person has done, the ChangeLogs included in the Gnus alpha distributions should give ample reading (550kB and counting).

Apologies to everybody that I've forgotten, of which there are many, I'm sure.

Gee, that's quite a list of people. I guess that must mean that there actually are people who are using Gnus. Who'd'a thunk it!

10.2.9 New Features

These lists are, of course, just short overviews of the most important new features. No, really. There are tons more. Yes, we have feeping creaturism in full effect.

10.2.9.1 (ding) Gnus

New features in Gnus 5.0/5.1:

• The look of all buffers can be changed by setting format-like variables (see section 2.1 Group Buffer Format and see section 3.1 Summary Buffer Format).

• Local spool and several NNTP servers can be used at once (see section 6. Select Methods).

• You can combine groups into virtual groups (see section 6.7.1 Virtual Groups).

• You can read a number of different mail formats (see section 6.3 Getting Mail). All the mail back ends implement a convenient mail expiry scheme (see section 6.3.9 Expiring Mail).

• Gnus can use various strategies for gathering threads that have lost their roots (thereby gathering loose sub-threads into one thread) or it can go back and retrieve enough headers to build a complete thread (see section 3.9.1 Customizing Threading).

• Killed groups can be displayed in the group buffer, and you can read them as well (see section 2.11 Listing Groups).

• Gnus can do partial group updates--you do not have to retrieve the entire active file just to check for new articles in a few groups (see section 1.9 The Active File).

• Gnus implements a sliding scale of subscribedness to groups (see section 2.6 Group Levels).

• You can score articles according to any number of criteria (see section 7. Scoring). You can even get Gnus to find out how to score articles for you (see section 7.6 Adaptive Scoring).

• Gnus maintains a dribble buffer that is auto-saved the normal Emacs manner, so it should be difficult to lose much data on what you have read if your machine should go down (see section 1.8 Auto Save).

• Gnus now has its own startup file (`~/.gnus.el') to avoid cluttering up the `.emacs' file.

• You can set the process mark on both groups and articles and perform operations on all the marked items (see section 8.1 Process/Prefix).

• You can grep through a subset of groups and create a group from the results (see section 6.7.2 Kibozed Groups).

• You can list subsets of groups according to, well, anything (see section 2.11 Listing Groups).

• You can browse foreign servers and subscribe to groups from those servers (see section 2.14 Browse Foreign Server).

• Gnus can fetch articles, asynchronously, on a second connection to the server (see section 3.11 Asynchronous Article Fetching).

• You can cache articles locally (see section 3.12 Article Caching).

• The uudecode functions have been expanded and generalized (see section 3.16 Decoding Articles).

• You can still post uuencoded articles, which was a little-known feature of GNUS' past (see section 3.16.5.3 Uuencoding and Posting).

• Fetching parents (and other articles) now actually works without glitches (see section 3.22 Finding the Parent).

• Gnus can fetch FAQs and group descriptions (see section 2.17.2 Group Information).

• Digests (and other files) can be used as the basis for groups (see section 6.6.3 Document Groups).

• Articles can be highlighted and customized (see section 4.3 Customizing Articles).

• URLs and other external references can be buttonized (see section 3.17.6 Article Buttons).

• You can do lots of strange stuff with the Gnus window & frame configuration (see section 8.5 Window Layout).

• You can click on buttons instead of using the keyboard (see section 8.10 Buttons).

10.2.9.2 September Gnus

New features in Gnus 5.2/5.3:

• A new message composition mode is used. All old customization variables for mail-mode, rnews-reply-mode and gnus-msg are now obsolete.

• Gnus is now able to generate sparse threads--threads where missing articles are represented by empty nodes (see section 3.9.1 Customizing Threading).

  (setq gnus-build-sparse-threads 'some)

• Outgoing articles are stored on a special archive server (see section 5.5 Archived Messages).

• Partial thread regeneration now happens when articles are referred.

• Gnus can make use of GroupLens predictions (see section 7.15 GroupLens).

• Picons (personal icons) can be displayed under XEmacs (see section 8.17.4 Picons).

• A trn-like tree buffer can be displayed (see section 3.24 Tree Display).

  (setq gnus-use-trees t)

• An nn-like pick-and-read minor mode is available for the summary buffers (see section 3.23.1 Pick and Read).

  (add-hook 'gnus-summary-mode-hook 'gnus-pick-mode)

• In binary groups you can use a special binary minor mode (see section 3.23.2 Binary Groups).

• Groups can be grouped in a folding topic hierarchy (see section 2.16 Group Topics).

  (add-hook 'gnus-group-mode-hook 'gnus-topic-mode)

• Gnus can re-send and bounce mail (see section 3.5.1 Summary Mail Commands).

• Groups can now have a score, and bubbling based on entry frequency is possible (see section 2.7 Group Score).

  (add-hook 'gnus-summary-exit-hook 'gnus-summary-bubble-group)

• Groups can be process-marked, and commands can be performed on groups of groups (see section 2.8 Marking Groups).

• Caching is possible in virtual groups.

• nndoc now understands all kinds of digests, mail boxes, rnews news batches, ClariNet briefs collections, and just about everything else (see section 6.6.3 Document Groups).

• Gnus has a new back end (nnsoup) to create/read SOUP packets (see section 6.6.4 SOUP).

• The Gnus cache is much faster.

• Groups can be sorted according to many criteria (see section 2.12 Sorting Groups).

• New group parameters have been introduced to set list-addresses and expiry times (see section 2.10 Group Parameters).

• All formatting specs allow specifying faces to be used (see section 8.4.5 Formatting Fonts).

• There are several more commands for setting/removing/acting on process marked articles on the M P submap (see section 3.7.6 Setting Process Marks).

• The summary buffer can be limited to show parts of the available articles based on a wide range of criteria. These commands have been bound to keys on the / submap (see section 3.8 Limiting).

• Articles can be made persistent with the * command (see section 3.13 Persistent Articles).

• All functions for hiding article elements are now toggles.

• Article headers can be buttonized (see section 3.17.4 Article Washing).

• All mail back ends support fetching articles by Message-ID.

• Duplicate mail can now be treated properly (see section 6.3.11 Duplicates).

• All summary mode commands are available directly from the article buffer (see section 4.4 Article Keymap).

• Frames can be part of gnus-buffer-configuration (see section 8.5 Window Layout).

• Mail can be re-scanned by a daemonic process (see section 8.11 Daemons).

• Gnus can make use of NoCeM files to weed out spam (see section 8.12 NoCeM).

  (setq gnus-use-nocem t)

• Groups can be made permanently visible (see section 2.11 Listing Groups).

  (setq gnus-permanently-visible-groups "^nnml:")

• Many new hooks have been introduced to make customizing easier.

• Gnus respects the Mail-Copies-To header.

• Threads can be gathered by looking at the References header (see section 3.9.1 Customizing Threading).

  (setq gnus-summary-thread-gathering-function 'gnus-gather-threads-by-references)

• Read articles can be stored in a special backlog buffer to avoid refetching (see section 3.14 Article Backlog).

  (setq gnus-keep-backlog 50)

• A clean copy of the current article is always stored in a separate buffer to allow easier treatment.

• Gnus can suggest where to save articles (see section 3.15 Saving Articles).

• Gnus doesn't have to do as much prompting when saving (see section 3.15 Saving Articles).

  (setq gnus-prompt-before-saving t)

• gnus-uu can view decoded files asynchronously while fetching articles (see section 3.16.5.2 Other Decode Variables).

  (setq gnus-uu-grabbed-file-functions 'gnus-uu-grab-view)

• Filling in the article buffer now works properly on cited text (see section 3.17.4 Article Washing).

• Hiding cited text adds buttons to toggle hiding, and how much cited text to hide is now customizable (see section 3.17.3 Article Hiding).

  (setq gnus-cited-lines-visible 2)

• Boring headers can be hidden (see section 3.17.3 Article Hiding).

• Default scoring values can now be set from the menu bar.

• Further syntax checking of outgoing articles have been added.

10.2.9.3 Red Gnus

New features in Gnus 5.4/5.5:

• `nntp.el' has been totally rewritten in an asynchronous fashion.

• Article prefetching functionality has been moved up into Gnus (see section 3.11 Asynchronous Article Fetching).

• Scoring can now be performed with logical operators like and, or, not, and parent redirection (see section 7.16 Advanced Scoring).

• Article washing status can be displayed in the article mode line (see section 4.5 Misc Article).

• `gnus.el' has been split into many smaller files.

• Suppression of duplicate articles based on Message-ID can be done (see section 3.29 Duplicate Suppression).

  (setq gnus-suppress-duplicates t)

• New variables for specifying what score and adapt files are to be considered home score and adapt files (see section 7.7 Home Score File) have been added.

• nndoc was rewritten to be easily extendable (see section 6.6.3.1 Document Server Internals).

• Groups can inherit group parameters from parent topics (see section 2.16.5 Topic Parameters).

• Article editing has been revamped and is now actually usable.

• Signatures can be recognized in more intelligent fashions (see section 3.17.10 Article Signature).

• Summary pick mode has been made to look more nn-like. Line numbers are displayed and the . command can be used to pick articles (Pick and Read).

• Commands for moving the `.newsrc.eld' from one server to another have been added (see section 1.6 Changing Servers).

• There's a way now to specify that "uninteresting" fields be suppressed when generating lines in buffers (see section 8.4.3 Advanced Formatting).

• Several commands in the group buffer can be undone with C-M-_ (see section 8.13 Undo).

• Scoring can be done on words using the new score type w (see section 7.4 Score File Format).

• Adaptive scoring can be done on a Subject word-by-word basis (see section 7.6 Adaptive Scoring).

  (setq gnus-use-adaptive-scoring '(word))

• Scores can be decayed (see section 7.17 Score Decays).

  (setq gnus-decay-scores t)

• Scoring can be performed using a regexp on the Date header. The Date is normalized to compact ISO 8601 format first (see section 7.4 Score File Format).

• A new command has been added to remove all data on articles from the native server (see section 1.6 Changing Servers).

• A new command for reading collections of documents (nndoc with nnvirtual on top) has been added---C-M-d (see section 3.26.4 Really Various Summary Commands).

• Process mark sets can be pushed and popped (see section 3.7.6 Setting Process Marks).

• A new mail-to-news back end makes it possible to post even when the NNTP server doesn't allow posting (see section 6.6.5 Mail-To-News Gateways).

• A new back end for reading searches from Web search engines (DejaNews, Alta Vista, InReference) has been added (see section 6.4.2 Web Searches).

• Groups inside topics can now be sorted using the standard sorting functions, and each topic can be sorted independently (see section 2.16.3 Topic Sorting).

• Subsets of the groups can be sorted independently (Sorting Groups).

• Cached articles can be pulled into the groups (see section 3.26.3 Summary Generation Commands).

• Score files are now applied in a more reliable order (see section 7.3 Score Variables).

• Reports on where mail messages end up can be generated (see section 6.3.3 Splitting Mail).

• More hooks and functions have been added to remove junk from incoming mail before saving the mail (see section 6.3.10 Washing Mail).

• Emphasized text can be properly fontisized:

10.2.9.4 Quassia Gnus

New features in Gnus 5.6:

• New functionality for using Gnus as an offline newsreader has been added. A plethora of new commands and modes have been added. See section 6.8 Gnus Unplugged, for the full story.

• The nndraft back end has returned, but works differently than before. All Message buffers are now also articles in the nndraft group, which is created automatically.

• gnus-alter-header-function can now be used to alter header values.

• gnus-summary-goto-article now accept Message-ID's.

• A new Message command for deleting text in the body of a message outside the region: C-c C-v.

• You can now post to component group in nnvirtual groups with C-u C-c C-c.

• nntp-rlogin-program---new variable to ease customization.

• C-u C-c C-c in gnus-article-edit-mode will now inhibit re-highlighting of the article buffer.

• New element in gnus-boring-article-headers---long-to.

• M-i symbolic prefix command. See section 8.3 Symbolic Prefixes, for details.

• L and I in the summary buffer now take the symbolic prefix a to add the score rule to the `all.SCORE' file.

• gnus-simplify-subject-functions variable to allow greater control over simplification.

• A T---new command for fetching the current thread.

• / T---new command for including the current thread in the limit.

• M-RET is a new Message command for breaking cited text.

• `\\1'-expressions are now valid in nnmail-split-methods.

• The custom-face-lookup function has been removed. If you used this function in your initialization files, you must rewrite them to use face-spec-set instead.

• Canceling now uses the current select method. Symbolic prefix a forces normal posting method.

• New command to translate M******** sm*rtq**t*s into proper text---W d.

• For easier debugging of nntp, you can set nntp-record-commands to a non-nil value.

• nntp now uses `~/.authinfo', a `.netrc'-like file, for controlling where and how to send AUTHINFO to NNTP servers.

• A command for editing group parameters from the summary buffer has been added.

• A history of where mails have been split is available.

• A new article date command has been added---article-date-iso8601.

• Subjects can be simplified when threading by setting gnus-score-thread-simplify.

• A new function for citing in Message has been added---message-cite-original-without-signature.

• article-strip-all-blank-lines---new article command.

• A new Message command to kill to the end of the article has been added.

• A minimum adaptive score can be specified by using the gnus-adaptive-word-minimum variable.

• The "lapsed date" article header can be kept continually updated by the gnus-start-date-timer command.

• Web listserv archives can be read with the nnlistserv back end.

• Old dejanews archives can now be read by nnweb.

10.2.9.5 Pterodactyl Gnus

New features in Gnus 5.8:

• The mail-fetching functions have changed. See the manual for the many details. In particular, all procmail fetching variables are gone.

  If you used procmail like in

  (setq nnmail-use-procmail t) (setq nnmail-spool-file 'procmail) (setq nnmail-procmail-directory "~/mail/incoming/") (setq nnmail-procmail-suffix "\\.in")

  this now has changed to

  (setq mail-sources '((directory :path "~/mail/incoming/" :suffix ".in")))

  See section 6.3.4.1 Mail Source Specifiers.

• Gnus is now a MIME-capable reader. This affects many parts of Gnus, and adds a slew of new commands. See the manual for details.

• Gnus has also been multilingualized. This also affects too many parts of Gnus to summarize here, and adds many new variables.

• gnus-auto-select-first can now be a function to be called to position point.

• The user can now decide which extra headers should be included in summary buffers and NOV files.

• gnus-article-display-hook has been removed. Instead, a number of variables starting with gnus-treat- have been added.

• The Gnus posting styles have been redone again and now works in a subtly different manner.

• New web-based back ends have been added: nnslashdot, nnwarchive and nnultimate. nnweb has been revamped, again, to keep up with ever-changing layouts.

• Gnus can now read IMAP mail via nnimap.

10.2.9.6 Oort Gnus

New features in Gnus 5.10:

• F (gnus-article-followup-with-original) and R (gnus-article-reply-with-original) only yank the text in the region if the region is active.

• gnus-group-read-ephemeral-group can be called interactively, using G M.

• In draft groups, e is now bound to gnus-draft-edit-message. Use B w for gnus-summary-edit-article instead.

• The revised Gnus FAQ is included in the manual, See section 10.9 Frequently Asked Questions.

• Upgrading from previous (stable) version if you have used Oort.

  If you have tried Oort (the unstable Gnus branch leading to this release) but went back to a stable version, be careful when upgrading to this version. In particular, you will probably want to remove all `.marks' (nnml) and `.mrk' (nnfolder) files, so that flags are read from your `.newsrc.eld' instead of from the `.marks'/`.mrk' file where this release store flags. See a later entry for more information about marks. Note that downgrading isn't save in general.

• Article Buttons

  More buttons for URLs, mail addresses, Message-IDs, Info links, man pages and Emacs or Gnus related references. See section 3.17.6 Article Buttons. The variables gnus-button-*-level can be used to control the appearance of all article buttons. See section 3.17.7 Article button levels.

• Dired integration

  gnus-dired-minor-mode (see 8.20 Interaction with other modes) installs key bindings in dired buffers to send a file as an attachment, open a file using the appropriate mailcap entry, and print a file using the mailcap entry.

• Gnus can display RSS newsfeeds as a newsgroup. See section 6.4.6 RSS.

• Single-part yenc encoded attachments can be decoded.

• Picons

  The picons code has been reimplemented to work in GNU Emacs--some of the previous options have been removed or renamed.

  Picons are small "personal icons" representing users, domain and newsgroups, which can be displayed in the Article buffer. See section 8.17.4 Picons.

• If the new option gnus-treat-body-boundary is non-nil, a boundary line is drawn at the end of the headers.

• Retrieval of charters and control messages

  There are new commands for fetching newsgroup charters (H c) and control messages (H C).

• Delayed articles

  You can delay the sending of a message with C-c C-j in the Message buffer. The messages are delivered at specified time. This is useful for sending yourself reminders. See section 3.6 Delayed Articles.

