docs/master

annotate ch03.roff @ 66:4c9f4d05d90e

Rework and extension of text from yesterday.
author markus schnalke <meillo@marmaro.de>
date Sat, 02 Jun 2012 10:32:26 +0200
parents abbaca05ee8e
children f80ced4e749b
rev   line source
meillo@58 1 .H0 "Discussion
meillo@0 2 .P
meillo@58 3 This main chapter discusses the practical work done in the mmh project.
meillo@58 4 It is structured along the goals to achieve. The concrete work done
meillo@58 5 is described in the examples of how the general goals were achieved.
meillo@58 6
meillo@58 7
meillo@58 8
meillo@58 9
meillo@58 10 .H1 "Stream-lining
meillo@58 11
meillo@0 12 .P
meillo@58 13 MH had been considered an all-in-one system for mail handling.
meillo@58 14 The community around nmh has a similar understanding.
meillo@58 15 In fundamental difference, I believe that mmh should be a MUA but
meillo@58 16 nothing more. I believe that all-in-one mail systems are not the way
meillo@58 17 to go. There are excellent specialized MTAs, like Postfix;
meillo@58 18 there are specialized MDAs, like Procmail; there are specialized
meillo@58 19 MRAs, like Fetchmail. I believe it's best to use them instead of
meillo@58 20 providing the same function ourselves. Doing something well requires to
meillo@58 21 focus on this particular aspect or a small set of aspects. The more
meillo@58 22 it is possible to focus, the better the result in this particular
meillo@58 23 area will be. The limiting resource in Free Software community development
meillo@58 24 usually is human power. If the low development power is even parted
meillo@58 25 into multiple development areas, it will hardly be possible to
meillo@58 26 compete with the specialists in the various fields. This is even
meillo@58 27 increased, given the small community \(en developers and users \(en
meillo@58 28 that MH-based mail systems have. In consequence, I believe that the
meillo@58 29 available resources should be concentrated at the point where MH is
meillo@58 30 most unique. This is clearly the MUA part.
meillo@58 31 .P
meillo@60 32 The goal for mmh was to remove peripheral parts and stream-line
meillo@60 33 it for the MUA task.
meillo@60 34
meillo@60 35
meillo@60 36 .H2 "Removal of Mail Transfer Facilities
meillo@60 37 .P
meillo@60 38 In contrast to nmh, which also provides mail submission and mail retrieval
meillo@60 39 facilities, mmh is a MUA only.
meillo@66 40 This general difference in the view on the character of nmh
meillo@66 41 strongly supported the development of mmh.
meillo@66 42 Removing the mail transfer facilities had been the first work task
meillo@66 43 for the mmh project.
meillo@60 44 .P
meillo@66 45 The MSA is called \fIMessage Transfer Service\fP (MTS) in nmh.
meillo@60 46 The facility establishes TCP/IP connections and speaks SMTP to submit
meillo@60 47 messages for relay to the outside world.
meillo@60 48 This part is implemented in the
meillo@60 49 .Pn post
meillo@60 50 command.
meillo@60 51 Demanded by the changes in
meillo@60 52 emailing, this part of nmh required changes in the last years.
meillo@60 53 Encrypted connections needed to be supported, hence SASL was introduced
meillo@60 54 into nmh. This added complexity to the nmh without improving it in
meillo@60 55 its core functions. Also, keeping up with recent developments in
meillo@60 56 this field needs requires development power and specialists.
meillo@60 57 Mmh cuts this whole facility off and depends on an external MTA instead.
meillo@60 58 The only outgoing interface available to mmh is the
meillo@60 59 .Pn sendmail
meillo@60 60 command.
meillo@60 61 Almost any MTA provides a
meillo@60 62 .Pn sendmail
meillo@60 63 command.
meillo@60 64 It not, any program can be substituted if it reads the
meillo@60 65 message from the standard input, extracts the recipient addresses
meillo@60 66 from the message header and does not conflict
meillo@60 67 with sendmail-specific command line arguments.
meillo@60 68 .P
meillo@60 69 To retrieve mail, the
meillo@60 70 .Pn inc
meillo@60 71 command in nmh has the ability to establish TCP/IP connections
meillo@60 72 and speaks POP3 to retrieve mail from remote servers.
meillo@60 73 As with mail submission, here encrypted connections are required
meillo@60 74 today, thus SASL support was added.
meillo@60 75 As POP3 is superseded by IMAP more and more, support for message
meillo@60 76 retrieval through IMAP will become necessary to be added soon.
meillo@60 77 Mmh has no support for retrieving mail from remote locations.
meillo@60 78 It depends on an external tool to cover this task.
meillo@60 79 There are two ways for messages to enter mmh's mail storage:
meillo@60 80 Incorporate them with
meillo@60 81 .Pn inc
meillo@60 82 from the system maildrop, or with
meillo@60 83 .Pn rcvstore
meillo@60 84 from the standard input.
meillo@66 85 In consequence, mmh has not any longer networking code
meillo@66 86 and thus does no more need to do transfer encryption and authentication.
meillo@66 87 Two large functional units are removed.
meillo@60 88 .P
meillo@60 89 With the removal of the MSA and MRA, mmh converted from an all-in-one
meillo@66 90 mail system to being only a MUA.
meillo@60 91 Following the Unix philosophy, it focuses on one job and to do that well.
meillo@60 92 Now, of course, mmh depends on third-party software.
meillo@60 93 An external MTA/MSA is required to transfer mail to the outside world;
meillo@60 94 an external MRA is required to retrieve mail from remote machines.
