docs/master: discussion.roff annotate

docs/master

annotate discussion.roff @ 224:b0e83cfde2bf

Further minor improvements.

author	markus schnalke <meillo@marmaro.de>
date	Sun, 15 Jul 2012 19:46:35 +0200
parents	1fa5a74bf138
children	d83ab437e3a6

rev	line source
meillo@58	1 .H0 "Discussion
meillo@0	2 .P
meillo@169	3 This main chapter discusses the practical work accomplished in the
meillo@169	4 mmh project.
meillo@217	5 It is structured along the goals chosen for the project.
meillo@217	6 A selection of the work undertaken
meillo@217	7 is described.
meillo@217	8 .P
meillo@217	9 This discussion compares the present version of mmh with the state of
meillo@217	10 nmh at the time when the mmh project had started, i.e. fall 2011.
meillo@224	11 Recent changes in nmh are rarely part of the discussion.
meillo@187	12 .P
meillo@187	13 For the reader's convenience, the structure of modern email systems
meillo@217	14 is depicted in the following figure.
meillo@187	15 It illustrates the path a message takes from sender to recipient.
meillo@217	16
meillo@217	17 .sp 1.5
meillo@187	18 .KS
meillo@187	19 .in 2c
meillo@187	20 .so input/mail-agents.pic
meillo@187	21 .KE
meillo@217	22 .sp 1.5
meillo@217	23
meillo@187	24 .LP
meillo@217	25 The ellipses denote mail agents, i.e. different jobs in email processing.
meillo@217	26 These are:
meillo@187	27 .IP "Mail User Agent (MUA)
meillo@217	28 The only program users directly interact with.
meillo@187	29 It includes functions to compose new mail, display received mail,
meillo@187	30 and to manage the mail storage.
meillo@217	31 It is called a \fImail client\fP as well.
meillo@187	32 .IP "Mail Submission Agent (MSA)
meillo@187	33 A special kind of Mail Transfer Agent, used to submit mail into the
meillo@187	34 mail transport system.
meillo@217	35 Often it is also called an MTA.
meillo@187	36 .IP "Mail Transfer Agent (MTA)
meillo@187	37 A node in the mail transport system.
meillo@217	38 It transfers incoming mail to a transport node nearer to the
meillo@217	39 final destination.
meillo@217	40 An MTA may be the final destination itself.
meillo@187	41 .IP "Mail Delivery Agent (MDA)
meillo@217	42 Delivers mail according to a set of rules.
meillo@217	43 Usually, the messages are stored to disk.
meillo@187	44 .IP "Mail Retrieval Agent (MRA)
meillo@217	45 Initiates the transfer of mail from a remote location to the local machine.
meillo@217	46 (The dashed arrow in the figure represents the pull request.)
meillo@217	47 .LP
meillo@217	48 The dashed boxes represent entities that usually reside on single machines.
meillo@217	49 The box on the lower left represents the sender's system.
meillo@187	50 The box on the upper left represents the first mail transfer node.
meillo@224	51 The box on the upper right represents the transfer node responsible
meillo@224	52 for the destination address.
meillo@217	53 The box on the lower right represents the recipient's system.
meillo@187	54 Often, the boxes above the dotted line are servers on the Internet.
meillo@217	55 Many mail clients, including nmh, include all of the components below
meillo@217	56 the dotted line.
meillo@217	57 This is not the case for mmh; it implements the MUA only.
meillo@187	58
meillo@187	59
meillo@187	60
meillo@58	61
meillo@58	62
meillo@58	63
meillo@133	64 .\" --------------------------------------------------------------
meillo@125	65 .H1 "Streamlining
meillo@58	66
meillo@0	67 .P
meillo@217	68 MH once provided a complete email system.
meillo@217	69 The community around nmh tries to keep nmh in similar shape.
meillo@178	70 In fundamental contrast, mmh shall be an MUA only.
meillo@87	71 I believe that the development of all-in-one mail systems is obsolete.
meillo@173	72 Today, email is too complex to be fully covered by a single project.
meillo@173	73 Such a project will not be able to excel in all aspects.
meillo@159	74 Instead, the aspects of email should be covered by multiple projects,
meillo@87	75 which then can be combined to form a complete system.
meillo@169	76 Excellent implementations for the various aspects of email already exist.
meillo@223	77 Just to name three examples: Postfix is a specialized MTA, Procmail
meillo@223	78 is a specialized MDA, and Fetchmail is a specialized MRA.
meillo@89	79 I believe that it is best to use such specialized tools instead of
meillo@217	80 providing the same function once more as a side component.
meillo@58	81 .P
meillo@169	82 Doing something well requires focusing on a small set of specific aspects.
meillo@217	83 Under the assumption that development which is focussed on a particular
meillo@217	84 area produces better results there, specialized projects will be superior
meillo@87	85 in their field of focus.
meillo@87	86 Hence, all-in-one mail system projects \(en no matter if monolithic
meillo@87	87 or modular \(en will never be the best choice in any of the fields.
meillo@217	88 Even in providing the most consistent all-in-one system, they are likely
meillo@217	89 to be beaten by projects that focus exclusively on the creation
meillo@217	90 of a homogeneous system by integrating existing mail components.
meillo@87	91 .P
meillo@217	92 Usually, the limiting resource in the community development of
meillo@217	93 free software is man power.
meillo@171	94 .\" XXX FIXME ref!
meillo@217	95 If the development effort is spread over a large development area,
meillo@217	96 it becomes more difficult to compete with the specialists in the
meillo@87	97 various fields.
meillo@87	98 The concrete situation for MH-based mail systems is even tougher,
meillo@169	99 given their small and aged community, concerning both developers and users.
meillo@87	100 .P
meillo@87	101 In consequence, I believe that the available development resources
meillo@100	102 should focus on the point where MH is most unique.
meillo@87	103 This is clearly the user interface \(en the MUA.
meillo@125	104 Peripheral parts should be removed to streamline mmh for the MUA task.
meillo@60	105
meillo@60	106
meillo@100	107 .H2 "Mail Transfer Facilities
meillo@154	108 .Id mail-transfer-facilities
meillo@60	109 .P
meillo@217	110 The removal of the mail transfer facilities, effectively dropping the
meillo@217	111 MSA and MRA, had been the first work task in the mmh project.
meillo@217	112 The desire for this change initiated the creation of the mmh project.
meillo@60	113 .P
meillo@169	114 Focusing on one mail agent role only, is motivated by Eric Allman's
meillo@105	115 experience with Sendmail.
meillo@217	116 He identified the limitation of Sendmail
meillo@217	117 to the MTA task as one reason for its success:
meillo@105	118 .[ [
meillo@105	119 costales sendmail
meillo@105	120 .], p. xviii]
meillo@105	121 .QS
meillo@105	122 Second, I limited myself to the routing function \(en
meillo@110	123 I wouldn't write user agents or delivery back-ends.
meillo@217	124 This was a departure of the dominant thought of the time,
meillo@105	125 in which routing logic, local delivery, and often the network code
meillo@105	126 were incorporated directly into the user agents.
meillo@105	127 .QE
meillo@105	128 .P
meillo@187	129 In nmh, the MSA is called \fIMessage Transfer Service\fP (MTS).