• If auto-compression-mode is enabled, attachments are automatically decompressed when activated.

• If the new option nnml-use-compressed-files is non-nil, the nnml back end allows compressed message files.

• Signed article headers (X-PGP-Sig) can be verified with W p.

• The Summary Buffer uses an arrow in the fringe to indicate the current article. Use (setq gnus-summary-display-arrow nil) to disable it.

• Warn about email replies to news

  Do you often find yourself replying to news by email by mistake? Then the new option gnus-confirm-mail-reply-to-news is just the thing for you.

• If the new option gnus-summary-display-while-building is non-nil, the summary buffer is shown and updated as it's being built.

• The new recent mark `.' indicates newly arrived messages (as opposed to old but unread messages).

• The new option gnus-gcc-mark-as-read automatically marks Gcc articles as read.

• The nndoc back end now supports mailman digests and exim bounces.

• Gnus supports RFC 2369 mailing list headers, and adds a number of related commands in mailing list groups. See section 3.31 Mailing List.

• The Date header can be displayed in a format that can be read aloud in English. See section 3.17.8 Article Date.

• The envelope sender address can be customized when using Sendmail. See section `Mail Variables' in Message Manual.

• diffs are automatically highlighted in groups matching mm-uu-diff-groups-regexp

• TLS wrapper shipped with Gnus

  TLS/SSL is now supported in IMAP and NNTP via `tls.el' and GNUTLS. The old TLS/SSL support via (external third party) `ssl.el' and OpenSSL still works.

• New `make.bat' for compiling and installing Gnus under MS Windows

  Use `make.bat' if you want to install Gnus under MS Windows, the first argument to the batch-program should be the directory where `xemacs.exe' respectively `emacs.exe' is located, iff you want to install Gnus after compiling it, give `make.bat' /copy as the second parameter.

  `make.bat' has been rewritten from scratch, it now features automatic recognition of XEmacs and GNU Emacs, generates `gnus-load.el', checks if errors occur while compilation and generation of info files and reports them at the end of the build process. It now uses makeinfo if it is available and falls back to `infohack.el' otherwise. `make.bat' should now install all files which are necessary to run Gnus and be generally a complete replacement for the configure; make; make install cycle used under Unix systems.

  The new `make.bat' makes `make-x.bat' superfluous, so it has been removed.

• Support for non-ASCII domain names

  Message supports non-ASCII domain names in From:, To: and Cc: and will query you whether to perform encoding when you try to send a message. The variable message-use-idna controls this. Gnus will also decode non-ASCII domain names in From:, To: and Cc: when you view a message. The variable gnus-use-idna controls this.

• Better handling of Microsoft citation styles

  Gnus now tries to recognize the mangled header block that some Microsoft mailers use to indicate that the rest of the message is a citation, even though it is not quoted in any way. The variable gnus-cite-unsightly-citation-regexp matches the start of these citations.

• gnus-article-skip-boring

  If you set gnus-article-skip-boring to t, then Gnus will not scroll down to show you a page that contains only boring text, which by default means cited text and signature. You can customize what is skippable using gnus-article-boring-faces.

  This feature is especially useful if you read many articles that consist of a little new content at the top with a long, untrimmed message cited below.

• The format spec %C for positioning point has changed to %*.

• The new variable gnus-parameters can be used to set group parameters.

  Earlier this was done only via G p (or G c), which stored the parameters in `~/.newsrc.eld', but via this variable you can enjoy the powers of customize, and simplified backups since you set the variable in `~/.gnus.el' instead of `~/.newsrc.eld'. The variable maps regular expressions matching group names to group parameters, a'la:

  (setq gnus-parameters '(("mail\\..*" (gnus-show-threads nil) (gnus-use-scoring nil)) ("^nnimap:\\(foo.bar\\)$" (to-group . "\\1"))))

• Smileys (`:-)', `;-)' etc) are now iconized for Emacs too.

  Put (setq gnus-treat-display-smileys nil) in `~/.gnus.el' to disable it.

• Gnus no longer generate the Sender: header automatically.

  Earlier it was generated iff the user configurable email address was different from the Gnus guessed default user address. As the guessing algorithm is rarely correct these days, and (more controversially) the only use of the Sender: header was to check if you are entitled to cancel/supersede news (which is now solved by Cancel Locks instead, see another entry), generation of the header has been disabled by default. See the variables message-required-headers, message-required-news-headers, and message-required-mail-headers.

• Features from third party `message-utils.el' added to `message.el'.

  Message now asks if you wish to remove `(was: <old subject>)' from subject lines (see message-subject-trailing-was-query). C-c M-m and C-c M-f inserts markers indicating included text. C-c C-f a adds a X-No-Archive: header. C-c C-f x inserts appropriate headers and a note in the body for cross-postings and followups (see the variables message-cross-post-*).

• References and X-Draft-From headers are no longer generated when you start composing messages and message-generate-headers-first is nil.

• Improved anti-spam features.

  Gnus is now able to take out spam from your mail and news streams using a wide variety of programs and filter rules. Among the supported methods are RBL blocklists, bogofilter and white/blacklists. Hooks for easy use of external packages such as SpamAssassin and Hashcash are also new. See section 8.19 Thwarting Email Spam.

• Easy inclusion of X-Faces headers.

• Face headers handling.

• In the summary buffer, the new command / N inserts new messages and / o inserts old messages.

• Gnus decodes morse encoded messages if you press W m.

• Unread count correct in nnimap groups.

  The estimated number of unread articles in the group buffer should now be correct for nnimap groups. This is achieved by calling nnimap-fixup-unread-after-getting-new-news from the gnus-setup-news-hook (called on startup) and gnus-after-getting-new-news-hook. (called after getting new mail). If you have modified those variables from the default, you may want to add nnimap-fixup-unread-after-getting-new-news again. If you were happy with the estimate and want to save some (minimal) time when getting new mail, remove the function.

• Group Carbon Copy (GCC) quoting

  To support groups that contains SPC and other weird characters, groups are quoted before they are placed in the Gcc: header. This means variables such as gnus-message-archive-group should no longer contain quote characters to make groups containing SPC work. Also, if you are using the string `nnml:foo, nnml:bar' (indicating Gcc into two groups) you must change it to return the list ("nnml:foo" "nnml:bar"), otherwise the Gcc: line will be quoted incorrectly. Note that returning the string `nnml:foo, nnml:bar' was incorrect earlier, it just didn't generate any problems since it was inserted directly.

• `~/News/overview/' not used.

  As a result of the following change, the `~/News/overview/' directory is not used any more. You can safely delete the entire hierarchy.

• gnus-agent

  The Gnus Agent has seen a major updated and is now enabled by default, and all nntp and nnimap servers from gnus-select-method and gnus-secondary-select-method are agentized by default. Earlier only the server in gnus-select-method was agentized by the default, and the agent was disabled by default. When the agent is enabled, headers are now also retrieved from the Agent cache instead of the back ends when possible. Earlier this only happened in the unplugged state. You can enroll or remove servers with J a and J r in the server buffer. Gnus will not download articles into the Agent cache, unless you instruct it to do so, though, by using J u or J s from the Group buffer. You revert to the old behavior of having the Agent disabled with (setq gnus-agent nil). Note that putting (gnus-agentize) in `~/.gnus.el' is not needed any more.

• gnus-summary-line-format

  The default value changed to `%U%R%z%I%(%[%4L: %-23,23f%]%) %s\n'. Moreover gnus-extra-headers, nnmail-extra-headers and gnus-ignored-from-addresses changed their default so that the users name will be replaced by the recipient's name or the group name posting to for NNTP groups.

• `deuglify.el' (gnus-article-outlook-deuglify-article)

  A new file from Raymond Scholz rscholz@zonix.de for deuglifying broken Outlook (Express) articles.

• (require 'gnus-load)

  If you use a stand-alone Gnus distribution, you'd better add (require 'gnus-load) into your `~/.emacs' after adding the Gnus lisp directory into load-path.

  File `gnus-load.el' contains autoload commands, functions and variables, some of which may not be included in distributions of Emacsen.

• gnus-slave-unplugged

  A new command which starts Gnus offline in slave mode.

• message-insinuate-rmail

  Adding (message-insinuate-rmail) and (setq mail-user-agent 'gnus-user-agent) in `.emacs' convinces Rmail to compose, reply and forward messages in message-mode, where you can enjoy the power of MML.

• message-minibuffer-local-map

  The line below enables BBDB in resending a message:

  (define-key message-minibuffer-local-map [(tab)] 'bbdb-complete-name)

• Externalizing and deleting of attachments.

  If gnus-gcc-externalize-attachments or message-fcc-externalize-attachments is non-nil, attach local files as external parts.

  The command gnus-mime-save-part-and-strip (bound to C-o on MIME buttons) saves a part and replaces the part with an external one. gnus-mime-delete-part (bound to d on MIME buttons) removes a part. It works only on back ends that support editing.

• gnus-default-charset

  The default value is determined from the current-language-environment variable, instead of iso-8859-1. Also the `.*' item in gnus-group-charset-alist is removed.

• gnus-posting-styles

  Add a new format of match like

  ((header "to" "larsi.*org") (Organization "Somewhere, Inc."))

  The old format like the lines below is obsolete, but still accepted.

  (header "to" "larsi.*org" (Organization "Somewhere, Inc."))

• message-ignored-news-headers and message-ignored-mail-headers

  `X-Draft-From' and `X-Gnus-Agent-Meta-Information' have been added into these two variables. If you customized those, perhaps you need add those two headers too.

• Gnus reads the NOV and articles in the Agent if plugged.

  If one reads an article while plugged, and the article already exists in the Agent, it won't get downloaded once more. (setq gnus-agent-cache nil) reverts to the old behavior.

• Gnus supports the "format=flowed" (RFC 2646) parameter. On composing messages, it is enabled by use-hard-newlines. Decoding format=flowed was present but not documented in earlier versions.

• The option mm-fill-flowed can be used to disable treatment of "format=flowed" messages. Also, flowed text is disabled when sending inline PGP signed messages. (New in Gnus 5.10.7)

• Gnus supports the generation of RFC 2298 Disposition Notification requests.

  This is invoked with the C-c M-n key binding from message mode.

• Gnus supports Maildir groups.

  Gnus includes a new back end `nnmaildir.el'. See section 6.3.13.5 Maildir.

• Printing capabilities are enhanced.

  Gnus supports Muttprint natively with O P from the Summary and Article buffers. Also, each individual MIME part can be printed using p on the MIME button.

• Message supports the Importance: (RFC 2156) header.

  In the message buffer, C-c C-f C-i or C-c C-u cycles through the valid values.

• Gnus supports Cancel Locks in News.

  This means a header `Cancel-Lock' is inserted in news posting. It is used to determine if you wrote an article or not (for canceling and superseding). Gnus generates a random password string the first time you post a message, and saves it in your `~/.emacs' using the Custom system. While the variable is called canlock-password, it is not security sensitive data. Publishing your canlock string on the web will not allow anyone to be able to anything she could not already do. The behavior can be changed by customizing message-insert-canlock.

• Gnus supports server-side mail filtering using Sieve.

  Sieve rules can be added as Group Parameters for groups, and the complete Sieve script is generated using D g from the Group buffer, and then uploaded to the server using C-c C-l in the generated Sieve buffer. See section 2.17.5 Sieve Commands, and the new Sieve manual section `Top' in Emacs Sieve.

• Extended format specs.

  Format spec `%&user-date;' is added into gnus-summary-line-format-alist. Also, user defined extended format specs are supported. The extended format specs look like `%u&foo;', which invokes function gnus-user-format-function-foo. Because `&' is used as the escape character, old user defined format `%u&' is no longer supported.

• / * (gnus-summary-limit-include-cached) is rewritten.

  It was aliased to Y c (gnus-summary-insert-cached-articles). The new function filters out other articles.

• Some limiting commands accept a C-u prefix to negate the match.

  If C-u is used on subject, author or extra headers, i.e., / s, / a, and / x (gnus-summary-limit-to-{subject,author,extra}) respectively, the result will be to display all articles that do not match the expression.

• Group names are treated as UTF-8 by default.

  This is supposedly what USEFOR wanted to migrate to. See gnus-group-name-charset-group-alist and gnus-group-name-charset-method-alist for customization.

• The nnml and nnfolder back ends store marks for each groups.

  This makes it possible to take backup of nnml/nnfolder servers/groups separately of `~/.newsrc.eld', while preserving marks. It also makes it possible to share articles and marks between users (without sharing the `~/.newsrc.eld' file) within e.g. a department. It works by storing the marks stored in `~/.newsrc.eld' in a per-group file `.marks' (for nnml) and `groupname.mrk' (for nnfolder, named groupname). If the nnml/nnfolder is moved to another machine, Gnus will automatically use the `.marks' or `.mrk' file instead of the information in `~/.newsrc.eld'. The new server variables nnml-marks-is-evil and nnfolder-marks-is-evil can be used to disable this feature.

• The menu bar item (in Group and Summary buffer) named "Misc" has been renamed to "Gnus".

• The menu bar item (in Message mode) named "MML" has been renamed to "Attachments". Note that this menu also contains security related stuff, like signing and encryption (see section `Security' in Message Manual).

• gnus-group-charset-alist and gnus-group-ignored-charsets-alist.

  The regexps in these variables are compared with full group names instead of real group names in 5.8. Users who customize these variables should change those regexps accordingly. For example:

  ("^han\\>" euc-kr) -> ("\\(^\\|:\\)han\\>" euc-kr)

• Gnus supports PGP (RFC 1991/2440), PGP/MIME (RFC 2015/3156) and S/MIME (RFC 2630-2633).

  It needs an external S/MIME and OpenPGP implementation, but no additional Lisp libraries. This add several menu items to the Attachments menu, and C-c RET key bindings, when composing messages. This also obsoletes gnus-article-hide-pgp-hook.

• Gnus inlines external parts (message/external).

• MML (Mime compose) prefix changed from M-m to C-c C-m.

  This change was made to avoid conflict with the standard binding of back-to-indentation, which is also useful in message mode.

• The default for message-forward-show-mml changed to symbol best.

  The behavior for the best value is to show MML (i.e., convert to MIME) when appropriate. MML will not be used when forwarding signed or encrypted messages, as the conversion invalidate the digital signature.

10.3 On Writing Manuals

I guess most manuals are written after-the-fact; documenting a program that's already there. This is not how this manual is written. When implementing something, I write the manual entry for that something straight away. I then see that it's difficult to explain the functionality, so I write how it's supposed to be, and then I change the implementation. Writing the documentation and writing the code goes hand in hand.

This, of course, means that this manual has no, or little, flow. It documents absolutely everything in Gnus, but often not where you're looking for it. It is a reference manual, and not a guide to how to get started with Gnus.

That would be a totally different book, that should be written using the reference manual as source material. It would look quite differently.

10.4 Terminology

news

This is what you are supposed to use this thing for--reading news. News is generally fetched from a nearby NNTP server, and is generally publicly available to everybody. If you post news, the entire world is likely to read just what you have written, and they'll all snigger mischievously. Behind your back.

mail

Everything that's delivered to you personally is mail. Some news/mail readers (like Gnus) blur the distinction between mail and news, but there is a difference. Mail is private. News is public. Mailing is not posting, and replying is not following up.

reply

Send a mail to the person who has written what you are reading.

follow up

Post an article to the current newsgroup responding to the article you are reading.

back end

Gnus considers mail and news to be mostly the same, really. The only difference is how to access the actual articles. News articles are commonly fetched via the protocol NNTP, whereas mail messages could be read from a file on the local disk. The internal architecture of Gnus thus comprises a "front end" and a number of "back ends". Internally, when you enter a group (by hitting RET, say), you thereby invoke a function in the front end in Gnus. The front end then "talks" to a back end and says things like "Give me the list of articles in the foo group" or "Show me article number 4711".

So a back end mainly defines either a protocol (the nntp back end accesses news via NNTP, the nnimap back end accesses mail via IMAP) or a file format and directory layout (the nnspool back end accesses news via the common "spool directory" format, the nnml back end access mail via a file format and directory layout that's quite similar).

Gnus does not handle the underlying media, so to speak--this is all done by the back ends. A back end is a collection of functions to access the articles.

However, sometimes the term "back end" is also used where "server" would have been more appropriate. And then there is the term "select method" which can mean either. The Gnus terminology can be quite confusing.

native

Gnus will always use one method (and back end) as the native, or default, way of getting news.

foreign

You can also have any number of foreign groups active at the same time. These are groups that use non-native non-secondary back ends for getting news.

secondary

Secondary back ends are somewhere half-way between being native and being foreign, but they mostly act like they are native.

article

A message that has been posted as news.

mail message

A message that has been mailed.

message

A mail message or news article

head

The top part of a message, where administrative information (etc.) is put.

body

The rest of an article. Everything not in the head is in the body.

header

A line from the head of an article.

headers

A collection of such lines, or a collection of heads. Or even a collection of NOV lines.