meillo@60 95 There exist excellent implementations of such software,
meillo@60 96 which do this specific task likely much better than the internal
meillo@60 97 versions of nmh do it. Also, this provides the choice for the best
meillo@60 98 suiting one of the available implementation.
meillo@60 99 .P
meillo@60 100 As it had already been possible to use an external MSA or MRA,
meillo@60 101 why not keep the internal version for convenience?
meillo@60 102 If this would question the sense in having a fall-back pager in all
meillo@60 103 the command line tools, in case
meillo@60 104 .Pn more
meillo@60 105 or
meillo@60 106 .Pn less
meillo@60 107 wouldn't be available, the answer is intuitively seen.
meillo@60 108 Now, an MSA or MRA is clearly more complex than a text pager, but
meillo@60 109 still the concept holds: If programs become complex, split them;
meillo@60 110 if projects become complex, split them.
meillo@60 111 Complexity is demanded by the problem to solve. Decades ago,
meillo@60 112 emailing had been small and simple.
meillo@60 113 (Remember,
meillo@60 114 .Pn /bin/mail
meillo@66 115 had once covered anything there was to email and still had been small.)
meillo@60 116 As the complexity in emailing increased, MH remainded mostly unchanged.
meillo@60 117 Nontheless, in nmh the POP server, which the original MH had included,
meillo@60 118 was removed. Now is the time to take one step further and remove
meillo@60 119 the MSA and MRA.
meillo@60 120 Not only does it decrease the code amount of the project,
meillo@60 121 but more important, it removes the whole field of message transfer
meillo@60 122 with all its implications from the project.
meillo@60 123 .P
meillo@66 124 If a project needs some kind of function, there's always the choice
meillo@66 125 between implementing the the function in the project directly or
meillo@66 126 depending on a library that provides the function or depending on
meillo@66 127 a program that provides the function.
meillo@66 128 Whereas adding the function directly to the project increases the
meillo@66 129 code size most, it makes the project most independent.
meillo@66 130 On the other end, interfacing external programs keeps the interface
meillo@66 131 smallest, but the depencency highest.
meillo@66 132 Using a library is in the middle.
meillo@66 133 Adding the function directly to the project is a bad choice for
meillo@66 134 any function of higher complexity, unless it's not available in other ways.
meillo@66 135 Hence, the dependencies only change in kind, not in their existence.
meillo@66 136 In mmh, library dependencies on
meillo@66 137 .Pn libsasl2
meillo@66 138 and
meillo@66 139 .Pn libcrypto /\c
meillo@66 140 .Pn libssl
meillo@66 141 were treated against program dependencies on an MSA and an MRA.
meillo@66 142 Besides program dependencies providing the stronger separation
meillo@66 143 and being more flexible, they also allowed
meillo@66 144 over 6\|000 lines of code to be removed from mmh.
meillo@66 145 This made mmh's code base about 12\|% smaller.
meillo@66 146 Reducing the projects code size by such an amount without actually
meillo@66 147 losing function is a convincing argument.
meillo@66 148 .P
meillo@66 149 Users of MH should have not problem to set up an external MSA and MRA.
meillo@60 150 Also, the popular MSAs and MRAs have large communities and a lot
meillo@60 151 of documentation available.
meillo@60 152 .P
meillo@60 153 Choices for MSAs range from the full-featured
meillo@60 154 .I Postfix
meillo@60 155 over mid-size solutions like
meillo@60 156 .I masqmail
meillo@60 157 and
meillo@60 158 .I dma
meillo@60 159 to small forwarders like
meillo@60 160 .I ssmtp
meillo@60 161 and
meillo@60 162 .I nullmailer .
meillo@60 163 Choices for MRAs include
meillo@60 164 .I fetchmail ,
meillo@60 165 .I getmail ,
meillo@60 166 .I mpop
meillo@60 167 and
meillo@60 168 .I fdm .
meillo@60 169
meillo@60 170
meillo@60 171 .H2 "Removal of non-MUA Tools
meillo@60 172 .P
meillo@62 173 Some of nmh's tools were removed from mmh because they didn't
meillo@58 174 match the main focus of adding to the MUA's task.
meillo@62 175 .BU
meillo@58 176 .Pn conflict
meillo@58 177 was removed because it is a mail system maintenance tool.
meillo@62 178 Besides, it even checks
meillo@58 179 .Fn /etc/passwd
meillo@58 180 and
meillo@58 181 .Fn /etc/group
meillo@62 182 for consistency, which has nothing at all to do with emailing.
meillo@58 183 The tool might be useful, but it should not be shipped with mmh.
meillo@62 184 .BU
meillo@58 185 .Pn rcvtty
meillo@58 186 was removed because its usecase of writing to the user's terminal
meillo@58 187 on receiving of mail is hardly wanted today. If users like to be
meillo@58 188 informed of new mail, then using the shell's
meillo@58 189 .Ev MAILPATH
meillo@62 190 variable or graphical notifications are likely more
meillo@62 191 appealing.
meillo@62 192 Writing directly to a terminals is hardly ever wanted today.
meillo@62 193 If though one wants to have it this way, the standard tool
meillo@58 194 .Pn write
meillo@58 195 can be used in a way similar to:
meillo@58 196 .DS
meillo@58 197 scan -file - | write `id -un`
meillo@58 198 .DE
meillo@62 199 .BU
meillo@58 200 .Pn viamail
meillo@62 201 was removed when the new attachment system was introduced, because
meillo@58 202 .Pn forw
meillo@62 203 could can now the task itself.