meillo@105	130 This facility, implemented by the
meillo@105	131 .Pn post
meillo@217	132 command, establishes network connections and spoke SMTP to submit
meillo@159	133 messages to be relayed to the outside world.
meillo@217	134 When email transfer changed, this part needed to be changed as well.
meillo@89	135 Encryption and authentication for network connections
meillo@87	136 needed to be supported, hence TLS and SASL were introduced into nmh.
meillo@217	137 This added complexity without improving the core functions.
meillo@217	138 Furthermore, keeping up with recent developments in the field of
meillo@87	139 mail transfer requires development power and specialists.
meillo@217	140 In mmh, this whole facility was simply cut off
meillo@76	141 .Ci f6aa95b724fd8c791164abe7ee5468bf5c34f226
meillo@76	142 .Ci fecd5d34f65597a4dfa16aeabea7d74b191532c3
meillo@217	143 .Ci 156d35f6425bea4c1ed3c4c79783dc613379c65b .
meillo@87	144 Instead, mmh depends on an external MSA.
meillo@217	145 All outgoing mail in mmh goes through the
meillo@60	146 .Pn sendmail
meillo@87	147 command, which almost any MSA provides.
meillo@87	148 If not, a wrapper program can be written.
meillo@87	149 It must read the message from the standard input, extract the
meillo@87	150 recipient addresses from the message header, and hand the message
meillo@87	151 over to the MSA.
meillo@223	152 For example, a wrapper script for qmail would be:
meillo@87	153 .VS
meillo@87	154 #!/bin/sh
meillo@138	155 exec qmail-inject # ignore command line arguments
meillo@87	156 VE
meillo@87	157 The requirement to parse the recipient addresses out of the message header
meillo@217	158 may be removed in the future.
meillo@217	159 Mmh could pass the recipient addresses as command line arguments.
meillo@100	160 This appears to be the better interface.
meillo@60	161 .P
meillo@60	162 To retrieve mail, the
meillo@60	163 .Pn inc
meillo@217	164 command in nmh acts as MRA.
meillo@217	165 It establishes network connections
meillo@217	166 and speaks POP3 to retrieve mail from remote servers.
meillo@76	167 As with mail submission, the network connections required encryption and
meillo@217	168 authentication, thus TLS and SASL were added to nmh.
meillo@169	169 Support for message retrieval through IMAP will soon become necessary
meillo@169	170 additions, too, and likewise for any other changes in mail transfer.
meillo@217	171 But not in mmh because it has dropped the support for retrieving mail
meillo@217	172 from remote locations
meillo@217	173 .Ci ab7b48411962d26439f92f35ed084d3d6275459c .
meillo@76	174 Instead, it depends on an external tool to cover this task.
meillo@169	175 Mmh has two paths for messages to enter mmh's mail storage:
meillo@100	176 (1) Mail can be incorporated with
meillo@60	177 .Pn inc
meillo@87	178 from the system maildrop, or (2) with
meillo@60	179 .Pn rcvstore
meillo@87	180 by reading them, one at a time, from the standard input.
meillo@60	181 .P
meillo@217	182 With the removal of the MSA and MRA, mmh converted from a complete
meillo@217	183 mail system to only an MUA.
meillo@60	184 Now, of course, mmh depends on third-party software.
meillo@87	185 An external MSA is required to transfer mail to the outside world;
meillo@60	186 an external MRA is required to retrieve mail from remote machines.
meillo@217	187 Excellent implementations of such software exist.
meillo@217	188 They likely are superior to the internal versions that were removed.
meillo@217	189 Additionally, the best suiting programs can be chosen freely.
meillo@60	190 .P
meillo@217	191 As it had already been possible to use an external MSA and MRA,
meillo@217	192 why should the internal version not be kept for convenience?
meillo@217	193 .\" XXX commas correct?
meillo@224	194 Transfered to a different area, the question,
meillo@224	195 whether there is sense in having a fall-back pager in all
meillo@224	196 the command line tools for the cases when
meillo@60	197 .Pn more
meillo@60	198 or
meillo@60	199 .Pn less
meillo@173	200 are not available, appears to be ridiculous.
meillo@100	201 Of course, MSAs and MRAs are more complex than text pagers
meillo@87	202 and not necessarily available but still the concept of orthogonal
meillo@217	203 design holds: ``Write programs that do one thing and do it well''.
meillo@87	204 .[
meillo@87	205 mcilroy unix phil
meillo@87	206 p. 53
meillo@87	207 .]
meillo@87	208 .[
meillo@87	209 mcilroy bstj foreword
meillo@87	210 .]
meillo@87	211 Here, this part of the Unix philosophy was applied not only
meillo@87	212 to the programs but to the project itself.
meillo@87	213 In other words:
meillo@164	214 Develop projects that focus on one thing and do it well.
meillo@169	215 Projects which have grown complex should be split, for the same
meillo@169	216 reasons that programs which have grown complex should be split.
meillo@100	217 If it is conceptionally more elegant to have the MSA and MRA as
meillo@87	218 separate projects then they should be separated.
meillo@217	219 In my opinion, this is the case.
meillo@217	220 The RFCs suggest this separation by clearly distinguishing the
meillo@217	221 different mail handling tasks [RFC\\|821].
meillo@217	222 The small interfaces between the mail agents support the
meillo@217	223 separation as well.
meillo@76	224 .P
meillo@217	225 Once, email had been small and simple.
meillo@100	226 At that time,
meillo@60	227 .Pn /bin/mail
meillo@169	228 had covered everything there was to email and still was small and simple.
meillo@100	229 Later, the essential complexity of email increased.
meillo@224	230 (Essential complexity is the complexity defined by the problem itself
meillo@217	231 .[ [
meillo@87	232 brooks no silver bullet
meillo@224	233 .]].)
meillo@217	234 Consequently, email systems grew.
meillo@100	235 RFCs started to introduce the concept of mail agents to separate the
meillo@217	236 various roles because they became more extensive and because
meillo@217	237 new roles appeared.
meillo@217	238 As mail system implementations grew, parts of them were split off.
meillo@169	239 For instance, a POP server was included in the original MH;
meillo@169	240 it was removed in nmh.
meillo@217	241 Now is the time to go one step further and split off the MSA and MRA,
meillo@217	242 as well.
meillo@87	243 Not only does this decrease the code size of the project,
meillo@169	244 more importantly, it unburdens mmh of the whole field of
meillo@217	245 message transfer, with all its implications for the project.
meillo@169	246 There is no more need for concern with changes in network transfer.
meillo@217	247 This independence is gained by depending on external components
meillo@217	248 that cover the field.
meillo@60	249 .P
meillo@217	250 In general, functionality can be added in three different ways:
meillo@171	251 .LI 1
meillo@217	252 By implementing the function in the project itself.
meillo@171	253 .LI 2
meillo@217	254 By depending on a library that provides the function.
meillo@171	255 .LI 3
meillo@217	256 By depending on a program that provides the function.
meillo@171	257 .LP
meillo@159	258 .\" XXX Rework sentence
meillo@169	259 While implementing the function in the project itself leads to the
meillo@169	260 largest increase in code size and requires the most maintenance
meillo@169	261 and development work,
meillo@217	262 it keeps the project's dependence on other software lowest.