NOV

When Gnus enters a group, it asks the back end for the headers of all unread articles in the group. Most servers support the News OverView format, which is more compact and much faster to read and parse than the normal HEAD format.

level

Each group is subscribed at some level or other (1-9). The ones that have a lower level are "more" subscribed than the groups with a higher level. In fact, groups on levels 1-5 are considered subscribed; 6-7 are unsubscribed; 8 are zombies; and 9 are killed. Commands for listing groups and scanning for new articles will all use the numeric prefix as working level.

killed groups

No information on killed groups is stored or updated, which makes killed groups much easier to handle than subscribed groups.

zombie groups

Just like killed groups, only slightly less dead.

active file

The news server has to keep track of what articles it carries, and what groups exist. All this information in stored in the active file, which is rather large, as you might surmise.

bogus groups

A group that exists in the `.newsrc' file, but isn't known to the server (i.e., it isn't in the active file), is a bogus group. This means that the group probably doesn't exist (any more).

activating

The act of asking the server for info on a group and computing the number of unread articles is called activating the group. Un-activated groups are listed with `*' in the group buffer.

spool

News servers store their articles locally in one fashion or other. One old-fashioned storage method is to have just one file per article. That's called a "traditional spool".

server

A machine one can connect to and get news (or mail) from.

select method

A structure that specifies the back end, the server and the virtual server settings.

virtual server

A named select method. Since a select method defines all there is to know about connecting to a (physical) server, taking the thing as a whole is a virtual server.

washing

Taking a buffer and running it through a filter of some sort. The result will (more often than not) be cleaner and more pleasing than the original.

ephemeral groups

Most groups store data on what articles you have read. Ephemeral groups are groups that will have no data stored--when you exit the group, it'll disappear into the aether.

solid groups

This is the opposite of ephemeral groups. All groups listed in the group buffer are solid groups.

sparse articles

These are article placeholders shown in the summary buffer when gnus-build-sparse-threads has been switched on.

threading

To put responses to articles directly after the articles they respond to--in a hierarchical fashion.

root

The first article in a thread is the root. It is the ancestor of all articles in the thread.

parent

An article that has responses.

child

An article that responds to a different article--its parent.

digest

A collection of messages in one file. The most common digest format is specified by RFC 1153.

splitting

The action of sorting your emails according to certain rules. Sometimes incorrectly called mail filtering.

10.5 Customization

All variables are properly documented elsewhere in this manual. This section is designed to give general pointers on how to customize Gnus for some quite common situations.

10.5.1 Slow/Expensive NNTP Connection

If you run Emacs on a machine locally, and get your news from a machine over some very thin strings, you want to cut down on the amount of data Gnus has to get from the NNTP server.

gnus-read-active-file

Set this to nil, which will inhibit Gnus from requesting the entire active file from the server. This file is often very large. You also have to set gnus-check-new-newsgroups and gnus-check-bogus-newsgroups to nil to make sure that Gnus doesn't suddenly decide to fetch the active file anyway.

gnus-nov-is-evil

This one has to be nil. If not, grabbing article headers from the NNTP server will not be very fast. Not all NNTP servers support XOVER; Gnus will detect this by itself.

10.5.2 Slow Terminal Connection

Let's say you use your home computer for dialing up the system that runs Emacs and Gnus. If your modem is slow, you want to reduce (as much as possible) the amount of data sent over the wires.

gnus-auto-center-summary

Set this to nil to inhibit Gnus from re-centering the summary buffer all the time. If it is vertical, do only vertical re-centering. If it is neither nil nor vertical, do both horizontal and vertical recentering.

gnus-visible-headers

Cut down on the headers included in the articles to the minimum. You can, in fact, make do without them altogether--most of the useful data is in the summary buffer, anyway. Set this variable to `^NEVVVVER' or `From:', or whatever you feel you need.

Use the following to enable all the available hiding features:

(setq gnus-treat-hide-headers 'head gnus-treat-hide-signature t gnus-treat-hide-citation t)

gnus-use-full-window

By setting this to nil, you can make all the windows smaller. While this doesn't really cut down much generally, it means that you have to see smaller portions of articles before deciding that you didn't want to read them anyway.

gnus-thread-hide-subtree

If this is non-nil, all threads in the summary buffer will be hidden initially.

gnus-updated-mode-lines

If this is nil, Gnus will not put information in the buffer mode lines, which might save some time.

10.5.3 Little Disk Space

The startup files can get rather large, so you may want to cut their sizes a bit if you are running out of space.

gnus-save-newsrc-file

If this is nil, Gnus will never save `.newsrc'---it will only save `.newsrc.eld'. This means that you will not be able to use any other newsreaders than Gnus. This variable is t by default.

gnus-read-newsrc-file

If this is nil, Gnus will never read `.newsrc'---it will only read `.newsrc.eld'. This means that you will not be able to use any other newsreaders than Gnus. This variable is t by default.

gnus-save-killed-list

If this is nil, Gnus will not save the list of dead groups. You should also set gnus-check-new-newsgroups to ask-server and gnus-check-bogus-newsgroups to nil if you set this variable to nil. This variable is t by default.

10.5.4 Slow Machine

If you have a slow machine, or are just really impatient, there are a few things you can do to make Gnus run faster.

Set gnus-check-new-newsgroups and gnus-check-bogus-newsgroups to nil to make startup faster.

Set gnus-show-threads, gnus-use-cross-reference and gnus-nov-is-evil to nil to make entering and exiting the summary buffer faster.

10.6 Troubleshooting

Gnus works so well straight out of the box--I can't imagine any problems, really.

Ahem.

1. Make sure your computer is switched on.

2. Make sure that you really load the current Gnus version. If you have been running GNUS, you need to exit Emacs and start it up again before Gnus will work.

3. Try doing an M-x gnus-version. If you get something that looks like `Gnus v5.10.6' you have the right files loaded. Otherwise you have some old `.el' files lying around. Delete these.

4. Read the help group (G h in the group buffer) for a FAQ and a how-to.

5. Gnus works on many recursive structures, and in some extreme (and very rare) cases Gnus may recurse down "too deeply" and Emacs will beep at you. If this happens to you, set max-lisp-eval-depth to 500 or something like that.

If all else fails, report the problem as a bug.

If you find a bug in Gnus, you can report it with the M-x gnus-bug command. M-x set-variable RET debug-on-error RET t RET, and send me the backtrace. I will fix bugs, but I can only fix them if you send me a precise description as to how to reproduce the bug.

You really can never be too detailed in a bug report. Always use the M-x gnus-bug command when you make bug reports, even if it creates a 10Kb mail each time you use it, and even if you have sent me your environment 500 times before. I don't care. I want the full info each time.

It is also important to remember that I have no memory whatsoever. If you send a bug report, and I send you a reply, and then you just send back "No, it's not! Moron!", I will have no idea what you are insulting me about. Always over-explain everything. It's much easier for all of us--if I don't have all the information I need, I will just mail you and ask for more info, and everything takes more time.

If the problem you're seeing is very visual, and you can't quite explain it, copy the Emacs window to a file (with xwd, for instance), put it somewhere it can be reached, and include the URL of the picture in the bug report.

If you would like to contribute a patch to fix bugs or make improvements, please produce the patch using `diff -u'.

If you want to debug your problem further before reporting, possibly in order to solve the problem yourself and send a patch, you can use edebug. Debugging Lisp code is documented in the Elisp manual (see section `Debugging Lisp Programs' in The GNU Emacs Lisp Reference Manual). To get you started with edebug, consider if you discover some weird behavior when pressing c, the first step is to do C-h k c and click on the hyperlink (Emacs only) in the documentation buffer that leads you to the function definition, then press M-x edebug-defun RET with point inside that function, return to Gnus and press c to invoke the code. You will be placed in the lisp buffer and can single step using SPC and evaluate expressions using M-: or inspect variables using C-h v, abort execution with q, and resume execution with c or g.

Sometimes, a problem do not directly generate an elisp error but manifests itself by causing Gnus to be very slow. In these cases, you can use M-x toggle-debug-on-quit and press C-g when things are slow, and then try to analyze the backtrace (repeating the procedure helps isolating the real problem areas).

A fancier approach is to use the elisp profiler, ELP. The profiler is (or should be) fully documented elsewhere, but to get you started there are a few steps that need to be followed. First, instrument the part of Gnus you are interested in for profiling, e.g. M-x elp-instrument-package RET gnus or M-x elp-instrument-package RET message. Then perform the operation that is slow and press M-x elp-results. You will then see which operations that takes time, and can debug them further. If the entire operation takes much longer than the time spent in the slowest function in the profiler output, you probably profiled the wrong part of Gnus. To reset profiling statistics, use M-x elp-reset-all. M-x elp-restore-all is supposed to remove profiling, but given the complexities and dynamic code generation in Gnus, it might not always work perfectly.

If you just need help, you are better off asking on `gnu.emacs.gnus'. I'm not very helpful. You can also ask on the ding mailing list. Write to ding-request@gnus.org to subscribe.

10.7 Gnus Reference Guide

It is my hope that other people will figure out smart stuff that Gnus can do, and that other people will write those smart things as well. To facilitate that I thought it would be a good idea to describe the inner workings of Gnus. And some of the not-so-inner workings, while I'm at it.

You can never expect the internals of a program not to change, but I will be defining (in some details) the interface between Gnus and its back ends (this is written in stone), the format of the score files (ditto), data structures (some are less likely to change than others) and general methods of operation.

10.7.1 Gnus Utility Functions

When writing small functions to be run from hooks (and stuff), it's vital to have access to the Gnus internal functions and variables. Below is a list of the most common ones.

gnus-newsgroup-name

This variable holds the name of the current newsgroup.

gnus-find-method-for-group

A function that returns the select method for group.

gnus-group-real-name

Takes a full (prefixed) Gnus group name, and returns the unprefixed name.

gnus-group-prefixed-name

Takes an unprefixed group name and a select method, and returns the full (prefixed) Gnus group name.

gnus-get-info

Returns the group info list for group.

gnus-group-unread

The number of unread articles in group, or t if that is unknown.

gnus-active

The active entry for group.

gnus-set-active

Set the active entry for group.

gnus-add-current-to-buffer-list

Adds the current buffer to the list of buffers to be killed on Gnus exit.

gnus-continuum-version

Takes a Gnus version string as a parameter and returns a floating point number. Earlier versions will always get a lower number than later versions.

gnus-group-read-only-p

Says whether group is read-only or not.

gnus-news-group-p

Says whether group came from a news back end.

gnus-ephemeral-group-p

Says whether group is ephemeral or not.

gnus-server-to-method

Returns the select method corresponding to server.

gnus-server-equal

Says whether two virtual servers are equal.

gnus-group-native-p

Says whether group is native or not.

gnus-group-secondary-p

Says whether group is secondary or not.

gnus-group-foreign-p

Says whether group is foreign or not.

gnus-group-find-parameter

Returns the parameter list of group. If given a second parameter, returns the value of that parameter for group.

gnus-group-set-parameter

Takes three parameters; group, parameter and value.

gnus-narrow-to-body

Narrows the current buffer to the body of the article.

gnus-check-backend-function

Takes two parameters, function and group. If the back end group comes from supports function, return non-nil.

(gnus-check-backend-function "request-scan" "nnml:misc") => t

gnus-read-method

Prompts the user for a select method.

10.7.2 Back End Interface

Gnus doesn't know anything about NNTP, spools, mail or virtual groups. It only knows how to talk to virtual servers. A virtual server is a back end and some back end variables. As examples of the first, we have nntp, nnspool and nnmbox. As examples of the latter we have nntp-port-number and nnmbox-directory.

When Gnus asks for information from a back end--say nntp---on something, it will normally include a virtual server name in the function parameters. (If not, the back end should use the "current" virtual server.) For instance, nntp-request-list takes a virtual server as its only (optional) parameter. If this virtual server hasn't been opened, the function should fail.

Note that a virtual server name has no relation to some physical server name. Take this example:

(nntp "odd-one"
      (nntp-address "ifi.uio.no")
      (nntp-port-number 4324))

Here the virtual server name is `odd-one' while the name of the physical server is `ifi.uio.no'.

The back ends should be able to switch between several virtual servers. The standard back ends implement this by keeping an alist of virtual server environments that they pull down/push up when needed.

There are two groups of interface functions: required functions, which must be present, and optional functions, which Gnus will always check for presence before attempting to call 'em.

All these functions are expected to return data in the buffer nntp-server-buffer (` *nntpd*'), which is somewhat unfortunately named, but we'll have to live with it. When I talk about resulting data, I always refer to the data in that buffer. When I talk about return value, I talk about the function value returned by the function call. Functions that fail should return nil as the return value.

Some back ends could be said to be server-forming back ends, and some might be said not to be. The latter are back ends that generally only operate on one group at a time, and have no concept of "server" ---they have a group, and they deliver info on that group and nothing more.

Gnus identifies each message by way of group name and article number. A few remarks about these article numbers might be useful. First of all, the numbers are positive integers. Secondly, it is normally not possible for later articles to "re-use" older article numbers without confusing Gnus. That is, if a group has ever contained a message numbered 42, then no other message may get that number, or Gnus will get mightily confused.(3) Third, article numbers must be assigned in order of arrival in the group; this is not necessarily the same as the date of the message.

The previous paragraph already mentions all the "hard" restrictions that article numbers must fulfill. But it seems that it might be useful to assign consecutive article numbers, for Gnus gets quite confused if there are holes in the article numbering sequence. However, due to the "no-reuse" restriction, holes cannot be avoided altogether. It's also useful for the article numbers to start at 1 to avoid running out of numbers as long as possible.

Note that by convention, back ends are named nnsomething, but Gnus also comes with some nnnotbackends, such as `nnheader.el', `nnmail.el' and `nnoo.el'.

In the examples and definitions I will refer to the imaginary back end nnchoke.

10.7.2.1 Required Back End Functions

(nnchoke-retrieve-headers ARTICLES &optional GROUP SERVER FETCH-OLD)

articles is either a range of article numbers or a list of Message-IDs. Current back ends do not fully support either--only sequences (lists) of article numbers, and most back ends do not support retrieval of Message-IDs. But they should try for both.

The result data should either be HEADs or NOV lines, and the result value should either be headers or nov to reflect this. This might later be expanded to various, which will be a mixture of HEADs and NOV lines, but this is currently not supported by Gnus.

If fetch-old is non-nil it says to try fetching "extra headers", in some meaning of the word. This is generally done by fetching (at most) fetch-old extra headers less than the smallest article number in articles, and filling the gaps as well. The presence of this parameter can be ignored if the back end finds it cumbersome to follow the request. If this is non-nil and not a number, do maximum fetches.

Here's an example HEAD:

221 1056 Article retrieved. Path: ifi.uio.no!sturles From: sturles@ifi.uio.no (Sturle Sunde) Newsgroups: ifi.discussion Subject: Re: Something very droll Date: 27 Oct 1994 14:02:57 +0100 Organization: Dept. of Informatics, University of Oslo, Norway Lines: 26 Message-ID: <38o8e1$a0o@holmenkollen.ifi.uio.no> References: <38jdmq$4qu@visbur.ifi.uio.no> NNTP-Posting-Host: holmenkollen.ifi.uio.no .

So a headers return value would imply that there's a number of these in the data buffer.

Here's a BNF definition of such a buffer:

headers = *head head = error / valid-head error-message = [ "4" / "5" ] 2number " " <error message> eol valid-head = valid-message *header "." eol valid-message = "221 " <number> " Article retrieved." eol header = <text> eol

(The version of BNF used here is the one used in RFC822.)

If the return value is nov, the data buffer should contain network overview database lines. These are basically fields separated by tabs.

nov-buffer = *nov-line nov-line = field 7*8[ <TAB> field ] eol field = <text except TAB>

For a closer look at what should be in those fields, see section 10.7.4 Headers.

(nnchoke-open-server SERVER &optional DEFINITIONS)

server is here the virtual server name. definitions is a list of (VARIABLE VALUE) pairs that define this virtual server.

If the server can't be opened, no error should be signaled. The back end may then choose to refuse further attempts at connecting to this server. In fact, it should do so.

If the server is opened already, this function should return a non-nil value. There should be no data returned.

(nnchoke-close-server &optional SERVER)

Close connection to server and free all resources connected to it. Return nil if the server couldn't be closed for some reason.

There should be no data returned.

(nnchoke-request-close)

Close connection to all servers and free all resources that the back end have reserved. All buffers that have been created by that back end should be killed. (Not the nntp-server-buffer, though.) This function is generally only called when Gnus is shutting down.