meillo@62 204 The program
meillo@58 205 .Pn sendfiles
meillo@62 206 was rewritten as a shell script wrapper around
meillo@58 207 .Pn forw .
meillo@62 208 .BU
meillo@58 209 .Pn msgchk
meillo@62 210 was removed, because it lost its usefulness when POP support was removed.
meillo@58 211 .Pn msgchk
meillo@62 212 provides hardly more information than:
meillo@58 213 .DS
meillo@58 214 ls -l /var/mail/meillo
meillo@58 215 .DE
meillo@62 216 It does separate between old and new mail, but that's merely a detail
meillo@62 217 and can be done with
meillo@58 218 .Pn stat (1)
meillo@62 219 too.
meillo@62 220 A very small shell script could be written to output the information
meillo@62 221 in a convenient way, if truly necessary.
meillo@58 222 As mmh's inc only incorporates mail from the user's local maildrop
meillo@62 223 and thus no data transfers over slow networks are involved,
meillo@62 224 there's hardly need to check for new mail before incorporating it.
meillo@62 225 .BU
meillo@58 226 .Pn msh
meillo@62 227 was removed because the tool was in conflict with the
meillo@58 228 philosophy of MH. It provided an interactive shell to access the
meillo@58 229 features of MH. One major feature of MH is being a tool chest.
meillo@58 230 .Pn msh
meillo@58 231 wouldn't be just another shell, tailored to the needs of mail
meillo@58 232 handling, but one large program to have the MH tools built in.
meillo@58 233 It's main use was for accessing Bulletin Boards, which have seized to
meillo@62 234 be popular.
meillo@62 235 .P
meillo@62 236 Removing
meillo@58 237 .Pn msh ,
meillo@62 238 together with the truly obsolete code relicts
meillo@58 239 .Pn vmh
meillo@58 240 and
meillo@58 241 .Pn wmh ,
meillo@62 242 saved more than 7\|000 lines of C code \(en
meillo@66 243 about 15\|% of the project's original source code amount.
meillo@63 244 Having the same functionality in less code (with equal readability,
meillo@63 245 of course) is an advantage.
meillo@63 246 Less code means less bugs and less maintenance work.
meillo@63 247 If
meillo@63 248 .Pn rcvtty
meillo@63 249 and
meillo@63 250 .Pn msgchk
meillo@63 251 are rarely used and can be implemented in different ways,
meillo@63 252 then why should one keep them?
meillo@63 253 .Pn viamail 's
meillo@63 254 use case is now partly obsolete and partly covered by
meillo@63 255 .Pn forw ,
meillo@63 256 hence there's no reason to still have
meillo@63 257 .Pn viamail
meillo@63 258 around.
meillo@63 259 .Pn conflict
meillo@63 260 is not related with the mail client, and
meillo@63 261 .Pn msh
meillo@63 262 conflicts with the basic concept of MH.
meillo@63 263 Both tools could still be useful, but not as part of mmh.
meillo@63 264 .P
meillo@63 265 It is a design goal of mmh to remove those parts that are rarely used.
meillo@63 266 The project shall become more stream-lined.
meillo@63 267 Rarely used and loosely related tools distract from the lean appearance.
meillo@63 268 They require maintenance cost without adding to the core task.
meillo@63 269 Therefore they were removed.
meillo@0 270
meillo@58 271
meillo@62 272 .H2 "Merge of \f(CWshow\fP and \f(CWmhshow\fP
meillo@58 273 .P
meillo@58 274 Since the very beginning, already in the first concept paper,
meillo@58 275 .Pn show
meillo@62 276 had been MH's message display program.
meillo@58 277 .Pn show
meillo@58 278 found out which pathnames the relevant messages had and invoked
meillo@58 279 .Pn mhl
meillo@62 280 then to have the content formated.
meillo@58 281 With the advent of MIME, this approach wasn't sufficient anymore.
meillo@58 282 MIME messages can consist of multiple parts, some of which aren't
meillo@62 283 directly displayable, and text content might be encoded in
meillo@58 284 foreign charsets.
meillo@58 285 .Pn show 's
meillo@58 286 simple approach and
meillo@58 287 .Pn mhl 's
meillo@58 288 limited display facilities couldn't cope with the task any longer.
meillo@62 289 .P
meillo@58 290 Instead of extending these tools, new ones were written from scratch
meillo@58 291 and then added to the MH tool chest. Doing so is encouraged by the
meillo@58 292 tool chest approach. The new tools could be added without interfering
meillo@62 293 with the existing ones. This is great. The ease of adding new tools
meillo@62 294 even made MH the first MUA to implement MIME.
meillo@58 295 .P
meillo@62 296 First, the new MIME features were added in form of the single program
meillo@58 297 .Pn mhn .
meillo@58 298 The command
meillo@62 299 .Cl "mhn \-show 42
meillo@58 300 would show the MIME message numbered 42.
meillo@58 301 With the 1.0 release of nmh in February 1999, Richard Coleman finished
meillo@58 302 the split of
meillo@58 303 .Pn mhn
meillo@58 304 into a set of specialized programs, which together covered the
meillo@62 305 multiple aspects of MIME. One of these resulting tools was
meillo@58 306 .Pn mhshow .
meillo@62 307 .Pn mhshow
meillo@62 308 resembled the
meillo@62 309 .Cl "mhn \-show
meillo@62 310 call.