meillo@169	263 Using libraries or external programs requires less maintenance work
meillo@217	264 but introduces dependencies on external projects.
meillo@169	265 Programs have the smallest interfaces and provide the best separation,
meillo@87	266 but possibly limit the information exchange.
meillo@169	267 External libraries are more strongly connected than external programs,
meillo@169	268 thus information can be exchanged in a more flexible manner.
meillo@87	269 Adding code to a project increases maintenance work.
meillo@87	270 .\" XXX ref
meillo@217	271 As implementing complex functions in the project itself adds
meillo@217	272 a lot of code, this should be avoided if possible.
meillo@217	273 Thus, the dependencies only change in their character,
meillo@169	274 not in their existence.
meillo@66	275 In mmh, library dependencies on
meillo@66	276 .Pn libsasl2
meillo@66	277 and
meillo@66	278 .Pn libcrypto /\c
meillo@66	279 .Pn libssl
meillo@159	280 were traded against program dependencies on an MSA and an MRA.
meillo@159	281 This also meant trading build-time dependencies against run-time
meillo@87	282 dependencies.
meillo@169	283 Besides providing stronger separation and greater flexibility,
meillo@169	284 program dependencies also allowed
meillo@66	285 over 6\\|000 lines of code to be removed from mmh.
meillo@66	286 This made mmh's code base about 12\\|% smaller.
meillo@87	287 Reducing the project's code size by such an amount without actually
meillo@87	288 losing functionality is a convincing argument.
meillo@87	289 Actually, as external MSAs and MRAs are likely superior to the
meillo@87	290 project's internal versions, the common user even gains functionality.
meillo@66	291 .P
meillo@169	292 Users of MH should not have problems setting up an external MSA and MRA.
meillo@60	293 Also, the popular MSAs and MRAs have large communities and a lot
meillo@169	294 of available documentation.
meillo@217	295
meillo@223	296 Choices for MSAs range from small forwarders such as \fIssmtp\fP and
meillo@223	297 \fInullmailer\fP, over mid-size MTAs including \fImasqmail\fP and
meillo@223	298 \fIdma\fP, up to full-featured MTAs as for instance \fIPostfix\fP.
meillo@223	299 MRAs are provided for example by \fIfetchmail\fP, \fIgetmail\fP,
meillo@223	300 \fImpop\fP, and \fIfdm\fP.
meillo@60	301
meillo@60	302
meillo@100	303 .H2 "Non-MUA Tools
meillo@60	304 .P
meillo@217	305 One goal of mmh is to remove the tools that do not significantly
meillo@217	306 contribute to the MUA's job.
meillo@217	307 Loosely related and rarely used tools distract from a lean appearance,
meillo@217	308 and require maintenance work without adding much to the core task.
meillo@217	309 By removing these tools, mmh became more streamlined and focused.
meillo@62	310 .BU
meillo@58	311 .Pn conflict
meillo@87	312 was removed
meillo@76	313 .Ci 8b235097cbd11d728c07b966cf131aa7133ce5a9
meillo@217	314 because it is a mail system maintenance tool and not MUA-related.
meillo@87	315 It even checked
meillo@58	316 .Fn /etc/passwd
meillo@58	317 and
meillo@58	318 .Fn /etc/group
meillo@87	319 for consistency, which is completely unrelated to email.
meillo@87	320 A tool like
meillo@87	321 .Pn conflict
meillo@87	322 is surely useful, but it should not be shipped with mmh.
meillo@76	323 .\" XXX historic reasons?
meillo@62	324 .BU
meillo@58	325 .Pn rcvtty
meillo@87	326 was removed
meillo@87	327 .Ci 14767c94b3827be7c867196467ed7aea5f6f49b0
meillo@89	328 because its use case of writing to the user's terminal
meillo@200	329 on reception of mail is obsolete.
meillo@87	330 If users like to be informed of new mail, the shell's
meillo@58	331 .Ev MAILPATH
meillo@87	332 variable or graphical notifications are technically more appealing.
meillo@217	333 Writing to terminals directly is hardly ever desired today.
meillo@169	334 If, though, one prefers this approach, the standard tool
meillo@58	335 .Pn write
meillo@58	336 can be used in a way similar to:
meillo@82	337 .VS
meillo@58	338 scan -file - \| write `id -un`
meillo@82	339 VE
meillo@62	340 .BU
meillo@58	341 .Pn viamail
meillo@159	342 .\" XXX was macht viamail
meillo@87	343 was removed
meillo@87	344 .Ci eda72d6a7a7c20ff123043fb7f19c509ea01f932
meillo@87	345 when the new attachment system was activated, because
meillo@58	346 .Pn forw
meillo@76	347 could then cover the task itself.
meillo@62	348 The program
meillo@58	349 .Pn sendfiles
meillo@62	350 was rewritten as a shell script wrapper around
meillo@58	351 .Pn forw .
meillo@76	352 .Ci 0e82199cf3c991a173e0ac8aa776efdb3ded61e6
meillo@62	353 .BU
meillo@58	354 .Pn msgchk
meillo@159	355 .\" XXX was macht msgchk
meillo@87	356 was removed
meillo@87	357 .Ci bb9360ead7eb7a3fedcce2eeedfc660014e41dbe ,
meillo@87	358 because it lost its use case when POP support was removed.
meillo@76	359 A call to
meillo@58	360 .Pn msgchk
meillo@87	361 provided hardly more information than:
meillo@82	362 .VS
meillo@58	363 ls -l /var/mail/meillo
meillo@82	364 VE
meillo@217	365 Yet, it distinguished between old and new mail, but
meillo@169	366 these details can be retrieved with
meillo@76	367 .Pn stat (1),
meillo@62	368 too.
meillo@100	369 A small shell script could be written to print the information
meillo@76	370 in a similar way, if truly necessary.
meillo@76	371 As mmh's
meillo@76	372 .Pn inc
meillo@87	373 only incorporates mail from the user's local maildrop,
meillo@62	374 and thus no data transfers over slow networks are involved,
meillo@169	375 there is hardly any need to check for new mail before incorporating it.
meillo@62	376 .BU
meillo@58	377 .Pn msh
meillo@87	378 was removed
meillo@76	379 .Ci 916690191222433a6923a4be54b0d8f6ac01bd02
meillo@87	380 because the tool was in conflict with the philosophy of MH.
meillo@217	381 It provided an interactive shell to access the features of MH.
meillo@217	382 However, it was not just a shell tailored to the needs of mail handling,
meillo@217	383 but one large program that had several MH tools built in.
meillo@217	384 This conflicted with the major feature of MH of being a tool chest.
meillo@76	385 .Pn msh 's
meillo@159	386 main use case had been accessing Bulletin Boards, which have ceased to
meillo@62	387 be popular.
meillo@62	388 .P
meillo@62	389 Removing
meillo@169	390 .Pn msh
meillo@212	391 together with the truly archaic code relics
meillo@58	392 .Pn vmh
meillo@58	393 and
meillo@169	394 .Pn wmh
meillo@62	395 saved more than 7\\|000 lines of C code \(en
meillo@66	396 about 15\\|% of the project's original source code amount.
meillo@100	397 Having less code \(en with equal readability, of course \(en
meillo@76	398 for the same functionality is an advantage.
meillo@63	399 Less code means less bugs and less maintenance work.