There should be no data returned.

(nnchoke-server-opened &optional SERVER)

If server is the current virtual server, and the connection to the physical server is alive, then this function should return a non-nil value. This function should under no circumstances attempt to reconnect to a server we have lost connection to.

There should be no data returned.

(nnchoke-status-message &optional SERVER)

This function should return the last error message from server.

There should be no data returned.

(nnchoke-request-article ARTICLE &optional GROUP SERVER TO-BUFFER)

The result data from this function should be the article specified by article. This might either be a Message-ID or a number. It is optional whether to implement retrieval by Message-ID, but it would be nice if that were possible.

If to-buffer is non-nil, the result data should be returned in this buffer instead of the normal data buffer. This is to make it possible to avoid copying large amounts of data from one buffer to another, while Gnus mainly requests articles to be inserted directly into its article buffer.

If it is at all possible, this function should return a cons cell where the car is the group name the article was fetched from, and the cdr is the article number. This will enable Gnus to find out what the real group and article numbers are when fetching articles by Message-ID. If this isn't possible, t should be returned on successful article retrieval.

(nnchoke-request-group GROUP &optional SERVER FAST)

Get data on group. This function also has the side effect of making group the current group.

If fast, don't bother to return useful data, just make group the current group.

Here's an example of some result data and a definition of the same:

211 56 1000 1059 ifi.discussion

The first number is the status, which should be 211. Next is the total number of articles in the group, the lowest article number, the highest article number, and finally the group name. Note that the total number of articles may be less than one might think while just considering the highest and lowest article numbers, but some articles may have been canceled. Gnus just discards the total-number, so whether one should take the bother to generate it properly (if that is a problem) is left as an exercise to the reader. If the group contains no articles, the lowest article number should be reported as 1 and the highest as 0.

group-status = [ error / info ] eol error = [ "4" / "5" ] 2<number> " " <Error message> info = "211 " 3* [ <number> " " ] <string>

(nnchoke-close-group GROUP &optional SERVER)

Close group and free any resources connected to it. This will be a no-op on most back ends.

There should be no data returned.

(nnchoke-request-list &optional SERVER)

Return a list of all groups available on server. And that means all.

Here's an example from a server that only carries two groups:

ifi.test 0000002200 0000002000 y ifi.discussion 3324 3300 n

On each line we have a group name, then the highest article number in that group, the lowest article number, and finally a flag. If the group contains no articles, the lowest article number should be reported as 1 and the highest as 0.

active-file = *active-line active-line = name " " <number> " " <number> " " flags eol name = <string> flags = "n" / "y" / "m" / "x" / "j" / "=" name

The flag says whether the group is read-only (`n'), is moderated (`m'), is dead (`x'), is aliased to some other group (`=other-group') or none of the above (`y').

(nnchoke-request-post &optional SERVER)

This function should post the current buffer. It might return whether the posting was successful or not, but that's not required. If, for instance, the posting is done asynchronously, it has generally not been completed by the time this function concludes. In that case, this function should set up some kind of sentinel to beep the user loud and clear if the posting could not be completed.

There should be no result data from this function.

10.7.2.2 Optional Back End Functions

(nnchoke-retrieve-groups GROUPS &optional SERVER)

groups is a list of groups, and this function should request data on all those groups. How it does it is of no concern to Gnus, but it should attempt to do this in a speedy fashion.

The return value of this function can be either active or group, which says what the format of the result data is. The former is in the same format as the data from nnchoke-request-list, while the latter is a buffer full of lines in the same format as nnchoke-request-group gives.

group-buffer = *active-line / *group-status

(nnchoke-request-update-info GROUP INFO &optional SERVER)

A Gnus group info (see section 10.7.6 Group Info) is handed to the back end for alterations. This comes in handy if the back end really carries all the information (as is the case with virtual and imap groups). This function should destructively alter the info to suit its needs, and should return a non-nil value.

There should be no result data from this function.

(nnchoke-request-type GROUP &optional ARTICLE)

When the user issues commands for "sending news" (F in the summary buffer, for instance), Gnus has to know whether the article the user is following up on is news or mail. This function should return news if article in group is news, mail if it is mail and unknown if the type can't be decided. (The article parameter is necessary in nnvirtual groups which might very well combine mail groups and news groups.) Both group and article may be nil.

There should be no result data from this function.

(nnchoke-request-set-mark GROUP ACTION &optional SERVER)

Set/remove/add marks on articles. Normally Gnus handles the article marks (such as read, ticked, expired etc) internally, and store them in `~/.newsrc.eld'. Some back ends (such as IMAP) however carry all information about the articles on the server, so Gnus need to propagate the mark information to the server.

action is a list of mark setting requests, having this format:

(RANGE ACTION MARK)

range is a range of articles you wish to update marks on. action is add or del, used to add marks or remove marks (preserving all marks not mentioned). mark is a list of marks; where each mark is a symbol. Currently used marks are read, tick, reply, expire, killed, dormant, save, download, unsend, forward and recent, but your back end should, if possible, not limit itself to these.

Given contradictory actions, the last action in the list should be the effective one. That is, if your action contains a request to add the tick mark on article 1 and, later in the list, a request to remove the mark on the same article, the mark should in fact be removed.

An example action list:

(((5 12 30) 'del '(tick)) ((10 . 90) 'add '(read expire)) ((92 94) 'del '(read)))

The function should return a range of articles it wasn't able to set the mark on (currently not used for anything).

There should be no result data from this function.

(nnchoke-request-update-mark GROUP ARTICLE MARK)

If the user tries to set a mark that the back end doesn't like, this function may change the mark. Gnus will use whatever this function returns as the mark for article instead of the original mark. If the back end doesn't care, it must return the original mark, and not nil or any other type of garbage.

The only use for this I can see is what nnvirtual does with it--if a component group is auto-expirable, marking an article as read in the virtual group should result in the article being marked as expirable.

There should be no result data from this function.

(nnchoke-request-scan &optional GROUP SERVER)

This function may be called at any time (by Gnus or anything else) to request that the back end check for incoming articles, in one way or another. A mail back end will typically read the spool file or query the POP server when this function is invoked. The group doesn't have to be heeded--if the back end decides that it is too much work just scanning for a single group, it may do a total scan of all groups. It would be nice, however, to keep things local if that's practical.

There should be no result data from this function.

(nnchoke-request-group-description GROUP &optional SERVER)

The result data from this function should be a description of group.

description-line = name <TAB> description eol name = <string> description = <text>

(nnchoke-request-list-newsgroups &optional SERVER)

The result data from this function should be the description of all groups available on the server.

description-buffer = *description-line

(nnchoke-request-newgroups DATE &optional SERVER)

The result data from this function should be all groups that were created after `date', which is in normal human-readable date format (i.e., the date format used in mail and news headers, and returned by the function message-make-date by default). The data should be in the active buffer format.

It is okay for this function to return "too many" groups; some back ends might find it cheaper to return the full list of groups, rather than just the new groups. But don't do this for back ends with many groups. Normally, if the user creates the groups herself, there won't be too many groups, so nnml and the like are probably safe. But for back ends like nntp, where the groups have been created by the server, it is quite likely that there can be many groups.

(nnchoke-request-create-group GROUP &optional SERVER)

This function should create an empty group with name group.

There should be no return data.

(nnchoke-request-expire-articles ARTICLES &optional GROUP SERVER FORCE)

This function should run the expiry process on all articles in the articles range (which is currently a simple list of article numbers.) It is left up to the back end to decide how old articles should be before they are removed by this function. If force is non-nil, all articles should be deleted, no matter how new they are.

This function should return a list of articles that it did not/was not able to delete.

There should be no result data returned.

(nnchoke-request-move-article ARTICLE GROUP SERVER ACCEPT-FORM &optional LAST)

This function should move article (which is a number) from group by calling accept-form.

This function should ready the article in question for moving by removing any header lines it has added to the article, and generally should "tidy up" the article. Then it should eval accept-form in the buffer where the "tidy" article is. This will do the actual copying. If this eval returns a non-nil value, the article should be removed.

If last is nil, that means that there is a high likelihood that there will be more requests issued shortly, so that allows some optimizations.

The function should return a cons where the car is the group name and the cdr is the article number that the article was entered as.

There should be no data returned.

(nnchoke-request-accept-article GROUP &optional SERVER LAST)

This function takes the current buffer and inserts it into group. If last in nil, that means that there will be more calls to this function in short order.

The function should return a cons where the car is the group name and the cdr is the article number that the article was entered as.

The group should exist before the back end is asked to accept the article for that group.

There should be no data returned.

(nnchoke-request-replace-article ARTICLE GROUP BUFFER)

This function should remove article (which is a number) from group and insert buffer there instead.

There should be no data returned.

(nnchoke-request-delete-group GROUP FORCE &optional SERVER)

This function should delete group. If force, it should really delete all the articles in the group, and then delete the group itself. (If there is such a thing as "the group itself".)

There should be no data returned.

(nnchoke-request-rename-group GROUP NEW-NAME &optional SERVER)

This function should rename group into new-name. All articles in group should move to new-name.

There should be no data returned.

10.7.2.3 Error Messaging

The back ends should use the function nnheader-report to report error conditions--they should not raise errors when they aren't able to perform a request. The first argument to this function is the back end symbol, and the rest are interpreted as arguments to format if there are multiple of them, or just a string if there is one of them. This function must always returns nil.

(nnheader-report 'nnchoke "You did something totally bogus")

(nnheader-report 'nnchoke "Could not request group %s" group)

Gnus, in turn, will call nnheader-get-report when it gets a nil back from a server, and this function returns the most recently reported message for the back end in question. This function takes one argument--the server symbol.

Internally, these functions access back-end-status-string, so the nnchoke back end will have its error message stored in nnchoke-status-string.

10.7.2.4 Writing New Back Ends

Many back ends are quite similar. nnml is just like nnspool, but it allows you to edit the articles on the server. nnmh is just like nnml, but it doesn't use an active file, and it doesn't maintain overview databases. nndir is just like nnml, but it has no concept of "groups", and it doesn't allow editing articles.

It would make sense if it were possible to "inherit" functions from back ends when writing new back ends. And, indeed, you can do that if you want to. (You don't have to if you don't want to, of course.)

All the back ends declare their public variables and functions by using a package called nnoo.

To inherit functions from other back ends (and allow other back ends to inherit functions from the current back end), you should use the following macros:

nnoo-declare

This macro declares the first parameter to be a child of the subsequent parameters. For instance:

(nnoo-declare nndir nnml nnmh)

nndir has declared here that it intends to inherit functions from both nnml and nnmh.

defvoo

This macro is equivalent to defvar, but registers the variable as a public server variable. Most state-oriented variables should be declared with defvoo instead of defvar.

In addition to the normal defvar parameters, it takes a list of variables in the parent back ends to map the variable to when executing a function in those back ends.

(defvoo nndir-directory nil "Where nndir will look for groups." nnml-current-directory nnmh-current-directory)

This means that nnml-current-directory will be set to nndir-directory when an nnml function is called on behalf of nndir. (The same with nnmh.)

nnoo-define-basics

This macro defines some common functions that almost all back ends should have.

(nnoo-define-basics nndir)

deffoo

This macro is just like defun and takes the same parameters. In addition to doing the normal defun things, it registers the function as being public so that other back ends can inherit it.

nnoo-map-functions

This macro allows mapping of functions from the current back end to functions from the parent back ends.

(nnoo-map-functions nndir (nnml-retrieve-headers 0 nndir-current-group 0 0) (nnmh-request-article 0 nndir-current-group 0 0))

This means that when nndir-retrieve-headers is called, the first, third, and fourth parameters will be passed on to nnml-retrieve-headers, while the second parameter is set to the value of nndir-current-group.

nnoo-import

This macro allows importing functions from back ends. It should be the last thing in the source file, since it will only define functions that haven't already been defined.

(nnoo-import nndir (nnmh nnmh-request-list nnmh-request-newgroups) (nnml))

This means that calls to nndir-request-list should just be passed on to nnmh-request-list, while all public functions from nnml that haven't been defined in nndir yet should be defined now.

Below is a slightly shortened version of the nndir back end.

;;; nndir.el --- single directory newsgroup access for Gnus
;; Copyright (C) 1995,96 Free Software Foundation, Inc.

;;; Code:

(require 'nnheader)
(require 'nnmh)
(require 'nnml)
(require 'nnoo)
(eval-when-compile (require 'cl))

(nnoo-declare nndir
  nnml nnmh)

(defvoo nndir-directory nil
  "Where nndir will look for groups."
  nnml-current-directory nnmh-current-directory)

(defvoo nndir-nov-is-evil nil
  "*Non-nil means that nndir will never retrieve NOV headers."
  nnml-nov-is-evil)

(defvoo nndir-current-group ""
  nil
  nnml-current-group nnmh-current-group)
(defvoo nndir-top-directory nil nil nnml-directory nnmh-directory)
(defvoo nndir-get-new-mail nil nil nnml-get-new-mail nnmh-get-new-mail)

(defvoo nndir-status-string "" nil nnmh-status-string)
(defconst nndir-version "nndir 1.0")

;;; Interface functions.

(nnoo-define-basics nndir)

(deffoo nndir-open-server (server &optional defs)
  (setq nndir-directory
        (or (cadr (assq 'nndir-directory defs))
            server))
  (unless (assq 'nndir-directory defs)
    (push `(nndir-directory ,server) defs))
  (push `(nndir-current-group
          ,(file-name-nondirectory
            (directory-file-name nndir-directory)))
        defs)
  (push `(nndir-top-directory
          ,(file-name-directory (directory-file-name nndir-directory)))
        defs)
  (nnoo-change-server 'nndir server defs))

(nnoo-map-functions nndir
  (nnml-retrieve-headers 0 nndir-current-group 0 0)
  (nnmh-request-article 0 nndir-current-group 0 0)
  (nnmh-request-group nndir-current-group 0 0)
  (nnmh-close-group nndir-current-group 0))

(nnoo-import nndir
  (nnmh
   nnmh-status-message
   nnmh-request-list
   nnmh-request-newgroups))

(provide 'nndir)

10.7.2.5 Hooking New Back Ends Into Gnus

Having Gnus start using your new back end is rather easy--you just declare it with the gnus-declare-backend functions. This will enter the back end into the gnus-valid-select-methods variable.

gnus-declare-backend takes two parameters--the back end name and an arbitrary number of abilities.

Here's an example:

(gnus-declare-backend "nnchoke" 'mail 'respool 'address)

The above line would then go in the `nnchoke.el' file.

The abilities can be:

mail

This is a mailish back end--followups should (probably) go via mail.

post

This is a newsish back end--followups should (probably) go via news.

post-mail

This back end supports both mail and news.

none

This is neither a post nor mail back end--it's something completely different.

respool

It supports respooling--or rather, it is able to modify its source articles and groups.

address

The name of the server should be in the virtual server name. This is true for almost all back ends.

prompt-address

The user should be prompted for an address when doing commands like B in the group buffer. This is true for back ends like nntp, but not nnmbox, for instance.

10.7.2.6 Mail-like Back Ends

One of the things that separate the mail back ends from the rest of the back ends is the heavy dependence by most of the mail back ends on common functions in `nnmail.el'. For instance, here's the definition of nnml-request-scan:

(deffoo nnml-request-scan (&optional group server)
  (setq nnml-article-file-alist nil)
  (nnmail-get-new-mail 'nnml 'nnml-save-nov nnml-directory group))

It simply calls nnmail-get-new-mail with a few parameters, and nnmail takes care of all the moving and splitting of the mail.

This function takes four parameters.

method

This should be a symbol to designate which back end is responsible for the call.

exit-function

This function should be called after the splitting has been performed.

temp-directory

Where the temporary files should be stored.

group

This optional argument should be a group name if the splitting is to be performed for one group only.

nnmail-get-new-mail will call back-end-save-mail to save each article. back-end-active-number will be called to find the article number assigned to this article.

The function also uses the following variables: back-end-get-new-mail (to see whether to get new mail for this back end); and back-end-group-alist and back-end-active-file to generate the new active file. back-end-group-alist should be a group-active alist, like this:

(("a-group" (1 . 10))
 ("some-group" (34 . 39)))

10.7.3 Score File Syntax

Score files are meant to be easily parseable, but yet extremely mallable. It was decided that something that had the same read syntax as an Emacs Lisp list would fit that spec.