meillo@62 311 .P
meillo@58 312
meillo@58 313
meillo@58 314 .H2 "Removal of Configure Options
meillo@58 315 .P
meillo@58 316
meillo@63 317
meillo@58 318 .H2 "Removal of switches
meillo@58 319 .P
meillo@58 320
meillo@58 321
meillo@58 322
meillo@58 323
meillo@58 324 .H1 "Moderizing
meillo@58 325
meillo@58 326
meillo@58 327 .H2 "Removal of Code Relicts
meillo@0 328 .P
meillo@51 329 The code base of mmh originates from the late Seventies,
meillo@51 330 had been extensively
meillo@51 331 worked on in the mid Eighties, and had been partly reorganized and extended
meillo@51 332 in the Nineties. Relicts of all those times had gathered in the code base.
meillo@12 333 My goal was to remove any ancient code parts. One part of the task was
meillo@12 334 converting obsolete code constructs to standard constructs, the other part
meillo@12 335 was dropping obsolete functions.
meillo@12 336 .P
meillo@12 337 As I'm not even thirty years old and have no more than seven years of
meillo@51 338 Unix experience, I needed to learn about the history in retrospective.
meillo@51 339 Older people likely have used those ancient constructs themselves
meillo@51 340 and have suffered from their incompatibilities and have longed for
meillo@12 341 standardization. Unfortunately, I have only read that others had done so.
meillo@12 342 This put me in a much more difficult positions when working on the old
meillo@12 343 code. I needed to recherche what other would have known by heart from
meillo@12 344 experience. All my programming experience comes from a time past ANSI C
meillo@12 345 and past POSIX. Although I knew about the times before, I took the
meillo@51 346 current state implicitly for granted most of the time.
meillo@12 347 .P
meillo@12 348 Being aware of
meillo@12 349 these facts, I rather let people with more historic experience solve the
meillo@12 350 task of converting the ancient code constructs to standardized ones.
meillo@12 351 Luckily, Lyndon Nerenberg focused on this task at the nmh project.
meillo@12 352 He converted large parts of the code to POSIX constructs, removing
meillo@12 353 the conditionals compilation for now standardized features.
meillo@12 354 I'm thankful for this task being solved. I only pulled the changes into
meillo@12 355 mmh.
meillo@12 356 .P
meillo@20 357 The other task \(en dropping ancient functionality to remove old code \(en
meillo@12 358 I did myself, though. My position to strip mmh to the bare minimum of
meillo@12 359 frequently used features is much more revolutional than the nmh community
meillo@20 360 likes it. Without the need to justify my decisions, I was able to quickly
meillo@20 361 remove functionality I considered ancient.
meillo@20 362 The need to discuss my decisions with
meillo@20 363 peers likely would have slowed this process down. Of course, I researched
meillo@12 364 if a particular feature really should be dropped. Having not had any
meillo@12 365 contact to this feature within my computer life was a first indicator to
meillo@12 366 drop it, but I also asked others and searched the literature for modern
meillo@12 367 usage of the feature. If it appeared to be truly ancient, I dropped it.
meillo@12 368 The reason for dropping is always part of the commit message in the
meillo@12 369 version control system. Thus, it is easy for others to check their
meillo@12 370 view on the topic with mine and possibly to argue for reinclusion.
meillo@12 371
meillo@12 372 .U2 "MMDF maildrop support
meillo@12 373 .P
meillo@12 374 I did drop any support for the MMDF maildrop format. This type of format
meillo@12 375 is conceptionally similar to the mbox format, but uses four bytes with
meillo@12 376 value 1 (\fL^A^A^A^A\fP) as message delimiter,
meillo@18 377 instead of the string ``\fLFrom\ \fP''.
meillo@12 378 Due to the similarity and mbox being the de-facto standard maildrop
meillo@12 379 format on Unix, but also due to the larger influence of Sendmail than MMDF,
meillo@12 380 the MMDF maildrop format had vanished.
meillo@12 381 .P
meillo@12 382 The simplifications within the code were only moderate. Switches could
meillo@12 383 be removed from tools like
meillo@12 384 .L packf ,
meillo@12 385 which generate packed mailboxes. Only one packed mailbox format remained:
meillo@12 386 mbox.
meillo@12 387 The most important changes affect the equally named mail parsing routine in
meillo@12 388 .L sbr/m_getfld.c .
meillo@12 389 The direct MMDF code had been removed, but as now only one packed mailbox
meillo@12 390 format is left, code structure simplifications are likely possible.
meillo@12 391 The reason why they are still outstanding is the heavily optimized code
meillo@18 392 of
meillo@18 393 .Fu m_getfld() .
meillo@18 394 Changes beyond a small local scope \(en
meillo@12 395 which restructuring in its core is \(en cause a high risk of damaging
meillo@12 396 the intricate workings of the optimized code. This problem is know
meillo@12 397 to the developers of nmh, too. They also avoid touching this minefield
meillo@12 398 if possible.
meillo@12 399
meillo@12 400 .U2 "UUCP Bang Paths
meillo@12 401 .P
meillo@12 402 More questionably than the former topic is the removal of support for the
meillo@12 403 UUCP bang path address style. However, the user may translate the bang
meillo@12 404 paths on retrieval to Internet addresses and the other way on posting
meillo@12 405 messages. The former can be done my an MDA like procmail; the latter
meillo@12 406 by a sendmail wrapper. This would ensure that any address handling would
meillo@12 407 work as expected. However, it might just work well without any
meillo@12 408 such modifications, as mmh does not touch addresses much, in general.
meillo@12 409 But I can't ensure as I have never used an environment with bang paths.
meillo@12 410 Also, the behavior might break at any point in further development.