meillo@76	400 As
meillo@63	401 .Pn rcvtty
meillo@63	402 and
meillo@63	403 .Pn msgchk
meillo@87	404 are assumed to be rarely used and can be implemented in different ways,
meillo@87	405 why should one keep them?
meillo@217	406 Removing them streamlined mmh.
meillo@63	407 .Pn viamail 's
meillo@63	408 use case is now partly obsolete and partly covered by
meillo@63	409 .Pn forw ,
meillo@217	410 hence there is no reason to still maintain it.
meillo@63	411 .Pn conflict
meillo@76	412 is not related to the mail client, and
meillo@63	413 .Pn msh
meillo@63	414 conflicts with the basic concept of MH.
meillo@169	415 These two tools might still be useful, but they should not be part of mmh.
meillo@63	416 .P
meillo@220	417 .Id slocal
meillo@169	418 Finally, there is
meillo@217	419 .Pn slocal ,
meillo@217	420 which is an MDA and thus not directly MUA-related.
meillo@217	421 It should be removed from mmh because including it conflicts with
meillo@178	422 the idea that mmh is an MUA only.
meillo@87	423 However,
meillo@76	424 .Pn slocal
meillo@76	425 provides rule-based processing of messages, like filing them into
meillo@76	426 different folders, which is otherwise not available in mmh.
meillo@87	427 Although
meillo@76	428 .Pn slocal
meillo@169	429 neither pulls in dependencies, nor does it include a separate
meillo@154	430 technical area (cf. Sec.
meillo@154	431 .Cf mail-transfer-facilities ),
meillo@169	432 it still accounts for about 1\\|000 lines of code that need to be maintained.
meillo@76	433 As
meillo@76	434 .Pn slocal
meillo@76	435 is almost self-standing, it should be split off into a separate project.
meillo@76	436 This would cut the strong connection between the MUA mmh and the MDA
meillo@76	437 .Pn slocal .
meillo@87	438 For anyone not using MH,
meillo@87	439 .Pn slocal
meillo@87	440 would become yet another independent MDA, like
meillo@87	441 .I procmail .
meillo@100	442 Then
meillo@87	443 .Pn slocal
meillo@217	444 could be installed without a complete MH system.
meillo@76	445 Likewise, mmh users could decide to use
meillo@76	446 .I procmail
meillo@217	447 without having a second, unused MDA, i.e.
meillo@87	448 .Pn slocal ,
meillo@76	449 installed.
meillo@100	450 That appears to be conceptionally the best solution.
meillo@76	451 Yet,
meillo@76	452 .Pn slocal
meillo@87	453 is not split off.
meillo@100	454 I defer the decision over
meillo@78	455 .Pn slocal
meillo@169	456 out of a need for deeper investigation.
meillo@217	457 In the meanwhile, it remains part of mmh
meillo@217	458 as its continued existence is not significant;
meillo@100	459 .Pn slocal
meillo@100	460 is unrelated to the rest of the project.
meillo@0	461
meillo@58	462
meillo@133	463
meillo@134	464 .H2 "Displaying Messages
meillo@155	465 .Id mhshow
meillo@131	466 .P
meillo@133	467 Since the very beginning, already in the first concept paper,
meillo@159	468 .\" XXX ref!!!
meillo@58	469 .Pn show
meillo@62	470 had been MH's message display program.
meillo@58	471 .Pn show
meillo@76	472 mapped message numbers and sequences to files and invoked
meillo@58	473 .Pn mhl
meillo@89	474 to have the files formatted.
meillo@173	475 With MIME, this approach was not sufficient anymore.
meillo@217	476 MIME messages can consist of multiple parts.
meillo@217	477 Some parts, like binary attachments or text content in foreign charsets,
meillo@217	478 are not directly displayable.
meillo@58	479 .Pn show 's
meillo@76	480 understanding of messages and
meillo@58	481 .Pn mhl 's
meillo@173	482 display capabilities could not cope with the task any longer.
meillo@62	483 .P
meillo@88	484 Instead of extending these tools, additional tools were written from
meillo@217	485 scratch and were added to the MH tool chest.
meillo@88	486 Doing so is encouraged by the tool chest approach.
meillo@88	487 Modular design is a great advantage for extending a system,
meillo@88	488 as new tools can be added without interfering with existing ones.
meillo@62	489 First, the new MIME features were added in form of the single program
meillo@58	490 .Pn mhn .
meillo@58	491 The command
meillo@82	492 .Cl "mhn -show 42
meillo@217	493 had then shown the message number
meillo@217	494 .Fn 42 ,
meillo@217	495 interpreting MIME.
meillo@58	496 With the 1.0 release of nmh in February 1999, Richard Coleman finished
meillo@58	497 the split of
meillo@58	498 .Pn mhn
meillo@88	499 into a set of specialized tools, which together covered the
meillo@88	500 multiple aspects of MIME.
meillo@88	501 One of them was
meillo@69	502 .Pn mhshow ,
meillo@88	503 which replaced
meillo@88	504 .Cl "mhn -show" .
meillo@88	505 It was capable of displaying MIME messages appropriately.
meillo@62	506 .P
meillo@88	507 From then on, two message display tools were part of nmh,
meillo@76	508 .Pn show
meillo@76	509 and
meillo@76	510 .Pn mhshow .
meillo@88	511 To ease the life of users,
meillo@69	512 .Pn show
meillo@69	513 was extended to automatically hand the job over to
meillo@69	514 .Pn mhshow
meillo@69	515 if displaying the message would be beyond
meillo@69	516 .Pn show 's
meillo@69	517 abilities.
meillo@88	518 In consequence, the user would simply invoke
meillo@69	519 .Pn show
meillo@69	520 (possibly through
meillo@69	521 .Pn next
meillo@69	522 or
meillo@69	523 .Pn prev )
meillo@69	524 and get the message printed with either
meillo@69	525 .Pn show
meillo@69	526 or
meillo@69	527 .Pn mhshow ,
meillo@69	528 whatever was more appropriate.
meillo@69	529 .P
meillo@217	530 Having two similar tools for basically the same task is redundancy.
meillo@217	531 Usually, users do not distinguish between
meillo@88	532 .Pn show
meillo@88	533 and
meillo@88	534 .Pn mhshow
meillo@88	535 in their daily mail reading.
meillo@217	536 Having two separate display programs was therefore unnecessary
meillo@88	537 from a user's point of view.
meillo@88	538 Besides, the development of both programs needed to be in sync,
meillo@76	539 to ensure that the programs behaved in a similar way,
meillo@76	540 because they were used like a single tool.
meillo@76	541 Different behavior would have surprised the user.
meillo@69	542 .P
meillo@69	543 Today, non-MIME messages are rather seen to be a special case of
meillo@100	544 MIME messages, although it is the other way round.
meillo@69	545 As
meillo@69	546 .Pn mhshow
meillo@217	547 already had been able to display non-MIME messages, it appeared natural
meillo@69	548 to drop
meillo@69	549 .Pn show
meillo@69	550 in favor of using
meillo@69	551 .Pn mhshow
meillo@217	552 exclusively
meillo@217	553 .Ci 4c1efddfd499300c7e74263e57d8aa137e84c853 .