Here's a typical score file:

(("summary"
  ("win95" -10000 nil s)
  ("Gnus"))
 ("from"
  ("Lars" -1000))
 (mark -100))

BNF definition of a score file:

score-file      = "" / "(" *element ")"
element         = rule / atom
rule            = string-rule / number-rule / date-rule
string-rule     = "(" quote string-header quote space *string-match ")"
number-rule     = "(" quote number-header quote space *number-match ")"
date-rule       = "(" quote date-header quote space *date-match ")"
quote           = <ascii 34>
string-header   = "subject" / "from" / "references" / "message-id" /
                  "xref" / "body" / "head" / "all" / "followup"
number-header   = "lines" / "chars"
date-header     = "date"
string-match    = "(" quote <string> quote [ "" / [ space score [ "" /
                  space date [ "" / [ space string-match-t ] ] ] ] ] ")"
score           = "nil" / <integer>
date            = "nil" / <natural number>
string-match-t  = "nil" / "s" / "substring" / "S" / "Substring" /
                  "r" / "regex" / "R" / "Regex" /
                  "e" / "exact" / "E" / "Exact" /
                  "f" / "fuzzy" / "F" / "Fuzzy"
number-match    = "(" <integer> [ "" / [ space score [ "" /
                  space date [ "" / [ space number-match-t ] ] ] ] ] ")"
number-match-t  = "nil" / "=" / "<" / ">" / ">=" / "<="
date-match      = "(" quote <string> quote [ "" / [ space score [ "" /
                  space date [ "" / [ space date-match-t ] ] ] ] ")"
date-match-t    = "nil" / "at" / "before" / "after"
atom            = "(" [ required-atom / optional-atom ] ")"
required-atom   = mark / expunge / mark-and-expunge / files /
                  exclude-files / read-only / touched
optional-atom   = adapt / local / eval
mark            = "mark" space nil-or-number
nil-or-number   = "nil" / <integer>
expunge         = "expunge" space nil-or-number
mark-and-expunge = "mark-and-expunge" space nil-or-number
files           = "files" *[ space <string> ]
exclude-files   = "exclude-files" *[ space <string> ]
read-only       = "read-only" [ space "nil" / space "t" ]
adapt        = "adapt" [ space "ignore" / space "t" / space adapt-rule ]
adapt-rule      = "(" *[ <string> *[ "(" <string> <integer> ")" ] ")"
local           = "local" *[ space "(" <string> space <form> ")" ]
eval            = "eval" space <form>
space           = *[ " " / <TAB> / <NEWLINE> ]

Any unrecognized elements in a score file should be ignored, but not discarded.

As you can see, white space is needed, but the type and amount of white space is irrelevant. This means that formatting of the score file is left up to the programmer--if it's simpler to just spew it all out on one looong line, then that's ok.

The meaning of the various atoms are explained elsewhere in this manual (see section 7.4 Score File Format).

10.7.4 Headers

Internally Gnus uses a format for storing article headers that corresponds to the NOV format in a mysterious fashion. One could almost suspect that the author looked at the NOV specification and just shamelessly stole the entire thing, and one would be right.

Header is a severely overloaded term. "Header" is used in RFC 1036 to talk about lines in the head of an article (e.g., From). It is used by many people as a synonym for "head"---"the header and the body". (That should be avoided, in my opinion.) And Gnus uses a format internally that it calls "header", which is what I'm talking about here. This is a 9-element vector, basically, with each header (ouch) having one slot.

These slots are, in order: number, subject, from, date, id, references, chars, lines, xref, and extra. There are macros for accessing and setting these slots--they all have predictable names beginning with mail-header- and mail-header-set-, respectively.

All these slots contain strings, except the extra slot, which contains an alist of header/value pairs (see section 3.1.2 To From Newsgroups).

10.7.5 Ranges

GNUS introduced a concept that I found so useful that I've started using it a lot and have elaborated on it greatly.

The question is simple: If you have a large amount of objects that are identified by numbers (say, articles, to take a wild example) that you want to qualify as being "included", a normal sequence isn't very useful. (A 200,000 length sequence is a bit long-winded.)

The solution is as simple as the question: You just collapse the sequence.

(1 2 3 4 5 6 10 11 12)

is transformed into

((1 . 6) (10 . 12))

To avoid having those nasty `(13 . 13)' elements to denote a lonesome object, a `13' is a valid element:

((1 . 6) 7 (10 . 12))

This means that comparing two ranges to find out whether they are equal is slightly tricky:

((1 . 5) 7 8 (10 . 12))

and

((1 . 5) (7 . 8) (10 . 12))

are equal. In fact, any non-descending list is a range:

(1 2 3 4 5)

is a perfectly valid range, although a pretty long-winded one. This is also valid:

(1 . 5)

and is equal to the previous range.

Here's a BNF definition of ranges. Of course, one must remember the semantic requirement that the numbers are non-descending. (Any number of repetition of the same number is allowed, but apt to disappear in range handling.)

range           = simple-range / normal-range
simple-range    = "(" number " . " number ")"
normal-range    = "(" start-contents ")"
contents        = "" / simple-range *[ " " contents ] /
                  number *[ " " contents ]

Gnus currently uses ranges to keep track of read articles and article marks. I plan on implementing a number of range operators in C if The Powers That Be are willing to let me. (I haven't asked yet, because I need to do some more thinking on what operators I need to make life totally range-based without ever having to convert back to normal sequences.)

10.7.6 Group Info

Gnus stores all permanent info on groups in a group info list. This list is from three to six elements (or more) long and exhaustively describes the group.

Here are two example group infos; one is a very simple group while the second is a more complex one:

("no.group" 5 ((1 . 54324)))

("nnml:my.mail" 3 ((1 . 5) 9 (20 . 55))
                ((tick (15 . 19)) (replied 3 6 (19 . 3)))
                (nnml "")
                ((auto-expire . t) (to-address . "ding@gnus.org")))

The first element is the group name---as Gnus knows the group, anyway. The second element is the subscription level, which normally is a small integer. (It can also be the rank, which is a cons cell where the car is the level and the cdr is the score.) The third element is a list of ranges of read articles. The fourth element is a list of lists of article marks of various kinds. The fifth element is the select method (or virtual server, if you like). The sixth element is a list of group parameters, which is what this section is about.

Any of the last three elements may be missing if they are not required. In fact, the vast majority of groups will normally only have the first three elements, which saves quite a lot of cons cells.

Here's a BNF definition of the group info format:

info          = "(" group space ralevel space read
                [ "" / [ space marks-list [ "" / [ space method [ "" /
                space parameters ] ] ] ] ] ")"
group         = quote <string> quote
ralevel       = rank / level
level         = <integer in the range of 1 to inf>
rank          = "(" level "." score ")"
score         = <integer in the range of 1 to inf>
read          = range
marks-lists   = nil / "(" *marks ")"
marks         = "(" <string> range ")"
method        = "(" <string> *elisp-forms ")"
parameters    = "(" *elisp-forms ")"

Actually that `marks' rule is a fib. A `marks' is a `<string>' consed on to a `range', but that's a bitch to say in pseudo-BNF.

If you have a Gnus info and want to access the elements, Gnus offers a series of macros for getting/setting these elements.

gnus-info-group

gnus-info-set-group

Get/set the group name.

gnus-info-rank

gnus-info-set-rank

Get/set the group rank (see section 2.7 Group Score).

gnus-info-level

gnus-info-set-level

Get/set the group level.

gnus-info-score

gnus-info-set-score

Get/set the group score (see section 2.7 Group Score).

gnus-info-read

gnus-info-set-read

Get/set the ranges of read articles.

gnus-info-marks

gnus-info-set-marks

Get/set the lists of ranges of marked articles.

gnus-info-method

gnus-info-set-method

Get/set the group select method.

gnus-info-params

gnus-info-set-params

Get/set the group parameters.

All the getter functions take one parameter--the info list. The setter functions take two parameters--the info list and the new value.

The last three elements in the group info aren't mandatory, so it may be necessary to extend the group info before setting the element. If this is necessary, you can just pass on a non-nil third parameter to the three final setter functions to have this happen automatically.

10.7.7 Extended Interactive

Gnus extends the standard Emacs interactive specification slightly to allow easy use of the symbolic prefix (see section 8.3 Symbolic Prefixes). Here's an example of how this is used:

(defun gnus-summary-increase-score (&optional score symp)
  (interactive (gnus-interactive "P\ny"))
  ...
  )

The best thing to do would have been to implement gnus-interactive as a macro which would have returned an interactive form, but this isn't possible since Emacs checks whether a function is interactive or not by simply doing an assq on the lambda form. So, instead we have gnus-interactive function that takes a string and returns values that are usable to interactive.

This function accepts (almost) all normal interactive specs, but adds a few more.

`y'

The current symbolic prefix--the gnus-current-prefix-symbol variable.

`Y'

A list of the current symbolic prefixes--the gnus-current-prefix-symbol variable.

`A'

The current article number--the gnus-summary-article-number function.

`H'

The current article header--the gnus-summary-article-header function.

`g'

The current group name--the gnus-group-group-name function.

10.7.8 Emacs/XEmacs Code

While Gnus runs under Emacs, XEmacs and Mule, I decided that one of the platforms must be the primary one. I chose Emacs. Not because I don't like XEmacs or Mule, but because it comes first alphabetically.

This means that Gnus will byte-compile under Emacs with nary a warning, while XEmacs will pump out gigabytes of warnings while byte-compiling. As I use byte-compilation warnings to help me root out trivial errors in Gnus, that's very useful.

I've also consistently used Emacs function interfaces, but have used Gnusey aliases for the functions. To take an example: Emacs defines a run-at-time function while XEmacs defines a start-itimer function. I then define a function called gnus-run-at-time that takes the same parameters as the Emacs run-at-time. When running Gnus under Emacs, the former function is just an alias for the latter. However, when running under XEmacs, the former is an alias for the following function:

(defun gnus-xmas-run-at-time (time repeat function &rest args)
  (start-itimer
   "gnus-run-at-time"
   `(lambda ()
      (,function ,@args))
   time repeat))

This sort of thing has been done for bunches of functions. Gnus does not redefine any native Emacs functions while running under XEmacs--it does this defalias thing with Gnus equivalents instead. Cleaner all over.

In the cases where the XEmacs function interface was obviously cleaner, I used it instead. For example gnus-region-active-p is an alias for region-active-p in XEmacs, whereas in Emacs it is a function.

Of course, I could have chosen XEmacs as my native platform and done mapping functions the other way around. But I didn't. The performance hit these indirections impose on Gnus under XEmacs should be slight.

10.7.9 Various File Formats

10.7.9.1 Active File Format

The active file lists all groups available on the server in question. It also lists the highest and lowest current article numbers in each group.

Here's an excerpt from a typical active file:

soc.motss 296030 293865 y
alt.binaries.pictures.fractals 3922 3913 n
comp.sources.unix 1605 1593 m
comp.binaries.ibm.pc 5097 5089 y
no.general 1000 900 y

Here's a pseudo-BNF definition of this file:

active      = *group-line
group-line  = group spc high-number spc low-number spc flag <NEWLINE>
group       = <non-white-space string>
spc         = " "
high-number = <non-negative integer>
low-number  = <positive integer>
flag        = "y" / "n" / "m" / "j" / "x" / "=" group

For a full description of this file, see the manual pages for `innd', in particular `active(5)'.

10.7.9.2 Newsgroups File Format

The newsgroups file lists groups along with their descriptions. Not all groups on the server have to be listed, and not all groups in the file have to exist on the server. The file is meant purely as information to the user.

The format is quite simple; a group name, a tab, and the description. Here's the definition:

newsgroups    = *line
line          = group tab description <NEWLINE>
group         = <non-white-space string>
tab           = <TAB>
description   = <string>

10.8 Emacs for Heathens

Believe it or not, but some people who use Gnus haven't really used Emacs much before they embarked on their journey on the Gnus Love Boat. If you are one of those unfortunates whom "C-M-a", "kill the region", and "set gnus-flargblossen to an alist where the key is a regexp that is used for matching on the group name" are magical phrases with little or no meaning, then this appendix is for you. If you are already familiar with Emacs, just ignore this and go fondle your cat instead.

10.8.1 Keystrokes

• Q: What is an experienced Emacs user?

• A: A person who wishes that the terminal had pedals.

Yes, when you use Emacs, you are apt to use the control key, the shift key and the meta key a lot. This is very annoying to some people (notably vile users), and the rest of us just love the hell out of it. Just give up and submit. Emacs really does stand for "Escape-Meta-Alt-Control-Shift", and not "Editing Macros", as you may have heard from other disreputable sources (like the Emacs author).

The shift keys are normally located near your pinky fingers, and are normally used to get capital letters and stuff. You probably use it all the time. The control key is normally marked "CTRL" or something like that. The meta key is, funnily enough, never marked as such on any keyboard. The one I'm currently at has a key that's marked "Alt", which is the meta key on this keyboard. It's usually located somewhere to the left hand side of the keyboard, usually on the bottom row.

Now, us Emacs people don't say "press the meta-control-m key", because that's just too inconvenient. We say "press the C-M-m key". M- is the prefix that means "meta" and "C-" is the prefix that means "control". So "press C-k" means "press down the control key, and hold it down while you press k". "Press C-M-k" means "press down and hold down the meta key and the control key and then press k". Simple, ay?

This is somewhat complicated by the fact that not all keyboards have a meta key. In that case you can use the "escape" key. Then M-k means "press escape, release escape, press k". That's much more work than if you have a meta key, so if that's the case, I respectfully suggest you get a real keyboard with a meta key. You can't live without it.

10.8.2 Emacs Lisp

Emacs is the King of Editors because it's really a Lisp interpreter. Each and every key you tap runs some Emacs Lisp code snippet, and since Emacs Lisp is an interpreted language, that means that you can configure any key to run any arbitrary code. You just, like, do it.

Gnus is written in Emacs Lisp, and is run as a bunch of interpreted functions. (These are byte-compiled for speed, but it's still interpreted.) If you decide that you don't like the way Gnus does certain things, it's trivial to have it do something a different way. (Well, at least if you know how to write Lisp code.) However, that's beyond the scope of this manual, so we are simply going to talk about some common constructs that you normally use in your `~/.gnus.el' file to customize Gnus. (You can also use the `~/.emacs' file, but in order to set things of Gnus up, it is much better to use the `~/.gnus.el' file, See section 1.7 Startup Files.)

If you want to set the variable gnus-florgbnize to four (4), you write the following:

(setq gnus-florgbnize 4)

This function (really "special form") setq is the one that can set a variable to some value. This is really all you need to know. Now you can go and fill your `~/.gnus.el' file with lots of these to change how Gnus works.

If you have put that thing in your `~/.gnus.el' file, it will be read and evaled (which is Lisp-ese for "run") the next time you start Gnus. If you want to change the variable right away, simply say C-x C-e after the closing parenthesis. That will eval the previous "form", which is a simple setq statement here.

Go ahead--just try it, if you're located at your Emacs. After you C-x C-e, you will see `4' appear in the echo area, which is the return value of the form you evaled.

Some pitfalls:

If the manual says "set gnus-read-active-file to some", that means:

(setq gnus-read-active-file 'some)

On the other hand, if the manual says "set gnus-nntp-server to `nntp.ifi.uio.no'", that means:

(setq gnus-nntp-server "nntp.ifi.uio.no")

So be careful not to mix up strings (the latter) with symbols (the former). The manual is unambiguous, but it can be confusing.

10.9 Frequently Asked Questions

Abstract

This is the new Gnus Frequently Asked Questions list. If you have a Web browser, the official hypertext version is at http://my.gnus.org/FAQ/, the Docbook source is available from http://sourceforge.net.

Please submit features and suggestions to the FAQ discussion list. The list is protected against junk mail with qconfirm. As a subscriber, your submissions will automatically pass. You can also subscribe to the list by sending a blank email to faq-discuss-subscribe@my.gnus.org and browse the archive (BROKEN).

Changes

• Updated FAQ to reflect release of Gnus 5.10 and start of No Gnus development.

Introduction

This is the Gnus Frequently Asked Questions list.

Gnus is a Usenet Newsreader and Electronic Mail User Agent implemented as a part of Emacs. It's been around in some form for almost a decade now, and has been distributed as a standard part of Emacs for much of that time. Gnus 5 is the latest (and greatest) incarnation. The original version was called GNUS, and was written by Masanobu UMEDA. When autumn crept up in '94, Lars Magne Ingebrigtsen grew bored and decided to rewrite Gnus.

Its biggest strength is the fact that it is extremely customizable. It is somewhat intimidating at first glance, but most of the complexity can be ignored until you're ready to take advantage of it. If you receive a reasonable volume of e-mail (you're on various mailing lists), or you would like to read high-volume mailing lists but cannot keep up with them, or read high volume newsgroups or are just bored, then Gnus is what you want.

This FAQ was maintained by Justin Sheehy until March 2002. He would like to thank Steve Baur and Per Abrahamsen for doing a wonderful job with this FAQ before him. We would like to do the same - thanks, Justin!