meillo@12 411
meillo@12 412 .U2 "Hardcopy terminal support
meillo@12 413 .P
meillo@12 414 More of a funny anecdote is the remaining of a check for printing to a
meillo@12 415 hardcopy terminal until Spring 2012, when I finally removed it.
meillo@12 416 I surely would be very happy to see such a terminal in action, maybe
meillo@12 417 actually being able to work on it, but I fear my chances are null.
meillo@12 418 .P
meillo@12 419 The check only prevented a pager to be placed between the outputting
meillo@18 420 program (\c
meillo@18 421 .Pn mhl )
meillo@18 422 and the terminal. This could have been ensured with
meillo@18 423 the
meillo@18 424 .Sw \-nomoreproc
meillo@18 425 at the command line statically, too.
meillo@12 426
meillo@12 427 .U2 "Removed support for header fields
meillo@12 428 .P
meillo@12 429 The `Encrypted' header had been introduced by RFC\^822, but already
meillo@12 430 marked legacy in RFC 2822. It was superseded by FIXME.
meillo@12 431 Mmh does no more support this header.
meillo@12 432 .P
meillo@21 433 Native support for `Face' headers
meillo@21 434 had been removed, as well.
meillo@21 435 The feature is similar to the `X-Face' header in its intent,
meillo@21 436 but takes a different approach to store the image.
meillo@21 437 Instead of encoding the image data directly into the header,
meillo@21 438 the the header contains the hostname and UDP port where the image
meillo@21 439 date could be retrieved.
meillo@21 440 Neither `X-Face' nor the here described `Face' system
meillo@21 441 \**
meillo@21 442 .FS
meillo@21 443 There is also a newer but different system, invented 2005,
meillo@21 444 using `Face' headers.
meillo@21 445 It is the successor of `X-Face' providing colored PNG images.
meillo@21 446 .FE
meillo@21 447 became well used in the large scale.
meillo@21 448 It's still possible to use a Face systems,
meillo@21 449 although mmh does not provide support for any of the different systems
meillo@21 450 anymore. It's fairly easy to write a small shell script to
meillo@21 451 extract the embedded or fetch the external Face data and display the image.
meillo@21 452 Own Face headers can be added into the draft template files.
meillo@21 453 .P
meillo@12 454 `Content-MD5' headers were introduced by RFC\^1864. They provide only
meillo@12 455 a verification of data corruption during the transfer. By no means can
meillo@12 456 they ensure verbatim end-to-end delivery of the contents. This is clearly
meillo@12 457 stated in the RFC. The proper approach to provide verificationability
meillo@12 458 of content in an end-to-end relationship is the use of digital cryptography
meillo@12 459 (RFCs FIXME). On the other hand, transfer protocols should ensure the
meillo@12 460 integrity of the transmission. In combinations these two approaches
meillo@12 461 make the `Content-MD5' header field useless. In consequence, I removed
meillo@12 462 the support for it. By this removal, MD5 computation is not needed
meillo@12 463 anywhere in mmh. Hence, over 500 lines of code were removed by this one
meillo@12 464 change. Even if the `Content-MD5' header field is useful sometimes,
meillo@12 465 I value its usefulnes less than the improvement in maintainability, caused
meillo@12 466 by the removal.
meillo@12 467
meillo@20 468 .U2 "Prompter's Control Keys
meillo@20 469 .P
meillo@20 470 The program
meillo@20 471 .Pn prompter
meillo@20 472 queries the user to fill in a message form. When used by
meillo@20 473 .Pn comp
meillo@20 474 as:
meillo@20 475 .DS
meillo@20 476 comp \-editor prompter
meillo@20 477 .DE
meillo@20 478 the resulting behavior is similar to
meillo@20 479 .Pn mailx .
meillo@51 480 Apparently,
meillo@20 481 .Pn prompter
meillo@20 482 hadn't been touched lately. Otherwise it's hardly explainable why it
meillo@20 483 still offered the switches
meillo@20 484 .Sn \-erase \fUchr\fP
meillo@20 485 and
meillo@20 486 .Sn \-kill \fUchr\fP
meillo@20 487 to name the characters for command line editing.
meillo@21 488 The times when this had been necessary are long time gone.
meillo@20 489 Today these things work out-of-the-box, and if not, are configured
meillo@20 490 with the standard tool
meillo@20 491 .Pn stty .
meillo@20 492
meillo@21 493 .U2 "Vfork and Retry Loops
meillo@21 494 .P
meillo@51 495 MH creates many processes, which is a consequence of the tool chest approach.
meillo@21 496 In earlier times
meillo@21 497 .Fu fork()
meillo@21 498 had been an expensive system call, as the process's whole image needed
meillo@21 499 to be duplicated. One common case is replacing the image with
meillo@21 500 .Fu exec()
meillo@21 501 right after having forked the child process.
meillo@21 502 To speed up this case, the
meillo@21 503 .Fu vfork()
meillo@21 504 system call was invented at Berkeley. It completely omits copying the
meillo@21 505 image. If the image gets replaced right afterwards then unnecessary
meillo@21 506 work is omited. On old systems this results in large speed ups.
meillo@21 507 MH uses
meillo@21 508 .Fu vfork()
meillo@21 509 whenever possible.
meillo@21 510 .P
meillo@21 511 Memory management units that support copy-on-write semantics make
meillo@21 512 .Fu fork()
meillo@21 513 almost as fast as
meillo@21 514 .Fu vfork()
meillo@21 515 in the cases when they can be exchanged.