meillo@88	554 Removing
meillo@88	555 .Pn show
meillo@217	556 is no loss in function, because
meillo@88	557 .Pn mhshow
meillo@88	558 covers it completely.
meillo@217	559 Yet, the old behavior of
meillo@88	560 .Pn show
meillo@88	561 can still be emulated with the simple command line:
meillo@88	562 .VS
meillo@88	563 mhl `mhpath c`
meillo@88	564 VE
meillo@88	565 .P
meillo@76	566 For convenience,
meillo@76	567 .Pn mhshow
meillo@88	568 was renamed to
meillo@88	569 .Pn show
meillo@88	570 after
meillo@88	571 .Pn show
meillo@88	572 was gone.
meillo@88	573 It is clear that such a rename may confuse future developers when
meillo@88	574 trying to understand the history.
meillo@88	575 Nevertheless, I consider the convenience on the user's side,
meillo@217	576 to outweigh the inconvenience for understanding the evolution
meillo@217	577 of the tools.
meillo@69	578 .P
meillo@88	579 To prepare for the transition,
meillo@69	580 .Pn mhshow
meillo@69	581 was reworked to behave more like
meillo@69	582 .Pn show
meillo@217	583 first (cf. Sec.
meillo@217	584 .Cf mhshow ).
meillo@164	585 .\" XXX code commits?
meillo@88	586 Once the tools behaved more alike, the replacing appeared to be
meillo@88	587 even more natural.
meillo@88	588 Today, mmh's new
meillo@69	589 .Pn show
meillo@217	590 has become the one single message display program once again,
meillo@159	591 with the difference
meillo@88	592 that today it handles MIME messages as well as non-MIME messages.
meillo@217	593 The outcomes of the transition are one program less to maintain,
meillo@88	594 no second display program for users to deal with,
meillo@88	595 and less system complexity.
meillo@69	596 .P
meillo@88	597 Still, removing the old
meillo@69	598 .Pn show
meillo@88	599 hurts in one regard: It had been such a simple program.
meillo@159	600 Its lean elegance is missing from the new
meillo@159	601 .Pn show ,
meillo@159	602 .\" XXX
meillo@159	603 however there is no alternative;
meillo@159	604 supporting MIME demands higher essential complexity.
meillo@58	605
meillo@134	606 .ig
meillo@134	607 XXX
meillo@134	608 Consider including text on scan listings here
meillo@58	609
meillo@134	610 Scan listings shall not contain body content. Hence, removed this feature.
meillo@134	611 Scan listings shall operator on message headers and non-message information
meillo@134	612 only. Displaying the beginning of the body complicates everything too much.
meillo@134	613 That's no surprise, because it's something completely different. If you
meillo@134	614 want to examine the body, then use show(1)/mhshow(1).
meillo@134	615 Changed the default scan formats accordingly.
meillo@134	616 .Ci 70b2643e0da8485174480c644ad9785c84f5bff4
meillo@134	617 ..
meillo@131	618
meillo@131	619
meillo@131	620
meillo@133	621
meillo@100	622 .H2 "Configure Options
meillo@58	623 .P
meillo@76	624 Customization is a double-edged sword.
meillo@76	625 It allows better suiting setups, but not for free.
meillo@76	626 There is the cost of code complexity to be able to customize.
meillo@76	627 There is the cost of less tested setups, because there are
meillo@171	628 more possible setups and especially corner cases.
meillo@159	629 Additionally, there is the cost of choice itself.
meillo@76	630 The code complexity directly affects the developers.
meillo@173	631 Less tested code affects both users and developers.
meillo@217	632 The problem of choice affects the users, for once by having to choose
meillo@159	633 but also by more complex interfaces that require more documentation.
meillo@159	634 Whenever options add few advantages but increase the complexity of the
meillo@159	635 system, they should be considered for removal.
meillo@72	636 I have reduced the number of project-specific configure options from
meillo@217	637 15 to 3.
meillo@74	638
meillo@76	639 .U3 "Mail Transfer Facilities
meillo@74	640 .P
meillo@217	641 With the removal of the mail transfer facilities 5 configure
meillo@85	642 options vanished:
meillo@85	643 .P
meillo@85	644 The switches
meillo@85	645 .Sw --with-tls
meillo@85	646 and
meillo@85	647 .Sw --with-cyrus-sasl
meillo@89	648 had activated the support for transfer encryption and authentication.
meillo@159	649 .\" XXX cf
meillo@159	650 .\" XXX gruende kurz wiederholen
meillo@217	651 They are not needed anymore.
meillo@85	652 .Ci fecd5d34f65597a4dfa16aeabea7d74b191532c3
meillo@85	653 .Ci 156d35f6425bea4c1ed3c4c79783dc613379c65b
meillo@85	654 .P
meillo@159	655 .\" XXX cf
meillo@159	656 .\" XXX ``For the same reason ...''
meillo@85	657 The configure switch
meillo@85	658 .Sw --enable-pop
meillo@85	659 activated the message retrieval facility.
meillo@217	660 Whereas the code area that had been conditionally compiled in
meillo@217	661 for TLS and SASL support was small,
meillo@217	662 the conditionally compiled code area for POP support was much larger.
meillo@217	663 The code base had only changed slightly on toggling TLS or SASL
meillo@217	664 support but it had changed much on toggling POP support.
meillo@85	665 The changes in the code base could hardly be overviewed.
meillo@159	666 By having POP support togglable, a second code base had been created,
meillo@85	667 one that needed to be tested.
meillo@85	668 This situation is basically similar for the conditional TLS and SASL
meillo@85	669 code, but there the changes are minor and can yet be overviewed.
meillo@85	670 Still, conditional compilation of a code base creates variations
meillo@85	671 of the original program.
meillo@85	672 More variations require more testing and maintenance work.
meillo@85	673 .P
meillo@85	674 Two other options only specified default configuration values:
meillo@100	675 .Sw --with-mts
meillo@217	676 defined the default transport service
meillo@217	677 .Ci f6aa95b724fd8c791164abe7ee5468bf5c34f226 .
meillo@85	678 With
meillo@100	679 .Sw --with-smtpservers
meillo@217	680 default SMTP servers could be specified
meillo@217	681 .Ci 128545e06224233b7e91fc4c83f8830252fe16c9 .
meillo@164	682 Both of them became irrelevant when the SMTP transport service was removed.
meillo@164	683 .\" XXX code ref
meillo@164	684 In mmh, all messages are handed over to
meillo@164	685 .Pn sendmail
meillo@164	686 for transportation.
meillo@164	687
meillo@72	688
meillo@74	689 .U3 "Backup Prefix
meillo@74	690 .P
meillo@76	691 The backup prefix is the string that was prepended to message
meillo@76	692 filenames to tag them as deleted.
meillo@173	693 By default it had been the comma character (`\fL,\fP').
meillo@159	694 .\" XXX Zeitlich ordnen
meillo@78	695 In July 2000, Kimmo Suominen introduced
meillo@78	696 the configure option
meillo@78	697 .Sw --with-hash-backup
meillo@173	698 to change the default to the hash character `\f(CW#\fP'.
meillo@217	699 This choice was probably personal preference, but,
meillo@217	700 being related or not, words that start with the hash character
meillo@78	701 introduce a comment in the Unix shell.