If you have a Web browser, the official hypertext version is at: http://my.gnus.org/FAQ/. This version is much nicer than the unofficial hypertext versions that are archived at Utrecht, Oxford, Smart Pages, Ohio State, and other FAQ archives. See the resources question below if you want information on obtaining it in another format.

The information contained here was compiled with the assistance of the Gnus development mailing list, and any errors or misprints are the my.gnus.org team's fault, sorry.

10.9.1 Installation FAQ

Question 1.1

What is the latest version of Gnus?

Answer

Jingle please: Gnus 5.10 is released, get it while it's hot! As well as the step in version number is rather small, Gnus 5.10 has tons of new features which you shouldn't miss. The current release (5.10.8) should be at least as stable as the latest release of the 5.8 series.

Question 1.2

What's new in 5.10?

Answer

First of all, you should have a look into the file GNUS-NEWS in the toplevel directory of the Gnus tarball, there the most important changes are listed. Here's a short list of the changes I find especially important/interesting:

• Major rewrite of the Gnus agent, Gnus agent is now active by default.

• Many new article washing functions for dealing with ugly formatted articles.

• Anti Spam features.

• Message-utils now included in Gnus.

• New format specifiers for summary lines, e.g. %B for a complex trn-style thread tree.

Question 1.3

Where and how to get Gnus?

Answer

The latest released version of Gnus isn't included in Emacs 21, therefor you should get the Gnus tarball from http://www.gnus.org/dist/gnus.tar.gz or via anonymous FTP from ftp://ftp.gnus.org/pub/gnus/gnus.tar.gz. If you use XEmacs instead of Emacs you can use XEmacs' package system instead.

Question 1.4

What to do with the tarball now?

Answer

Untar it via `tar xvzf gnus.tar.gz' and do the common `./configure; make; make install' circle. (under MS-Windows either get the Cygwin environment from http://www.cygwin.com which allows you to do what's described above or unpack the tarball with some packer (e.g. Winace from http://www.winace.com) and use the batch-file make.bat included in the tarball to install Gnus.) If you don't want to (or aren't allowed to) install Gnus system-wide, you can install it in your home directory and add the following lines to your ~/.xemacs/init.el or ~/.emacs:

(add-to-list 'load-path "/path/to/gnus/lisp")
(if (featurep 'xemacs)
    (add-to-list 'Info-directory-list "/path/to/gnus/texi/")
  (add-to-list 'Info-default-directory-list "/path/to/gnus/texi/"))

Make sure that you don't have any Gnus related stuff before this line, on MS Windows use something like "C:/path/to/lisp" (yes, "/").

Question 1.5

I sometimes read references to No Gnus and Oort Gnus, what are those?

Answer

Oort Gnus was the name of the development version of Gnus, which became Gnus 5.10 in autumn 2003. No Gnus is the name of the current development version which will once become Gnus 5.12 or Gnus 6. (If you're wondering why not 5.11, the odd version numbers are normally used for the Gnus versions bundled with Emacs)

Question 1.6

Which version of Emacs do I need?

Answer

Gnus 5.10 requires an Emacs version that is greater than or equal to Emacs 20.7 or XEmacs 21.1. The development versions of Gnus (aka No Gnus) requires Emacs 21 or XEmacs 21.4.

Question 1.7

How do I run Gnus on both Emacs and XEmacs?

Answer

You can't use the same copy of Gnus in both as the Lisp files are byte-compiled to a format which is different depending on which Emacs did the compilation. Get one copy of Gnus for Emacs and one for XEmacs.

10.9.2 Startup / Group buffer

Question 2.1

Every time I start Gnus I get a message "Gnus auto-save file exists. Do you want to read it?", what does this mean and how to prevent it?

Answer

This message means that the last time you used Gnus, it wasn't properly exited and therefor couldn't write its informations to disk (e.g. which messages you read), you are now asked if you want to restore those informations from the auto-save file.

To prevent this message make sure you exit Gnus via `q' in group buffer instead of just killing Emacs.

Question 2.2

Gnus doesn't remember which groups I'm subscribed to, what's this?

Answer

You get the message described in the q/a pair above while starting Gnus, right? It's an other symptom for the same problem, so read the answer above.

Question 2.3

How to change the format of the lines in Group buffer?

Answer

You've got to tweak the value of the variable gnus-group-line-format. See the manual node "Group Line Specification" for information on how to do this. An example for this (guess from whose .gnus :-)):

(setq gnus-group-line-format "%P%M%S[%5t]%5y : %(%g%)\n")

Question 2.4

My group buffer becomes a bit crowded, is there a way to sort my groups into categories so I can easier browse through them?

Answer

Gnus offers the topic mode, it allows you to sort your groups in, well, topics, e.g. all groups dealing with Linux under the topic linux, all dealing with music under the topic music and all dealing with scottish music under the topic scottish which is a subtopic of music.

To enter topic mode, just hit t while in Group buffer. Now you can use `T n' to create a topic at point and `T m' to move a group to a specific topic. For more commands see the manual or the menu. You might want to include the %P specifier at the beginning of your gnus-group-line-format variable to have the groups nicely indented.

Question 2.5

How to manually sort the groups in Group buffer? How to sort the groups in a topic?

Answer

Move point over the group you want to move and hit `C-k', now move point to the place where you want the group to be and hit `C-y'.

10.9.3 Getting Messages

Question 3.1

I just installed Gnus, started it via `M-x gnus' but it only says "nntp (news) open error", what to do?

Answer

You've got to tell Gnus where to fetch the news from. Read the documentation for information on how to do this. As a first start, put those lines in ~/.gnus.el:

(setq gnus-select-method '(nntp "news.yourprovider.net"))
(setq user-mail-address "you@yourprovider.net")
(setq user-full-name "Your Name")

Question 3.2

I'm working under Windows and have no idea what ~/.gnus.el means.

Answer

The ~/ means the home directory where Gnus and Emacs look for the configuration files. However, you don't really need to know what this means, it suffices that Emacs knows what it means :-) You can type `C-x C-f ~/.gnus.el RET ' (yes, with the forward slash, even on Windows), and Emacs will open the right file for you. (It will most likely be new, and thus empty.) However, I'd discourage you from doing so, since the directory Emacs chooses will most certainly not be what you want, so let's do it the correct way. The first thing you've got to do is to create a suitable directory (no blanks in directory name please) e.g. c:\myhome. Then you must set the environment variable HOME to this directory. To do this under Win9x or Me include the line

SET HOME=C:\myhome

in your autoexec.bat and reboot. Under NT, 2000 and XP, hit Winkey+Pause/Break to enter system options (if it doesn't work, go to Control Panel -> System). There you'll find the possibility to set environment variables, create a new one with name HOME and value C:\myhome, a reboot is not necessary.

Now to create ~/.gnus.el, say `C-x C-f ~/.gnus.el RET C-x C-s'. in Emacs.

Question 3.3

My news server requires authentication, how to store user name and password on disk?

Answer

Create a file ~/.authinfo which includes for each server a line like this

machine news.yourprovider.net login YourUserName password YourPassword

chmod 600 ~/.authinfo

in a shell.)

Question 3.4

Gnus seems to start up OK, but I can't find out how to subscribe to a group.

Answer

If you know the name of the group say `U name.of.group RET' in group buffer (use the tab-completion Luke). Otherwise hit ^ in group buffer, this brings you to the server buffer. Now place point (the cursor) over the server which carries the group you want, hit `RET', move point to the group you want to subscribe to and say `u' to subscribe to it.

Question 3.5

Gnus doesn't show all groups / Gnus says I'm not allowed to post on this server as well as I am, what's that?

Answer

Some providers allow restricted anonymous access and full access only after authorization. To make Gnus send authinfo to those servers append

force yes

Question 3.6

I want Gnus to fetch news from several servers, is this possible?

Answer

Of course. You can specify more sources for articles in the variable gnus-secondary-select-methods. Add something like this in ~/.gnus.el:

(add-to-list 'gnus-secondary-select-methods
             '(nntp "news.yourSecondProvider.net"))
(add-to-list 'gnus-secondary-select-methods
             '(nntp "news.yourThirdProvider.net"))

Question 3.7

And how about local spool files?

Answer

No problem, this is just one more select method called nnspool, so you want this:

(add-to-list 'gnus-secondary-select-methods '(nnspool ""))

Or this if you don't want an NNTP Server as primary news source:

(setq gnus-select-method '(nnspool ""))

Gnus will look for the spool file in /usr/spool/news, if you want something different, change the line above to something like this:

(add-to-list 'gnus-secondary-select-methods
             '(nnspool ""
		       (nnspool-directory "/usr/local/myspoolddir")))

This sets the spool directory for this server only. You might have to specify more stuff like the program used to post articles, see the Gnus manual on how to do this.

Question 3.8

OK, reading news works now, but I want to be able to read my mail with Gnus, too. How to do it?

Answer

That's a bit harder since there are many possible sources for mail, many possible ways for storing mail and many different ways for sending mail. The most common cases are these: 1: You want to read your mail from a pop3 server and send them directly to a SMTP Server 2: Some program like fetchmail retrieves your mail and stores it on disk from where Gnus shall read it. Outgoing mail is sent by Sendmail, Postfix or some other MTA. Sometimes, you even need a combination of the above cases.

However, the first thing to do is to tell Gnus in which way it should store the mail, in Gnus terminology which back end to use. Gnus supports many different back ends, the most commonly used one is nnml. It stores every mail in one file and is therefor quite fast. However you might prefer a one file per group approach if your file system has problems with many small files, the nnfolder back end is then probably the choice for you. To use nnml add the following to ~/.gnus.el:

(add-to-list 'gnus-secondary-select-methods '(nnml ""))

As you might have guessed, if you want nnfolder, it's

(add-to-list 'gnus-secondary-select-methods '(nnfolder ""))

Now we need to tell Gnus, where to get it's mail from. If it's a POP3 server, then you need something like this:

(eval-after-load "mail-source"
  '(add-to-list 'mail-sources '(pop :server "pop.YourProvider.net"
                                    :user "yourUserName"
                                    :password "yourPassword")))

Make sure ~/.gnus.el isn't readable to others if you store your password there. If you want to read your mail from a traditional spool file on your local machine, it's

(eval-after-load "mail-source"
  '(add-to-list 'mail-sources '(file :path "/path/to/spool/file"))

If it's a Maildir, with one file per message as used by postfix, Qmail and (optionally) fetchmail it's

(eval-after-load "mail-source"
  '(add-to-list 'mail-sources '(maildir :path "/path/to/Maildir/"
                                        :subdirs ("cur" "new")))

And finally if you want to read your mail from several files in one directory, for example because procmail already split your mail, it's

(eval-after-load "mail-source"
  '(add-to-list 'mail-sources
		'(directory :path "/path/to/procmail-dir/"
			    :suffix ".prcml")))

Where :suffix ".prcml" tells Gnus only to use files with the suffix .prcml.

OK, now you only need to tell Gnus how to send mail. If you want to send mail via sendmail (or whichever MTA is playing the role of sendmail on your system), you don't need to do anything. However, if you want to send your mail to an SMTP Server you need the following in your ~/.gnus.el

(setq send-mail-function 'smtpmail-send-it)
(setq message-send-mail-function 'smtpmail-send-it)
(setq smtpmail-default-smtp-server "smtp.yourProvider.net")

Question 3.9

And what about IMAP?

Answer

There are two ways of using IMAP with Gnus. The first one is to use IMAP like POP3, that means Gnus fetches the mail from the IMAP server and stores it on disk. If you want to do this (you don't really want to do this) add the following to ~/.gnus.el

(add-to-list 'mail-sources '(imap :server "mail.mycorp.com"
                                  :user "username"
                                  :pass "password"
                                  :stream network
                                  :authentication login
                                  :mailbox "INBOX"
                                  :fetchflag "\\Seen"))

You might have to tweak the values for stream and/or authentification, see the Gnus manual node "Mail Source Specifiers" for possible values.

If you want to use IMAP the way it's intended, you've got to follow a different approach. You've got to add the nnimap back end to your select method and give the information about the server there.

(add-to-list 'gnus-secondary-select-methods
	     '(nnimap "Give the baby a name"
		      (nnimap-address "imap.yourProvider.net")
		      (nnimap-port 143)
		      (nnimap-list-pattern "archive.*")))

Again, you might have to specify how to authenticate to the server if Gnus can't guess the correct way, see the Manual Node "IMAP" for detailed information.

Question 3.10

At the office we use one of those MS Exchange servers, can I use Gnus to read my mail from it?

Answer

Offer your administrator a pair of new running shoes for activating IMAP on the server and follow the instructions above.

Question 3.11

Can I tell Gnus not to delete the mails on the server it retrieves via POP3?

Answer

First of all, that's not the way POP3 is intended to work, if you have the possibility, you should use the IMAP Protocol if you want your messages to stay on the server. Nevertheless there might be situations where you need the feature, but sadly Gnus itself has no predefined functionality to do so.

However this is Gnus county so there are possibilities to achieve what you want. The easiest way is to get an external program which retrieves copies of the mail and stores them on disk, so Gnus can read it from there. On Unix systems you could use e.g. fetchmail for this, on MS Windows you can use Hamster, an excellent local news and mail server.

The other solution would be, to replace the method Gnus uses to get mail from POP3 servers by one which is capable of leaving the mail on the server. If you use XEmacs, get the package mail-lib, it includes an enhanced pop3.el, look in the file, there's documentation on how to tell Gnus to use it and not to delete the retrieved mail. For GNU Emacs look for the file epop3.el which can do the same (If you know the home of this file, please send me an e-mail). You can also tell Gnus to use an external program (e.g. fetchmail) to fetch your mail, see the info node "Mail Source Specifiers" in the Gnus manual on how to do it.

10.9.4 Reading messages

Question 4.1

When I enter a group, all read messages are gone. How to view them again?

Answer

If you enter the group by saying `RET' in group buffer with point over the group, only unread and ticked messages are loaded. Say `C-u RET' instead to load all available messages. If you want only the e.g. 300 newest say `C-u 300 RET'

Loading only unread messages can be annoying if you have threaded view enabled, say

(setq gnus-fetch-old-headers 'some)

If you already use Gnus 5.10, you can say `/o N' In summary buffer to load the last N messages, this feature is not available in 5.8.8

If you don't want all old messages, but the parent of the message you're just reading, you can say `^', if you want to retrieve the whole thread the message you're just reading belongs to, `A T' is your friend.

Question 4.2

How to tell Gnus to show an important message every time I enter a group, even when it's read?

Answer

You can tick important messages. To do this hit `u' while point is in summary buffer over the message. When you want to remove the mark, hit either `d' (this deletes the tick mark and set's unread mark) or `M c' (which deletes all marks for the message).

Question 4.3

How to view the headers of a message?

Answer

Say `t' to show all headers, one more `t' hides them again.

Question 4.4

How to view the raw unformatted message?

Answer

Say `C-u g' to show the raw message `g' returns to normal view.

Question 4.5

How can I change the headers Gnus displays by default at the top of the article buffer?

Answer

The variable gnus-visible-headers controls which headers are shown, its value is a regular expression, header lines which match it are shown. So if you want author, subject, date, and if the header exists, Followup-To and MUA / NUA say this in ~/.gnus.el:

(setq gnus-visible-headers
      '("^From" "^Subject" "^Date" "^Newsgroups" "^Followup-To"
	"^User-Agent" "^X-Newsreader" "^X-Mailer"))

Question 4.6

I'd like Gnus NOT to render HTML-mails but show me the text part if it's available. How to do it?

Answer

Say

(eval-after-load "mm-decode"
 '(progn 
      (add-to-list 'mm-discouraged-alternatives "text/html")
      (add-to-list 'mm-discouraged-alternatives "text/richtext")))

in ~/.gnus.el. If you don't want HTML rendered, even if there's no text alternative add

(setq mm-automatic-display (remove "text/html" mm-automatic-display))

too.

Question 4.7

Can I use some other browser than w3 to render my HTML-mails?

Answer

Only if you use Gnus 5.10 or younger. In this case you've got the choice between w3, w3m, links, lynx and html2text, which one is used can be specified in the variable mm-text-html-renderer, so if you want links to render your mail say

(setq mm-text-html-renderer 'links)

Question 4.8

Is there anything I can do to make poorly formatted mails more readable?

Answer

Gnus offers you several functions to "wash" incoming mail, you can find them if you browse through the menu, item Article->Washing. The most interesting ones are probably "Wrap long lines" (`W w'), "Decode ROT13" (`W r') and "Outlook Deuglify" which repairs the dumb quoting used by many users of Microsoft products (`W Y f' gives you full deuglify. See `W Y C-h' or have a look at the menus for other deuglifications). Outlook deuglify is only available since Gnus 5.10.

Question 4.9

Is there a way to automatically ignore posts by specific authors or with specific words in the subject? And can I highlight more interesting ones in some way?