meillo@21 516 With
meillo@21 517 .Fu vfork()
meillo@51 518 being more error-prone and hardly faster, it's preferable to simply
meillo@21 519 use
meillo@21 520 .Fu fork()
meillo@21 521 instead.
meillo@21 522 .P
meillo@21 523 Related to the costs of
meillo@21 524 .Fu fork()
meillo@21 525 is the probability of its success.
meillo@21 526 Today on modern systems, the system call will succeed almost always.
meillo@51 527 In the Eighties on heavy loaded systems, as they were common at
meillo@21 528 universities, this had been different. Thus, many of the
meillo@21 529 .Fu fork()
meillo@21 530 calls were wrapped into loops to retry to fork several times in
meillo@21 531 short intervals, in case of previous failure.
meillo@21 532 In mmh, the program aborts at once if the fork failed.
meillo@21 533 The user can reexecute the command then. This is expected to be a
meillo@21 534 very rare case on modern systems, especially personal ones, which are
meillo@21 535 common today.
meillo@21 536
meillo@12 537
meillo@58 538 .H2 "Attachments
meillo@22 539 .P
meillo@58 540 MIME
meillo@58 541
meillo@58 542
meillo@58 543 .H2 "Digital Cryptography
meillo@22 544 .P
meillo@58 545 Signing and encryption.
meillo@58 546
meillo@58 547
meillo@58 548 .H2 "Good Defaults
meillo@22 549 .P
meillo@58 550 foo
meillo@58 551
meillo@58 552
meillo@58 553
meillo@58 554
meillo@58 555 .H1 "Code style
meillo@22 556 .P
meillo@58 557 foo
meillo@58 558
meillo@58 559
meillo@58 560 .H2 "Standard Code
meillo@22 561 .P
meillo@58 562 POSIX
meillo@22 563
meillo@22 564
meillo@58 565 .H2 "Separation
meillo@14 566
meillo@58 567 .U2 "MH Directory Split
meillo@0 568 .P
meillo@19 569 In MH and nmh, a personal setup had consisted of two parts:
meillo@19 570 The MH profile, named
meillo@19 571 .Fn \&.mh_profile
meillo@19 572 and being located directly in the user's home directory.
meillo@19 573 And the MH directory, where all his mail messages and also his personal
meillo@19 574 forms, scan formats, other configuration files are stored. The location
meillo@19 575 of this directory could be user-chosen. The default was to name it
meillo@19 576 .Fn Mail
meillo@19 577 and have it directly in the home directory.
meillo@19 578 .P
meillo@19 579 I've never liked the data storage and the configuration to be intermixed.
meillo@19 580 They are different kinds of data. One part, are the messages,
meillo@19 581 which are the data to operate on. The other part, are the personal
meillo@19 582 configuration files, which are able to change the behavior of the operations.
meillo@19 583 The actual operations are defined in the profile, however.
meillo@19 584 .P
meillo@19 585 When storing data, one should try to group data by its type.
meillo@19 586 There's sense in the Unix file system hierarchy, where configuration
meillo@19 587 file are stored separate (\c
meillo@19 588 .Fn /etc )
meillo@19 589 to the programs (\c
meillo@19 590 .Fn /bin
meillo@19 591 and
meillo@19 592 .Fn /usr/bin )
meillo@19 593 to their sources (\c
meillo@19 594 .Fn /usr/src ).
meillo@19 595 Such separation eases the backup management, for instance.
meillo@19 596 .P
meillo@19 597 In mmh, I've reorganized the file locations.
meillo@19 598 Still there are two places:
meillo@19 599 There's the mail storage directory, which, like in MH, contains all the
meillo@19 600 messages, but, unlike in MH, nothing else.
meillo@19 601 Its location still is user-chosen, with the default name
meillo@19 602 .Fn Mail ,
meillo@19 603 in the user's home directory. This is much similar to the case in nmh.
meillo@19 604 The configuration files, however, are grouped together in the new directory
meillo@19 605 .Fn \&.mmh
meillo@19 606 in the user's home directory.
meillo@19 607 The user's profile now is a file, named
meillo@19 608 .Fn profile ,
meillo@19 609 in this mmh directory.
meillo@19 610 Consistently, the context file and all the personal forms, scan formats,
meillo@19 611 and the like, are also there.
meillo@19 612 .P
meillo@19 613 The naming changed with the relocation.
meillo@19 614 The directory where everything, except the profile, had been stored (\c
meillo@19 615 .Fn $HOME/Mail ),
meillo@19 616 used to be called \fIMH directory\fP. Now, this directory is called the
meillo@19 617 user's \fImail storage\fP. The name \fImmh directory\fP is now given to
meillo@19 618 the new directory
meillo@19 619 (\c
meillo@19 620 .Fn $HOME/.mmh ),
meillo@19 621 containing all the personal configuration files.
meillo@19 622 .P
meillo@19 623 The separation of the files by type of content is logical and convenient.
meillo@19 624 There are no functional differences as any possible setup known to me
meillo@19 625 can be implemented with both approaches, although likely a bit easier
meillo@19 626 with the new approach. The main goal of the change had been to provide
meillo@19 627 sensible storage locations for any type of personal mmh file.
meillo@19 628 .P
meillo@19 629 In order for one user to have multiple MH setups, he can use the
meillo@19 630 environment variable
meillo@19 631 .Ev MH
meillo@19 632 the point to a different profile file.
meillo@19 633 The MH directory (mail storage plus personal configuration files) is
meillo@19 634 defined by the
meillo@19 635 .Pe Path
meillo@19 636 profile entry.