meillo@72	702 Thus, the command line
meillo@72	703 .Cl "rm #13 #15
meillo@72	704 calls
meillo@72	705 .Pn rm
meillo@217	706 without arguments because the first hash character starts a comment
meillo@72	707 that reaches until the end of the line.
meillo@72	708 To delete the backup files,
meillo@72	709 .Cl "rm ./#13 ./#15"
meillo@72	710 needs to be used.
meillo@217	711 Thus, using the hash as backup prefix may be seen as a precaution
meillo@217	712 against backup loss.
meillo@78	713 .P
meillo@159	714 First, I removed the configure option but added the profile entry
meillo@217	715 .Pe Backup-Prefix ,
meillo@217	716 which allowed to specify an arbitrary string as backup prefix
meillo@217	717 .Ci 6c40d481d661d532dd527eaf34cebb6d3f8ed086 .
meillo@76	718 This change did not remove the choice but moved it to a location where
meillo@217	719 it suited better, in my eyes.
meillo@76	720 .P
meillo@217	721 Eventually however, the new trash folder concept
meillo@154	722 (cf. Sec.
meillo@154	723 .Cf trash-folder )
meillo@164	724 removed the need for the backup prefix completely.
meillo@78	725 .Ci 8edc5aaf86f9f77124664f6801bc6c6cdf258173
meillo@133	726 .Ci ca0b3e830b86700d9e5e31b1784de2bdcaf58fc5
meillo@133	727
meillo@76	728
meillo@76	729 .U3 "Editor and Pager
meillo@74	730 .P
meillo@74	731 The two configure options
meillo@74	732 .CW --with-editor=EDITOR
meillo@74	733 .CW --with-pager=PAGER
meillo@74	734 were used to specify the default editor and pager at configure time.
meillo@109	735 Doing so at configure time made sense in the eighties,
meillo@76	736 when the set of available editors and pagers varied much across
meillo@76	737 different systems.
meillo@89	738 Today, the situation is more homogeneous.
meillo@74	739 The programs
meillo@74	740 .Pn vi
meillo@74	741 and
meillo@74	742 .Pn more
meillo@76	743 can be expected to be available on every Unix system,
meillo@74	744 as they are specified by POSIX since two decades.
meillo@74	745 (The specifications for
meillo@74	746 .Pn vi
meillo@74	747 and
meillo@74	748 .Pn more
meillo@74	749 appeared in
meillo@74	750 .[
meillo@74	751 posix 1987
meillo@74	752 .]
meillo@74	753 and,
meillo@74	754 .[
meillo@74	755 posix 1992
meillo@74	756 .]
meillo@74	757 respectively.)
meillo@217	758 As a first step, these two tools were hard-coded as defaults
meillo@217	759 .Ci 5d43a99db70c12a673028c7758c20cbe3e13ef5f .
meillo@74	760 Not changed were the
meillo@74	761 .Pe editor
meillo@74	762 and
meillo@74	763 .Pe moreproc
meillo@76	764 profile entries, which allowed the user to override the system defaults.
meillo@217	765 Later, the concept was reworked again to respect the standard
meillo@217	766 environment variables
meillo@74	767 .Ev VISUAL
meillo@74	768 and
meillo@74	769 .Ev PAGER
meillo@76	770 if they are set.
meillo@74	771 Today, mmh determines the editor to use in the following order,
meillo@217	772 taking the first available and non-empty item
meillo@217	773 .Ci f85f4b7ae62e3d05a945dcd46ead51f0a2a89a9b :
meillo@171	774 .LI 1
meillo@74	775 Environment variable
meillo@74	776 .Ev MMHEDITOR
meillo@171	777 .LI 2
meillo@74	778 Profile entry
meillo@74	779 .Pe Editor
meillo@171	780 .LI 3
meillo@74	781 Environment variable
meillo@74	782 .Ev VISUAL
meillo@171	783 .LI 4
meillo@74	784 Environment variable
meillo@74	785 .Ev EDITOR
meillo@171	786 .LI 5
meillo@74	787 Command
meillo@74	788 .Pn vi .
meillo@171	789 .LP
meillo@217	790 The pager to use is determined in a similar order
meillo@217	791 .Ci 0c4214ea2aec6497d0d67b436bbee9bc1d225f1e :
meillo@171	792 .LI 1
meillo@74	793 Environment variable
meillo@74	794 .Ev MMHPAGER
meillo@171	795 .LI 2
meillo@74	796 Profile entry
meillo@74	797 .Pe Pager
meillo@74	798 (replaces
meillo@74	799 .Pe moreproc )
meillo@171	800 .LI 3
meillo@74	801 Environment variable
meillo@74	802 .Ev PAGER
meillo@171	803 .LI 4
meillo@74	804 Command
meillo@74	805 .Pn more .
meillo@171	806 .LP
meillo@76	807 By respecting the
meillo@74	808 .Ev VISUAL /\c
meillo@74	809 .Ev EDITOR
meillo@74	810 and
meillo@74	811 .Ev PAGER
meillo@76	812 environment variables,
meillo@217	813 the new behavior complies with the common style on Unix systems.
meillo@217	814 It is more uniform and clearer for users.
meillo@72	815
meillo@72	816
meillo@76	817 .U3 "ndbm
meillo@72	818 .P
meillo@74	819 .Pn slocal
meillo@217	820 used to depend on the database library
meillo@217	821 .I ndbm .
meillo@217	822 The database is used to store the
meillo@217	823 .Hd Message-ID
meillo@217	824 header field values of all messages delivered.
meillo@217	825 This enabled
meillo@74	826 .Pn slocal
meillo@74	827 to suppress delivering the same message to the same user twice.
meillo@217	828 This features was enabled by the
meillo@74	829 .Sw -suppressdup
meillo@217	830 switch.
meillo@74	831 .P
meillo@217	832 As a variety of versions of the database library exist,
meillo@78	833 .[
meillo@78	834 wolter unix incompat notes dbm
meillo@78	835 .]
meillo@217	836 complicated autoconf code was needed to detect them correctly.
meillo@181	837 Furthermore, the configure switches
meillo@74	838 .Sw --with-ndbm=ARG
meillo@74	839 and
meillo@74	840 .Sw --with-ndbmheader=ARG
meillo@74	841 were added to help with difficult setups that would
meillo@217	842 not be detected automatically or not correctly.
meillo@74	843 .P
meillo@74	844 By removing the suppress duplicates feature of
meillo@74	845 .Pn slocal ,
meillo@74	846 the dependency on
meillo@74	847 .I ndbm
meillo@217	848 vanished and 120 lines of complex autoconf code could be saved
meillo@217	849 .Ci ecd6d6a20cb7a1507e3a20d6c4cb3a1cf14c6bbf .
meillo@217	850 The change removed functionality but that is considered minor to the
meillo@217	851 improvement of dropping the dependency and the complex autoconf code.
meillo@159	852 .\" XXX argument: slocal ist sowieso nicht teil vom mmh kern
meillo@72	853
meillo@217	854 .U3 "MH-E Support
meillo@72	855 .P
meillo@74	856 The configure option
meillo@74	857 .Sw --disable-mhe
meillo@217	858 was removed when the MH-E support was reworked.
meillo@217	859 MH-E is the Emacs front-end to MH.
meillo@76	860 It requires MH to provide minor additional functions.