Answer

You want Scoring. Scoring means, that you define rules which assign each message an integer value. Depending on the value the message is highlighted in summary buffer (if it's high, say +2000) or automatically marked read (if the value is low, say -800) or some other action happens.

There are basically three ways of setting up rules which assign the scoring-value to messages. The first and easiest way is to set up rules based on the article you are just reading. Say you're reading a message by a guy who always writes nonsense and you want to ignore his messages in the future. Hit `L', to set up a rule which lowers the score. Now Gnus asks you which the criteria for lowering the Score shall be. Hit `?' twice to see all possibilities, we want `a' which means the author (the from header). Now Gnus wants to know which kind of matching we want. Hit either `e' for an exact match or `s' for substring-match and delete afterwards everything but the name to score down all authors with the given name no matter which email address is used. Now you need to tell Gnus when to apply the rule and how long it should last, hit e.g. `p' to apply the rule now and let it last forever. If you want to raise the score instead of lowering it say `I' instead of `L'.

You can also set up rules by hand. To do this say `V f' in summary buffer. Then you are asked for the name of the score file, it's name.of.group.SCORE for rules valid in only one group or all.Score for rules valid in all groups. See the Gnus manual for the exact syntax, basically it's one big list whose elements are lists again. the first element of those lists is the header to score on, then one more list with what to match, which score to assign, when to expire the rule and how to do the matching. If you find me very interesting, you could e.g. add the following to your all.Score:

(("references" ("hschmi22.userfqdn.rz-online.de" 500 nil s))
 ("message-id" ("hschmi22.userfqdn.rz-online.de" 999 nil s)))

This would add 999 to the score of messages written by me and 500 to the score of messages which are a (possibly indirect) answer to a message written by me. Of course nobody with a sane mind would do this :-)

The third alternative is adaptive scoring. This means Gnus watches you and tries to find out what you find interesting and what annoying and sets up rules which reflect this. Adaptive scoring can be a huge help when reading high traffic groups. If you want to activate adaptive scoring say

(setq gnus-use-adaptive-scoring t)

in ~/.gnus.el.

Question 4.10

How can I disable threading in some (e.g. mail-) groups, or set other variables specific for some groups?

Answer

While in group buffer move point over the group and hit `G c', this opens a buffer where you can set options for the group. At the bottom of the buffer you'll find an item that allows you to set variables locally for the group. To disable threading enter gnus-show-threads as name of variable and nil as value. Hit button done at the top of the buffer when you're ready.

Question 4.11

Can I highlight messages written by me and follow-ups to those?

Answer

Stop those "Can I ..." questions, the answer is always yes in Gnus Country :-). It's a three step process: First we make faces (specifications of how summary-line shall look like) for those postings, then we'll give them some special score and finally we'll tell Gnus to use the new faces. You can find detailed instructions on how to do it on my.gnus.org

Question 4.12

The number of total messages in a group which Gnus displays in group buffer is by far to high, especially in mail groups. Is this a bug?

Answer

No, that's a matter of design of Gnus, fixing this would mean reimplementation of major parts of Gnus' back ends. Gnus thinks "highest-article-number - lowest-article-number = total-number-of-articles". This works OK for Usenet groups, but if you delete and move many messages in mail groups, this fails. To cure the symptom, enter the group via `C-u RET' (this makes Gnus get all messages), then hit `M P b' to mark all messages and then say `B m name.of.group' to move all messages to the group they have been in before, they get new message numbers in this process and the count is right again (until you delete and move your mail to other groups again).

Question 4.13

I don't like the layout of summary and article buffer, how to change it? Perhaps even a three pane display?

Answer

You can control the windows configuration by calling the function gnus-add-configuration. The syntax is a bit complicated but explained very well in the manual node "Window Layout". Some popular examples:

Instead 25% summary 75% article buffer 35% summary and 65% article (the 1.0 for article means "take the remaining space"):

(gnus-add-configuration
 '(article (vertical 1.0 (summary .35 point) (article 1.0))))

A three pane layout, Group buffer on the left, summary buffer top-right, article buffer bottom-right:

(gnus-add-configuration
 '(article
   (horizontal 1.0
	       (vertical 25
			 (group 1.0))
	       (vertical 1.0
			 (summary 0.25 point)
			 (article 1.0)))))
(gnus-add-configuration
 '(summary
   (horizontal 1.0
	       (vertical 25
			 (group 1.0))
	       (vertical 1.0
			 (summary 1.0 point)))))

Question 4.14

I don't like the way the Summary buffer looks, how to tweak it?

Answer

You've got to play around with the variable gnus-summary-line-format. It's value is a string of symbols which stand for things like author, date, subject etc. A list of the available specifiers can be found in the manual node "Summary Buffer Lines" and the often forgotten node "Formatting Variables" and it's sub-nodes. There you'll find useful things like positioning the cursor and tabulators which allow you a summary in table form, but sadly hard tabulators are broken in 5.8.8.

Since 5.10, Gnus offers you some very nice new specifiers, e.g. %B which draws a thread-tree and %&user-date which gives you a date where the details are dependent of the articles age. Here's an example which uses both:

(setq gnus-summary-line-format ":%U%R %B %s %-60=|%4L |%-20,20f |%&user-date; \n")

resulting in:

:O     Re: [Richard Stallman] rfc2047.el          |  13 |Lars Magne Ingebrigt |Sat 23:06
:O     Re: Revival of the ding-patches list       |  13 |Lars Magne Ingebrigt |Sat 23:12
:R  >  Re: Find correct list of articles for a gro|  25 |Lars Magne Ingebrigt |Sat 23:16
:O  \->  ...                                      |  21 |Kai Grossjohann      | 0:01
:R  >  Re: Cry for help: deuglify.el - moving stuf|  28 |Lars Magne Ingebrigt |Sat 23:34
:O  \->  ...                                      | 115 |Raymond Scholz       | 1:24
:O    \->  ...                                    |  19 |Lars Magne Ingebrigt |15:33
:O     Slow mailing list                          |  13 |Lars Magne Ingebrigt |Sat 23:49
:O     Re: `@' mark not documented                |  13 |Lars Magne Ingebrigt |Sat 23:50
:R  >  Re: Gnus still doesn't count messages prope|  23 |Lars Magne Ingebrigt |Sat 23:57
:O  \->  ...                                      |  18 |Kai Grossjohann      | 0:35
:O    \->  ...                                    |  13 |Lars Magne Ingebrigt | 0:56

Question 4.15

How to split incoming mails in several groups?

Answer

Gnus offers two possibilities for splitting mail, the easy nnmail-split-methods and the more powerful Fancy Mail Splitting. I'll only talk about the first one, refer to the manual, node "Fancy Mail Splitting" for the latter.

The value of nnmail-split-methods is a list, each element is a list which stands for a splitting rule. Each rule has the form "group where matching articles should go to", "regular expression which has to be matched", the first rule which matches wins. The last rule must always be a general rule (regular expression .*) which denotes where articles should go which don't match any other rule. If the folder doesn't exist yet, it will be created as soon as an article lands there. By default the mail will be send to all groups whose rules match. If you don't want that (you probably don't want), say

(setq nnmail-crosspost nil)

in ~/.gnus.el.

An example might be better than thousand words, so here's my nnmail-split-methods. Note that I send duplicates in a special group and that the default group is spam, since I filter all mails out which are from some list I'm subscribed to or which are addressed directly to me before. Those rules kill about 80% of the Spam which reaches me (Email addresses are changed to prevent spammers from using them):

(setq nnmail-split-methods
  '(("duplicates" "^Gnus-Warning:.*duplicate")
    ("XEmacs-NT" "^\\(To:\\|CC:\\).*localpart@xemacs.invalid.*")
    ("Gnus-Tut" "^\\(To:\\|CC:\\).*localpart@socha.invalid.*")
    ("tcsh" "^\\(To:\\|CC:\\).*localpart@mx.gw.invalid.*")
    ("BAfH" "^\\(To:\\|CC:\\).*localpart@.*uni-muenchen.invalid.*")
    ("Hamster-src" "^\\(CC:\\|To:\\).*hamster-sourcen@yahoogroups.\\(de\\|com\\).*")
    ("Tagesschau" "^From: tagesschau <localpart@www.tagesschau.invalid>$")
    ("Replies" "^\\(CC:\\|To:\\).*localpart@Frank-Schmitt.invalid.*")
    ("EK" "^From:.*\\(localpart@privateprovider.invalid\\|localpart@workplace.invalid\\).*")
    ("Spam" "^Content-Type:.*\\(ks_c_5601-1987\\|EUC-KR\\|big5\\|iso-2022-jp\\).*")
    ("Spam" "^Subject:.*\\(This really work\\|XINGA\\|ADV:\\|XXX\\|adult\\|sex\\).*")
    ("Spam" "^Subject:.*\\(\=\?ks_c_5601-1987\?\\|\=\?euc-kr\?\\|\=\?big5\?\\).*")
    ("Spam" "^X-Mailer:\\(.*BulkMailer.*\\|.*MIME::Lite.*\\|\\)")
    ("Spam" "^X-Mailer:\\(.*CyberCreek Avalanche\\|.*http\:\/\/GetResponse\.com\\)")
    ("Spam" "^From:.*\\(verizon\.net\\|prontomail\.com\\|money\\|ConsumerDirect\\).*")
    ("Spam" "^Delivered-To: GMX delivery to spamtrap@gmx.invalid$")
    ("Spam" "^Received: from link2buy.com")
    ("Spam" "^CC: .*azzrael@t-online.invalid")
    ("Spam" "^X-Mailer-Version: 1.50 BETA")
    ("Uni" "^\\(CC:\\|To:\\).*localpart@uni-koblenz.invalid.*")
    ("Inbox" "^\\(CC:\\|To:\\).*\\(my\ name\\|address@one.invalid\\|adress@two.invalid\\)")
    ("Spam" "")))

10.9.5 Composing messages

Question 5.1

What are the basic commands I need to know for sending mail and postings?

Answer

To start composing a new mail hit `m' either in Group or Summary buffer, for a posting, it's either `a' in Group buffer and filling the Newsgroups header manually or `a' in the Summary buffer of the group where the posting shall be send to. Replying by mail is `r' if you don't want to cite the author, or import the cited text manually and `R' to cite the text of the original message. For a follow up to a newsgroup, it's `f' and `F' (analogously to `r' and `R').

Enter new headers above the line saying "--text follows this line--", enter the text below the line. When ready hit `C-c C-c', to send the message, if you want to finish it later hit `C-c C-d' to save it in the drafts group, where you can start editing it again by saying `D e'.

Question 5.2

How to enable automatic word-wrap when composing messages?

Answer

Say

(add-hook 'message-mode-hook
	  (lambda ()
	    (setq fill-column 72)
	    (turn-on-auto-fill)))

in ~/.gnus.el. You can reformat a paragraph by hitting `M-q' (as usual)

Question 5.3

How to set stuff like From, Organization, Reply-To, signature...?

Answer

There are other ways, but you should use posting styles for this. (See below why). This example should make the syntax clear:

(setq gnus-posting-styles
  '((".*"
     (name "Frank Schmitt")
     (address "me@there.invalid")
     (organization "Hamme net, kren mer och nimmi")
     (signature-file "~/.signature")
     ("X-SampleHeader" "foobar")
     (eval (setq some-variable "Foo bar")))))

The ".*" means that this settings are the default ones (see below), valid values for the first element of the following lists are signature, signature-file, organization, address, name or body. The attribute name can also be a string. In that case, this will be used as a header name, and the value will be inserted in the headers of the article; if the value is `nil', the header name will be removed. You can also say (eval (foo bar)), then the function foo will be evaluated with argument bar and the result will be thrown away.

Question 5.4

Can I set things like From, Signature etc group based on the group I post too?

Answer

That's the strength of posting styles. Before, we used ".*" to set the default for all groups. You can use a regexp like "^gmane" and the following settings are only applied to postings you send to the gmane hierarchy, use ".*binaries" instead and they will be applied to postings send to groups containing the string binaries in their name etc.

You can instead of specifying a regexp specify a function which is evaluated, only if it returns true, the corresponding settings take effect. Two interesting candidates for this are message-news-p which returns t if the current Group is a newsgroup and the corresponding message-mail-p.

Note that all forms that match are applied, that means in the example below, when I post to gmane.mail.spam.spamassassin.general, the settings under ".*" are applied and the settings under message-news-p and those under "^gmane" and those under "^gmane\\.mail\\.spam\\.spamassassin\\.general$". Because of this put general settings at the top and specific ones at the bottom.

(setq gnus-posting-styles
      '((".*" ;;default
         (name "Frank Schmitt")
         (organization "Hamme net, kren mer och nimmi")
         (signature-file "~/.signature"))
        ((message-news-p) ;;Usenet news?
         (address "mySpamTrap@Frank-Schmitt.invalid")
         (reply-to "hereRealRepliesOnlyPlease@Frank-Schmitt.invalid"))
        ((message-mail-p) ;;mail?
         (address "usedForMails@Frank-Schmitt.invalid"))
        ("^gmane" ;;this is mail, too in fact
         (address "usedForMails@Frank-Schmitt.invalid")
         (reply-to nil))
        ("^gmane\\.mail\\.spam\\.spamassassin\\.general$"
         (eval (set (make-local-variable 'message-sendmail-envelope-from)
                    "Azzrael@rz-online.de")))))

Question 5.5

Is there a spell-checker? Perhaps even on-the-fly spell-checking?

Answer

You can use ispell.el to spell-check stuff in Emacs. So the first thing to do is to make sure that you've got either ispell or aspell installed and in your Path. Then you need ispell.el and for on-the-fly spell-checking flyspell.el. Ispell.el is shipped with Emacs and available through the XEmacs package system, flyspell.el is shipped with Emacs and part of XEmacs text-modes package which is available through the package system, so there should be no need to install them manually.

Ispell.el assumes you use ispell, if you choose aspell say

(setq ispell-program-name "aspell")

If you want your outgoing messages to be spell-checked, say

(add-hook 'message-send-hook 'ispell-message)

In your ~/.gnus.el, if you prefer on-the-fly spell-checking say

(add-hook 'message-mode-hook (lambda () (flyspell-mode 1)))

Question 5.6

Can I set the dictionary based on the group I'm posting to?

Answer

Yes, say something like

(add-hook 'gnus-select-group-hook
          (lambda ()
            (cond
             ((string-match
               "^de\\." (gnus-group-real-name gnus-newsgroup-name))
              (ispell-change-dictionary "deutsch8"))
             (t
              (ispell-change-dictionary "english")))))

Question 5.7

Is there some kind of address-book, so I needn't remember all those email addresses?

Answer

There's an very basic solution for this, mail aliases. You can store your mail addresses in a ~/.mailrc file using a simple alias syntax:

alias al	"Al <al@english-heritage.invalid>"

Then typing your alias (followed by a space or punctuation character) on a To: or Cc: line in the message buffer will cause Gnus to insert the full address for you. See the node "Mail Aliases" in Message (not Gnus) manual for details.

However, what you really want is the Insidious Big Brother Database bbdb. Get it through the XEmacs package system or from bbdb's homepage. Now place the following in ~/.gnus.el, to activate bbdb for Gnus:

(require 'bbdb)
(bbdb-initialize 'gnus 'message)

Now you probably want some general bbdb configuration, place them in ~/.emacs:

(require 'bbdb)
;;If you don't live in Northern America, you should disable the 
;;syntax check for telephone numbers by saying
(setq bbdb-north-american-phone-numbers-p nil)
;;Tell bbdb about your email address:
(setq bbdb-user-mail-names
      (regexp-opt '("Your.Email@here.invalid"
                    "Your.other@mail.there.invalid")))
;;cycling while completing email addresses
(setq bbdb-complete-name-allow-cycling t)
;;No popup-buffers
(setq bbdb-use-pop-up nil)

Now you should be ready to go. Say `M-x bbdb RET RET' to open a bbdb buffer showing all entries. Say `c' to create a new entry, `b' to search your BBDB and `C-o' to add a new field to an entry. If you want to add a sender to the BBDB you can also just hit `:' on the posting in the summary buffer and you are done. When you now compose a new mail, hit `TAB' to cycle through know recipients.

Question 5.8

Sometimes I see little images at the top of article buffer. What's that and how can I send one with my postings, too?