meillo@19 637 The context file could be defined by the
meillo@19 638 .Pe context
meillo@19 639 profile entry or by the
meillo@19 640 .Ev MHCONTEXT
meillo@19 641 environment variable.
meillo@19 642 The latter is useful to have a distinct context (e.g. current folders)
meillo@19 643 in each terminal window, for instance.
meillo@19 644 In mmh, there are three environment variables now.
meillo@19 645 .Ev MMH
meillo@19 646 may be used to change the location of the mmh directory.
meillo@19 647 .Ev MMHP
meillo@19 648 and
meillo@19 649 .Ev MMHC
meillo@19 650 change the profile and context files, respectively.
meillo@19 651 Besides providing a more consistent feel (which simply is the result
meillo@19 652 of being designed anew), the set of personal configuration files can
meillo@19 653 be chosen independently from the profile (including mail storage location)
meillo@19 654 and context, now. Being it relevant for practical use or not, it
meillo@19 655 de-facto is an improvement. However, the main achievement is the
meillo@19 656 split between mail storage and personal configuration files.
meillo@17 657
meillo@0 658
meillo@58 659 .H2 "Modularization
meillo@0 660 .P
meillo@58 661 whatnowproc
meillo@0 662 .P
meillo@49 663 The \fIMH library\fP
meillo@49 664 .Fn libmh.a
meillo@49 665 collects a bunch of standard functions that many of the MH tools need,
meillo@49 666 like reading the profile or context files.
meillo@49 667 This doesn't hurt the separation.
meillo@49 668
meillo@58 669
meillo@58 670 .H2 "Style
meillo@58 671 .P
meillo@58 672 Code layout, goto, ...
meillo@58 673
meillo@58 674
meillo@58 675
meillo@58 676
meillo@58 677 .H1 "Concept Exploitation/Homogeniety
meillo@58 678
meillo@58 679
meillo@58 680 .H2 "Draft Folder
meillo@58 681 .P
meillo@58 682 Historically, MH provided exactly one draft message, named
meillo@58 683 .Fn draft
meillo@58 684 and
meillo@58 685 being located in the MH directory. When starting to compose another message
meillo@58 686 before the former one was sent, the user had been questioned whether to use,
meillo@58 687 refile or replace the old draft. Working on multiple drafts at the same time
meillo@58 688 was impossible. One could only work on them in alteration by refiling the
meillo@58 689 previous one to some directory and fetching some other one for reediting.
meillo@58 690 This manual draft management needed to be done each time the user wanted
meillo@58 691 to switch between editing one draft to editing another.
meillo@58 692 .P
meillo@58 693 To allow true parallel editing of drafts, in a straight forward way, the
meillo@58 694 draft folder facility exists. It had been introduced already in July 1984
meillo@58 695 by Marshall T. Rose. The facility was deactivated by default.
meillo@58 696 Even in nmh, the draft folder facility remained deactivated by default.
meillo@58 697 At least, Richard Coleman added the man page
meillo@58 698 .Mp mh-draft(5)
meillo@58 699 to document
meillo@58 700 the feature well.
meillo@58 701 .P
meillo@58 702 The only advantage of not using the draft folder facility is the static
meillo@58 703 name of the draft file. This could be an issue for MH frontends like mh-e.
meillo@58 704 But as they likely want to provide working on multiple drafts in parallel,
meillo@58 705 the issue is only concerning compatibility. The aim of nmh to stay compatible
meillo@58 706 prevented the default activation of the draft folder facility.
meillo@58 707 .P
meillo@58 708 On the other hand, a draft folder is the much more natural concept than
meillo@58 709 a draft message. MH's mail storage consists of folders and messages,
meillo@58 710 the messages named with ascending numbers. A draft message breaks with this
meillo@58 711 concept by introducing a message in a file named
meillo@58 712 .Fn draft .
meillo@58 713 This draft
meillo@58 714 message is special. It can not be simply listed with the available tools,
meillo@58 715 but instead requires special switches. I.e. corner-cases were
meillo@58 716 introduced. A draft folder, in contrast, does not introduce such
meillo@58 717 corner-cases. The available tools can operate on the messages within that
meillo@58 718 folder like on any messages within any mail folders. The only difference
meillo@58 719 is the fact that the default folder for
meillo@58 720 .Pn send
meillo@58 721 is the draft folder,
meillo@58 722 instead of the current folder, like for all other tools.
meillo@58 723 .P
meillo@58 724 The trivial part of the change was activating the draft folder facility
meillo@58 725 by default and setting a default name for this folder. Obviously, I chose
meillo@58 726 the name
meillo@58 727 .Fn +drafts .
meillo@58 728 This made the
meillo@58 729 .Sw \-draftfolder
meillo@58 730 and
meillo@58 731 .Sw \-draftmessage
meillo@58 732 switches useless, and I could remove them.
meillo@58 733 The more difficult but also the part that showed the real improvement,
meillo@58 734 was updating the tools to the new concept.
meillo@58 735 .Sw \-draft
meillo@58 736 switches could
meillo@58 737 be dropped, as operating on a draft message became indistinguishable to
meillo@58 738 operating on any other message for the tools.
meillo@58 739 .Pn comp
meillo@58 740 still has its
meillo@58 741 .Sw \-use
meillo@58 742 switch for switching between its two modes: (1) Compose a new
meillo@58 743 draft, possibly by taking some existing message as a form. (2) Modify
meillo@58 744 an existing draft. In either case, the behavior of
meillo@58 745 .Pn comp is
meillo@58 746 deterministic. There is no more need to query the user. I consider this
meillo@58 747 a major improvement. By making
meillo@58 748 .Pn send
meillo@58 749 simply operate on the current
meillo@58 750 message in the draft folder by default, with message and folder both
meillo@58 751 overridable by specifying them on the command line, it is now possible
meillo@58 752 to send a draft anywhere within the storage by simply specifying its folder
meillo@58 753 and name.