meillo@76	861 The
meillo@76	862 .Sw --disable-mhe
meillo@217	863 configure option had switched off these extensions.
meillo@217	864 After removing the support for old versions of MH-E,
meillo@74	865 only the
meillo@74	866 .Sw -build
meillo@76	867 switches of
meillo@74	868 .Pn forw
meillo@74	869 and
meillo@74	870 .Pn repl
meillo@217	871 are left to be MH-E extensions.
meillo@76	872 They are now always built in because they add little code and complexity.
meillo@76	873 In consequence, the
meillo@74	874 .Sw --disable-mhe
meillo@76	875 configure option was removed
meillo@217	876 .Ci a7ce7b4a580d77b6c2c4d980812beb589aa4c643 .
meillo@217	877 Dropping the option also removed a variant of the code base
meillo@217	878 that would have needed to be tested.
meillo@217	879 This change was undertaken in January 2012 in nmh and
meillo@217	880 thereafter merged into mmh.
meillo@217	881
meillo@72	882
meillo@74	883 .U3 "Masquerading
meillo@72	884 .P
meillo@74	885 The configure option
meillo@74	886 .Sw --enable-masquerade
meillo@76	887 could take up to three arguments:
meillo@217	888 .Ar draft_from ,
meillo@217	889 .Ar mmailid ,
meillo@217	890 and
meillo@217	891 .Ar username_extension .
meillo@74	892 They activated different types of address masquerading.
meillo@74	893 All of them were implemented in the SMTP-speaking
meillo@74	894 .Pn post
meillo@217	895 command.
meillo@76	896 Address masquerading is an MTA's task and mmh does not cover
meillo@76	897 this field anymore.
meillo@76	898 Hence, true masquerading needs to be implemented in the external MTA.
meillo@74	899 .P
meillo@74	900 The
meillo@74	901 .I mmailid
meillo@74	902 masquerading type is the oldest one of the three and the only one
meillo@74	903 available in the original MH.
meillo@74	904 It provided a
meillo@74	905 .I username
meillo@74	906 to
meillo@74	907 .I fakeusername
meillo@217	908 mapping, based on the
meillo@217	909 .Fn passwd 's
meillo@217	910 GECOS field.
meillo@217	911 Nmh's man page
meillo@181	912 .Mp mh-tailor (5)
meillo@74	913 described the use case as being the following:
meillo@98	914 .QS
meillo@74	915 This is useful if you want the messages you send to always
meillo@74	916 appear to come from the name of an MTA alias rather than your
meillo@74	917 actual account name. For instance, many organizations set up
meillo@74	918 `First.Last' sendmail aliases for all users. If this is
meillo@74	919 the case, the GECOS field for each user should look like:
meillo@74	920 ``First [Middle] Last <First.Last>''
meillo@98	921 .QE
meillo@74	922 .P
meillo@74	923 As mmh sends outgoing mail via the local MTA only,
meillo@76	924 the best location to do such global rewrites is there.
meillo@74	925 Besides, the MTA is conceptionally the right location because it
meillo@74	926 does the reverse mapping for incoming mail (aliasing), too.
meillo@181	927 Furthermore, masquerading set up there is readily available for all
meillo@74	928 mail software on the system.
meillo@76	929 Hence, mmailid masquerading was removed.
meillo@74	930 .Ci 0836c8000ccb34b59410ef1c15b1b7feac70ce5f
meillo@74	931 .P
meillo@74	932 The
meillo@74	933 .I username_extension
meillo@76	934 masquerading type did not replace the username but would append a suffix,
meillo@76	935 specified by the
meillo@74	936 .Ev USERNAME_EXTENSION
meillo@76	937 environment variable, to it.
meillo@76	938 This provided support for the
meillo@74	939 .I user-extension
meillo@217	940 feature of qmail
meillo@217	941 .[ [
meillo@217	942 sill qmail handbook
meillo@217	943 .], p. 141]
meillo@217	944 and the similar
meillo@74	945 .I "plussed user
meillo@217	946 processing of Sendmail.
meillo@217	947 .[ [
meillo@217	948 sendmail costales
meillo@217	949 .], p. 476]
meillo@217	950 The decision to remove this username_extension masquerading
meillo@217	951 was motivated by the fact that
meillo@74	952 .Pn spost
meillo@217	953 had not supported it yet.
meillo@217	954 Username extensions can be used in mmh, but less convenient.
meillo@159	955 .\" XXX covered by next paragraph
meillo@76	956 .\" XXX format file %(getenv USERNAME_EXTENSION)
meillo@217	957 .Ci 2abae0bfd0ad5bf898461e50aa4b466d641f23d9
meillo@74	958 .P
meillo@74	959 The
meillo@74	960 .I draft_from
meillo@74	961 masquerading type instructed
meillo@74	962 .Pn post
meillo@84	963 to use the value of the
meillo@84	964 .Hd From
meillo@84	965 header field as SMTP envelope sender.
meillo@76	966 Sender addresses could be replaced completely.
meillo@76	967 Mmh offers a kind of masquerading similar in effect, but
meillo@74	968 with technical differences.
meillo@76	969 As mmh does not transfer messages itself, the local MTA has final control
meillo@217	970 over the sender's address.
meillo@217	971 Any masquerading mmh introduces may be reverted by the MTA.
meillo@76	972 In times of pedantic spam checking, an MTA will take care to use
meillo@76	973 sensible envelope sender addresses to keep its own reputation up.
meillo@84	974 Nonetheless, the MUA can set the
meillo@84	975 .Hd From
meillo@217	976 header field and thereby propose a sender address to the MTA.
meillo@74	977 The MTA may then decide to take that one or generate the canonical sender
meillo@74	978 address for use as envelope sender address.
meillo@217	979 .Ci b14ea6073f77b4359aaf3fddd0e105989db9
meillo@74	980 .P
meillo@74	981 In mmh, the MTA will always extract the recipient and sender from the
meillo@84	982 message header (\c
meillo@74	983 .Pn sendmail 's
meillo@74	984 .Sw -t
meillo@74	985 switch).
meillo@84	986 The
meillo@84	987 .Hd From
meillo@84	988 header field of the draft may be set arbitrary by the user.
meillo@74	989 If it is missing, the canonical sender address will be generated by the MTA.
meillo@74	990
meillo@74	991 .U3 "Remaining Options
meillo@74	992 .P
meillo@74	993 Two configure options remain in mmh.
meillo@74	994 One is the locking method to use:
meillo@74	995 .Sw --with-locking=[dot\|fcntl\|flock\|lockf] .
meillo@217	996 The idea of removing all methods except the portable
meillo@217	997 .I "dot locking
meillo@76	998 and having that one as the default is appealing, but this change
meillo@76	999 requires deeper technical investigation into the topic.
meillo@76	1000 The other option,
meillo@74	1001 .Sw --enable-debug ,
meillo@217	1002 compiles the programs with debugging symbols.
meillo@74	1003 This option is likely to stay.
meillo@72	1004
meillo@72	1005
meillo@58	1006
meillo@63	1007
meillo@100	1008 .H2 "Command Line Switches
meillo@58	1009 .P
meillo@217	1010 The command line switches of MH tools follow a style similar to
meillo@217	1011 the X Window System style.