Answer

Those images are called X-Faces. They are 48*48 pixel b/w pictures, encoded in a header line. If you want to include one in your posts, you've got to convert some image to a X-Face. So fire up some image manipulation program (say Gimp), open the image you want to include, cut out the relevant part, reduce color depth to 1 bit, resize to 48*48 and save as bitmap. Now you should get the compface package from this site. and create the actual X-face by saying

cat file.xbm | xbm2ikon | compface > file.face
cat file.face | sed 's/\\/\\\\/g;s/\"/\\\"/g;' > file.face.quoted

If you can't use compface, there's an online X-face converter at http://www.dairiki.org/xface/. If you use MS Windows, you could also use the WinFace program from http://www.xs4all.nl/~walterln/winface/. Now you only have to tell Gnus to include the X-face in your postings by saying

(setq message-default-headers
        (with-temp-buffer
          (insert "X-Face: ")
          (insert-file-contents "~/.xface")
          (buffer-string)))

in ~/.gnus.el. If you use Gnus 5.10, you can simply add an entry

(x-face-file "~/.xface")

to gnus-posting-styles.

Question 5.9

Sometimes I accidentally hit r instead of f in newsgroups. Can Gnus warn me, when I'm replying by mail in newsgroups?

Answer

Put this in ~/.gnus.el:

(setq gnus-confirm-mail-reply-to-news t)

if you already use Gnus 5.10, if you still use 5.8.8 or 5.9 try this instead:

(eval-after-load "gnus-msg"
  '(unless (boundp 'gnus-confirm-mail-reply-to-news)
     (defadvice gnus-summary-reply (around reply-in-news activate)
       "Request confirmation when replying to news."
       (interactive)
       (when (or (not (gnus-news-group-p gnus-newsgroup-name))
                 (y-or-n-p "Really reply by mail to article author? "))
         ad-do-it))))

Question 5.10

How to tell Gnus not to generate a sender header?

Answer

Since 5.10 Gnus doesn't generate a sender header by default. For older Gnus' try this in ~/.gnus.el:

(eval-after-load "message"
      '(add-to-list 'message-syntax-checks '(sender . disabled)))

Question 5.11

I want Gnus to locally store copies of my send mail and news, how to do it?

Answer

You must set the variable gnus-message-archive-group to do this. You can set it to a string giving the name of the group where the copies shall go or like in the example below use a function which is evaluated and which returns the group to use.

(setq gnus-message-archive-group
	'((if (message-news-p)
	      "nnml:Send-News"
	    "nnml:Send-Mail")))

Question 5.12

People tell me my Message-IDs are not correct, why aren't they and how to fix it?

Answer

The message-ID is an unique identifier for messages you send. To make it unique, Gnus need to know which machine name to put after the "@". If the name of the machine where Gnus is running isn't suitable (it probably isn't at most private machines) you can tell Gnus what to use by saying:

(setq message-user-fqdn "yourmachine.yourdomain.tld")

in ~/.gnus.el. If you use Gnus 5.9 or ealier, you can use this instead (works for newer versions a well):

(eval-after-load "message"
  '(let ((fqdn "yourmachine.yourdomain.tld"));; <-- Edit this!
     (if (boundp 'message-user-fqdn)
         (setq message-user-fqdn fqdn)
       (gnus-message 1 "Redefining `message-make-fqdn'.")
       (defun message-make-fqdn ()
         "Return user's fully qualified domain name."
         fqdn))))

If you have no idea what to insert for "yourmachine.yourdomain.tld", you've got several choices. You can either ask your provider if he allows you to use something like yourUserName.userfqdn.provider.net, or you can use somethingUnique.yourdomain.tld if you own the domain yourdomain.tld, or you can register at a service which gives private users a FQDN for free, e.g. http://www.stura.tu-freiberg.de/~dlx/addfqdn.html. (Sorry but this website is in German, if you know of an English one offering the same, drop me a note).

Finally you can tell Gnus not to generate a Message-ID for News at all (and letting the server do the job) by saying

(setq message-required-news-headers
  (remove' Message-ID message-required-news-headers))

you can also tell Gnus not to generate Message-IDs for mail by saying

(setq message-required-mail-headers
  (remove' Message-ID message-required-mail-headers))

, however some mail servers don't generate proper Message-IDs, too, so test if your Mail Server behaves correctly by sending yourself a Mail and looking at the Message-ID.

10.9.6 Old messages

Question 6.1

How to import my old mail into Gnus?

Answer

The easiest way is to tell your old mail program to export the messages in mbox format. Most Unix mailers are able to do this, if you come from the MS Windows world, you may find tools at http://mbx2mbox.sourceforge.net/.

Now you've got to import this mbox file into Gnus. To do this, create a nndoc group based on the mbox file by saying `G f /path/file.mbox RET' in Group buffer. You now have read-only access to your mail. If you want to import the messages to your normal Gnus mail groups hierarchy, enter the nndoc group you've just created by saying `C-u RET' (thus making sure all messages are retrieved), mark all messages by saying `M P b' and either copy them to the desired group by saying `B c name.of.group RET' or send them through nnmail-split-methods (respool them) by saying `B r'.

Question 6.2

How to archive interesting messages?

Answer

If you stumble across an interesting message, say in gnu.emacs.gnus and want to archive it there are several solutions. The first and easiest is to save it to a file by saying `O f'. However, wouldn't it be much more convenient to have more direct access to the archived message from Gnus? If you say yes, put this snippet by Frank Haun <pille3003@fhaun.de> in ~/.gnus.el:

(defun my-archive-article (&optional n)
  "Copies one or more article(s) to a corresponding `nnml:' group, e.g.
`gnus.ding' goes to `nnml:1.gnus.ding'. And `nnml:List-gnus.ding' goes
to `nnml:1.List-gnus-ding'.

Use process marks or mark a region in the summary buffer to archive
more then one article."
  (interactive "P")
  (let ((archive-name
         (format
          "nnml:1.%s"
          (if (featurep 'xemacs)
              (replace-in-string gnus-newsgroup-name "^.*:" "")
            (replace-regexp-in-string "^.*:" "" gnus-newsgroup-name)))))
    (gnus-summary-copy-article n archive-name)))

You can now say `M-x my-archive-article' in summary buffer to archive the article under the cursor in a nnml group. (Change nnml to your preferred back end)

Of course you can also make sure the cache is enabled by saying

(setq gnus-use-cache t)

then you only have to set either the tick or the dormant mark for articles you want to keep, setting the read mark will remove them from cache.

Question 6.3

How to search for a specific message?

Answer

There are several ways for this, too. For a posting from a Usenet group the easiest solution is probably to ask groups.google.com, if you found the posting there, tell Google to display the raw message, look for the message-id, and say `M-^ the@message.id RET' in a summary buffer. Since Gnus 5.10 there's also a Gnus interface for groups.google.com which you can call with `G W') in group buffer.

Another idea which works for both mail and news groups is to enter the group where the message you are searching is and use the standard Emacs search `C-s', it's smart enough to look at articles in collapsed threads, too. If you want to search bodies, too try `M-s' instead. Further on there are the gnus-summary-limit-to-foo functions, which can help you, too.

Of course you can also use grep to search through your local mail, but this is both slow for big archives and inconvenient since you are not displaying the found mail in Gnus. Here comes nnir into action. Nnir is a front end to search engines like swish-e or swish++ and others. You index your mail with one of those search engines and with the help of nnir you can search trough the indexed mail and generate a temporary group with all messages which met your search criteria. If this sound cool to you get nnir.el from ftp://ls6-ftp.cs.uni-dortmund.de/pub/src/emacs/ or ftp://ftp.is.informatik.uni-duisburg.de/pub/src/emacs/. Instructions on how to use it are at the top of the file.

Question 6.4

How to get rid of old unwanted mail?

Answer

You can of course just mark the mail you don't need anymore by saying `#' with point over the mail and then say `B DEL' to get rid of them forever. You could also instead of actually deleting them, send them to a junk-group by saying `B m nnml:trash-bin' which you clear from time to time, but both are not the intended way in Gnus.

In Gnus, we let mail expire like news expires on a news server. That means you tell Gnus the message is expirable (you tell Gnus "I don't need this mail anymore") by saying `E' with point over the mail in summary buffer. Now when you leave the group, Gnus looks at all messages which you marked as expirable before and if they are old enough (default is older than a week) they are deleted.

Question 6.5

I want that all read messages are expired (at least in some groups). How to do it?

Answer

If you want all read messages to be expired (e.g. in mailing lists where there's an online archive), you've got two choices: auto-expire and total-expire. Auto-expire means, that every article which has no marks set and is selected for reading is marked as expirable, Gnus hits `E' for you every time you read a message. Total-expire follows a slightly different approach, here all article where the read mark is set are expirable.

To activate auto-expire, include auto-expire in the Group parameters for the group. (Hit `G c' in summary buffer with point over the group to change group parameters). For total-expire add total-expire to the group-parameters.

Which method you choose is merely a matter of taste: Auto-expire is faster, but it doesn't play together with Adaptive Scoring, so if you want to use this feature, you should use total-expire.

If you want a message to be excluded from expiration in a group where total or auto expire is active, set either tick (hit `u') or dormant mark (hit `u'), when you use auto-expire, you can also set the read mark (hit `d').

Question 6.6

I don't want expiration to delete my mails but to move them to another group.

Answer

Say something like this in ~/.gnus.el:

(setq nnmail-expiry-target "nnml:expired")

(If you want to change the value of nnmail-expiry-target on a per group basis see the question "How can I disable threading in some (e.g. mail-) groups, or set other variables specific for some groups?")

10.9.7 Gnus in a dial-up environment

Question 7.1

I don't have a permanent connection to the net, how can I minimize the time I've got to be connected?

Answer

You've got basically two options: Either you use the Gnus Agent (see below) for this, or you can install programs which fetch your news and mail to your local disk and Gnus reads the stuff from your local machine.

If you want to follow the second approach, you need a program which fetches news and offers them to Gnus, a program which does the same for mail and a program which receives the mail you write from Gnus and sends them when you're online.

Let's talk about Unix systems first: For the news part, the easiest solution is a small nntp server like Leafnode or sn, of course you can also install a full featured news server like inn. Then you want to fetch your Mail, popular choices are fetchmail and getmail. You should tell those to write the mail to your disk and Gnus to read it from there. Last but not least the mail sending part: This can be done with every MTA like sendmail, postfix, exim or qmail.

On windows boxes I'd vote for Hamster, it's a small freeware, open-source program which fetches your mail and news from remote servers and offers them to Gnus (or any other mail and/or news reader) via nntp respectively POP3 or IMAP. It also includes a smtp server for receiving mails from Gnus.

Question 7.2

So what was this thing about the Agent?

Answer

The Gnus agent is part of Gnus, it allows you to fetch mail and news and store them on disk for reading them later when you're offline. It kind of mimics offline newsreaders like e.g. Forte Agent. If you want to use the Agent place the following in ~/.gnus.el if you are still using 5.8.8 or 5.9 (it's the default since 5.10):

(setq gnus-agent t)

Now you've got to select the servers whose groups can be stored locally. To do this, open the server buffer (that is press `^' while in the group buffer). Now select a server by moving point to the line naming that server. Finally, agentize the server by typing `J a'. If you make a mistake, or change your mind, you can undo this action by typing `J r'. When you're done, type 'q' to return to the group buffer. Now the next time you enter a group on a agentized server, the headers will be stored on disk and read from there the next time you enter the group.

Question 7.3

I want to store article bodies on disk, too. How to do it?

Answer

You can tell the agent to automatically fetch the bodies of articles which fulfill certain predicates, this is done in a special buffer which can be reached by saying `J c' in group buffer. Please refer to the documentation for information which predicates are possible and how exactly to do it.

Further on you can tell the agent manually which articles to store on disk. There are two ways to do this: Number one: In the summary buffer, process mark a set of articles that shall be stored in the agent by saying `#' with point over the article and then type `J s'. The other possibility is to set, again in the summary buffer, downloadable (%) marks for the articles you want by typing `@' with point over the article and then typing `J u'. What's the difference? Well, process marks are erased as soon as you exit the summary buffer while downloadable marks are permanent. You can actually set downloadable marks in several groups then use fetch session ('J s' in the GROUP buffer) to fetch all of those articles. The only downside is that fetch session also fetches all of the headers for every selected group on an agentized server. Depending on the volume of headers, the initial fetch session could take hours.

Question 7.4

How to tell Gnus not to try to send mails / postings while I'm offline?

Answer

All you've got to do is to tell Gnus when you are online (plugged) and when you are offline (unplugged), the rest works automatically. You can toggle plugged/unplugged state by saying `J j' in group buffer. To start Gnus unplugged say `M-x gnus-unplugged' instead of `M-x gnus'. Note that for this to work, the agent must be active.

10.9.8 Getting help

Question 8.1

How to find information and help inside Emacs?

Answer

The first stop should be the Gnus manual (Say `C-h i d m Gnus RET' to start the Gnus manual, then walk through the menus or do a full-text search with `s'). Then there are the general Emacs help commands starting with C-h, type `C-h ? ?' to get a list of all available help commands and their meaning. Finally `M-x apropos-command' lets you search through all available functions and `M-x apropos' searches the bound variables.

Question 8.2

I can't find anything in the Gnus manual about X (e.g. attachments, PGP, MIME...), is it not documented?

Answer

There's not only the Gnus manual but also the manuals for message, emacs-mime, sieve and pgg. Those packages are distributed with Gnus and used by Gnus but aren't really part of core Gnus, so they are documented in different info files, you should have a look in those manuals, too.

Question 8.3

Which websites should I know?

Answer

The two most important ones are the official Gnus website. and it's sister site my.gnus.org (MGO), hosting an archive of lisp snippets, howtos, a (not really finished) tutorial and this FAQ.

Tell me about other sites which are interesting.

Question 8.4

Which mailing lists and newsgroups are there?

Answer

There's the newsgroup gnu.emacs.gnus (also available as gmane.emacs.gnus.user) which deals with general Gnus questions. The ding mailing list (ding@gnus.org) deals with development of Gnus. You can read the ding list via NNTP, too under the name gmane.emacs.gnus.general from news.gmane.org.

If you want to stay in the big8, news.software.newssreaders is also read by some Gnus users (but chances for qualified help are much better in the above groups) and if you speak German, there's de.comm.software.gnus.

Question 8.5

Where to report bugs?

Answer

Say `M-x gnus-bug', this will start a message to the gnus bug mailing list including information about your environment which make it easier to help you.

Question 8.6

I need real-time help, where to find it?

Answer

Point your IRC client to irc.my.gnus.org channel #mygnus. Don't be afraid if people there speak German, they are willing and capable of switching to English when people from outside Germany enter.

10.9.9 Tuning Gnus

Question 9.1

Starting Gnus is really slow, how to speed it up?

Answer

The reason for this could be the way Gnus reads it's active file, see the node "The Active File" in the Gnus manual for things you might try to speed the process up. An other idea would be to byte compile your ~/.gnus.el (say `M-x byte-compile-file RET ~/.gnus.el RET' to do it). Finally, if you have require statements in your .gnus, you could replace them with eval-after-load, which loads the stuff not at startup time, but when it's needed. Say you've got this in your ~/.gnus.el:

(require 'message)
(add-to-list 'message-syntax-checks '(sender . disabled))

then as soon as you start Gnus, message.el is loaded. If you replace it with

(eval-after-load "message"
      '(add-to-list 'message-syntax-checks '(sender . disabled)))

it's loaded when it's needed.

Question 9.2

How to speed up the process of entering a group?

Answer

A speed killer is setting the variable gnus-fetch-old-headers to anything different from nil, so don't do this if speed is an issue. To speed up building of summary say

(gnus-compile)

at the bottom of your ~/.gnus.el, this will make gnus byte-compile things like gnus-summary-line-format. then you could increase the value of gc-cons-threshold by saying something like

(setq gc-cons-threshold 3500000)

in ~/.emacs. If you don't care about width of CJK characters or use Gnus 5.10 or younger together with a recent GNU Emacs, you should say

(setq gnus-use-correct-string-widths nil)

Question 9.3

Sending mail becomes slower and slower, what's up?

Answer

The reason could be that you told Gnus to archive the messages you wrote by setting gnus-message-archive-group. Try to use a nnml group instead of an archive group, this should bring you back to normal speed.

10.9.10 Glossary

~/.gnus.el

When the term ~/.gnus.el is used it just means your Gnus configuration file. You might as well call it ~/.gnus or specify another name.

Back End

In Gnus terminology a back end is a virtual server, a layer between core Gnus and the real NNTP-, POP3-, IMAP- or whatever-server which offers Gnus a standardized interface to functions like "get message", "get Headers" etc.

Emacs

When the term Emacs is used in this FAQ, it means either GNU Emacs or XEmacs.

Message

In this FAQ message means a either a mail or a posting to a Usenet Newsgroup or to some other fancy back end, no matter of which kind it is.

MUA

MUA is an acronym for Mail User Agent, it's the program you use to read and write e-mails.

NUA

NUA is an acronym for News User Agent, it's the program you use to read and write Usenet news.

This document was generated by XEmacs shared group account on December, 19 2009 using texi2html 1.65.