meillo@58 754 .P
meillo@58 755 All theses changes converted special cases to regular cases, thus
meillo@58 756 simplifying the tools and increasing the flexibility.
meillo@58 757
meillo@58 758
meillo@58 759 .H2 "Trash Folder
meillo@58 760 .P
meillo@58 761 Similar to the situation for drafts is the situation for removed messages.
meillo@58 762 Historically, a message was deleted by renaming. A specific
meillo@58 763 \fIbackup prefix\fP, often comma (\c
meillo@58 764 .Fn , )
meillo@58 765 or hash (\c
meillo@58 766 .Fn # ),
meillo@58 767 being prepended to the file name. Thus, MH wouldn't recognize the file
meillo@58 768 as a message anymore, as only files whose name consists of digits only
meillo@58 769 are treated as messages. The removed messages remained as files in the
meillo@58 770 same directory and needed some maintenance job to truly delete them after
meillo@58 771 some grace time. Usually, by running a command similar to
meillo@58 772 .DS
meillo@58 773 find /home/user/Mail \-ctime +7 \-name ',*' | xargs rm
meillo@58 774 .DE
meillo@58 775 in a cron job. Within the grace time interval
meillo@58 776 the original message could be restored by stripping the
meillo@58 777 the backup prefix from the file name. If however, the last message of
meillo@58 778 a folder is been removed \(en say message
meillo@58 779 .Fn 6
meillo@58 780 becomes file
meillo@58 781 .Fn ,6
meillo@58 782 \(en and a new message enters the same folder, thus the same
meillo@58 783 numbered being given again \(en in our case
meillo@58 784 .Fn 6
meillo@58 785 \(en, if that one
meillo@58 786 is removed too, then the backup of the former message gets overwritten.
meillo@58 787 Thus, the ability to restore removed messages does not only depend on
meillo@58 788 the ``sweeping cron job'' but also on the removing of further messages.
meillo@58 789 This is undesirable, because the real mechanism is hidden from the user
meillo@58 790 and the consequences of further removals are not always obvious.
meillo@58 791 Further more, the backup files are scattered within the whole mail
meillo@58 792 storage, instead of being collected at one place.
meillo@58 793 .P
meillo@58 794 To improve the situation, the profile entry
meillo@58 795 .Pe rmmproc
meillo@58 796 (previously named
meillo@58 797 .Pe Delete-Prog )
meillo@58 798 was introduced, very early.
meillo@58 799 It could be set to any command, which would care for the mail removal
meillo@58 800 instead of taking the default action, described above.
meillo@58 801 Refiling the to-be-removed files to some garbage folder was a common
meillo@58 802 example. Nmh's man page
meillo@58 803 .Mp rmm(1)
meillo@58 804 proposes
meillo@58 805 .Cl "refile +d
meillo@58 806 to move messages to the garbage folder and
meillo@58 807 .Cl "rm `mhpath +d all`
meillo@58 808 the empty the garbage folder.
meillo@58 809 Managing the message removal this way is a sane approach. It keeps
meillo@58 810 the removed messages in one place, makes it easy to remove the backup
meillo@58 811 files, and, most important, enables the user to use the tools of MH
meillo@58 812 itself to operate on the removed messages. One can
meillo@58 813 .Pn scan
meillo@58 814 them,
meillo@58 815 .Pn show
meillo@58 816 them, and restore them with
meillo@58 817 .Pn refile .
meillo@58 818 There's no more
meillo@58 819 need to use
meillo@58 820 .Pn mhpath
meillo@58 821 to switch over from MH tools to Unix tools \(en MH can do it all itself.
meillo@58 822 .P
meillo@58 823 This approach matches perfect with the concepts of MH, thus making
meillo@58 824 it powerful. Hence, I made it the default. And even more, I also
meillo@58 825 removed the old backup prefix approach, as it is clearly less powerful.
meillo@58 826 Keeping unused alternative in the code is a bad choice as they likely
meillo@58 827 gather bugs, by not being constantly tested. Also, the increased code
meillo@58 828 size and more conditions crease the maintenance costs. By strictly
meillo@58 829 converting to the trash folder approach, I simplified the code base.
meillo@58 830 .Pn rmm
meillo@58 831 calls
meillo@58 832 .Pn refile
meillo@58 833 internally to move the to-be-removed
meillo@58 834 message to the trash folder (\c
meillo@58 835 .Fn +trash
meillo@58 836 by default). Messages
meillo@58 837 there can be operated on like on any other message in the storage.
meillo@58 838 The sweep clean, one can use
meillo@58 839 .Cl "rmm \-unlink +trash a" ,
meillo@58 840 where the
meillo@58 841 .Sw \-unlink
meillo@58 842 switch causes the files to be truly unliked instead
meillo@58 843 of moved to the trash folder.
meillo@58 844
meillo@58 845
meillo@58 846 .H2 "Path Notations
meillo@58 847 .P
meillo@58 848 foo
meillo@58 849
meillo@58 850
meillo@58 851 .H2 "MIME Integration
meillo@58 852 .P
meillo@58 853 user-visible access to whole messages and MIME parts are inherently
meillo@58 854 different
meillo@58 855
meillo@58 856
meillo@58 857 .H2 "Of One Cast
meillo@58 858 .P