meillo@171	1012 .\" XXX ref
meillo@217	1013 The switches consist of a single dash (`\fL-\fP') followed by a word.
meillo@217	1014 For example
meillo@217	1015 .Cl -truncate .
meillo@212	1016 To ease typing, the word can be abbreviated, given the remaining
meillo@217	1017 prefix is unambiguous.
meillo@212	1018 If no other switch starts with the letter `t', then any of
meillo@212	1019 .Cl "-truncate" ,
meillo@212	1020 .Cl "-trunc" ,
meillo@212	1021 .Cl "-tr" ,
meillo@212	1022 and
meillo@212	1023 .Cl "-t
meillo@212	1024 is equal.
meillo@212	1025 As a result, switches can neither be grouped (as in
meillo@212	1026 .Cl "ls -ltr" )
meillo@212	1027 nor can switch arguments be appended directly to the switch (as in
meillo@212	1028 .Cl "sendmail -q30m" ).
meillo@212	1029 Many switches have negating counter-parts, which start with `no'.
meillo@212	1030 For example
meillo@212	1031 .Cl "-notruncate
meillo@212	1032 inverts the
meillo@212	1033 .Cl "-truncate
meillo@212	1034 switch.
meillo@212	1035 They exist to override the effect of default switches in the profile.
meillo@93	1036 Every program in mmh has two generic switches:
meillo@93	1037 .Sw -help ,
meillo@93	1038 to print a short message on how to use the program, and
meillo@159	1039 .Sw -Version
meillo@164	1040 (with capital `V'), to tell what version of mmh the program belongs to.
meillo@93	1041 .P
meillo@93	1042 Switches change the behavior of programs.
meillo@93	1043 Programs that do one thing in one way require no switches.
meillo@93	1044 In most cases, doing something in exactly one way is too limiting.
meillo@217	1045 If one task should be accomplished in various ways,
meillo@217	1046 switches are a good approach to alter the behavior of a program.
meillo@93	1047 Changing the behavior of programs provides flexibility and customization
meillo@217	1048 to users, but at the same time it complicates the code,
meillo@217	1049 the documentation, and the usage of the program.
meillo@97	1050 .\" XXX: Ref
meillo@93	1051 Therefore, the number of switches should be kept small.
meillo@217	1052 A small set of well-chosen switches is best.
meillo@217	1053 Usually, the number of switches increases over time.
meillo@93	1054 Already in 1985, Rose and Romine have identified this as a major
meillo@93	1055 problem of MH:
meillo@93	1056 .[ [
meillo@93	1057 rose romine real work
meillo@93	1058 .], p. 12]
meillo@98	1059 .QS
meillo@93	1060 A complaint often heard about systems which undergo substantial development
meillo@93	1061 by many people over a number of years, is that more and more options are
meillo@93	1062 introduced which add little to the functionality but greatly increase the
meillo@93	1063 amount of information a user needs to know in order to get useful work done.
meillo@93	1064 This is usually referred to as creeping featurism.
meillo@93	1065 .QP
meillo@93	1066 Unfortunately MH, having undergone six years of off-and-on development by
meillo@93	1067 ten or so well-meaning programmers (the present authors included),
meillo@93	1068 suffers mightily from this.
meillo@98	1069 .QE
meillo@93	1070 .P
meillo@217	1071 Being reluctant to adding new switches (or \fIoptions\fP,
meillo@217	1072 as Rose and Romine call them) is one part of a counter-action,
meillo@97	1073 the other part is removing hardly used switches.
meillo@217	1074 Nmh's tools have lots of switches already implemented.
meillo@217	1075 Hence, cleaning up by removing some of them was the more important part
meillo@97	1076 of the counter-action.
meillo@93	1077 Removing existing functionality is always difficult because it
meillo@93	1078 breaks programs that use these functions.
meillo@93	1079 Also, for every obsolete feature, there'll always be someone who still
meillo@93	1080 uses it and thus opposes its removal.
meillo@93	1081 This puts the developer into the position,
meillo@93	1082 where sensible improvements to style are regarded as destructive acts.
meillo@97	1083 Yet, living with the featurism is far worse, in my eyes, because
meillo@97	1084 future needs will demand adding further features,
meillo@93	1085 worsening the situation more and more.
meillo@93	1086 Rose and Romine added in a footnote,
meillo@93	1087 ``[...]
meillo@93	1088 .Pn send
meillo@217	1089 will no doubt acquire an endless number of switches in the years to come''
meillo@217	1090 .[ [
meillo@217	1091 rose romine real work
meillo@217	1092 .], p. 12].
meillo@97	1093 Although clearly humorous, the comment points to the nature of the problem.
meillo@97	1094 Refusing to add any new switches would encounter the problem at its root,
meillo@97	1095 but this is not practical.
meillo@97	1096 New needs will require new switches and it would be unwise to block
meillo@97	1097 them strictly.
meillo@97	1098 Nevertheless, removing obsolete switches still is an effective approach
meillo@97	1099 to deal with the problem.
meillo@97	1100 Working on an experimental branch without an established user base,
meillo@97	1101 eased my work because I did not offend users when I removed existing
meillo@110	1102 functions.
meillo@93	1103 .P
meillo@93	1104 Rose and Romine counted 24 visible and 9 more hidden switches for
meillo@93	1105 .Pn send .
meillo@97	1106 In nmh, they increased up to 32 visible and 12 hidden ones.
meillo@182	1107 At the time of writing, no more than 4 visible switches and 1 hidden switch
meillo@97	1108 have remained in mmh's
meillo@97	1109 .Pn send .
meillo@217	1110 These numbers include the two generic switches,
meillo@182	1111 .Sw -help
meillo@182	1112 and
meillo@183	1113 .Sw -Version .
meillo@217	1114 .P
meillo@183	1115 Hidden switches are ones not documented.
meillo@183	1116 In mmh, 12 tools have hidden switches.
meillo@183	1117 9 of them are
meillo@183	1118 .Sw -debug
meillo@183	1119 switches, the other 6 provide special interfaces for internal use.
meillo@93	1120 .P
meillo@217	1121 The following figure displays the number of switches for each of the tools
meillo@159	1122 that is available in both nmh and mmh.
meillo@100	1123 The tools are sorted by the number of switches they had in nmh.
meillo@217	1124 Both visible and hidden switches were counted,
meillo@97	1125 but not the generic help and version switches.
meillo@93	1126 Whereas in the beginning of the project, the average tool had 11 switches,
meillo@93	1127 now it has no more than 5 \(en only half as many.
meillo@93	1128 If the `no' switches and similar inverse variant are folded onto
meillo@100	1129 their counter-parts, the average tool had 8 switches in pre-mmh times and
meillo@100	1130 has 4 now.
meillo@93	1131 The total number of functional switches in mmh dropped from 465
meillo@182	1132 to 233.
meillo@58	1133
meillo@93	1134 .KS
meillo@93	1135 .in 1c
meillo@93	1136 .so input/switches.grap
meillo@93	1137 .KE
meillo@58	1138
meillo@93	1139 .P
meillo@93	1140 A part of the switches vanished after functions were removed.
meillo@93	1141 This was the case for network mail transfer, for instance.
meillo@97	1142 Sometimes, however, the work flow was the other way:
meillo@97	1143 I looked through the
meillo@97	1144 .Mp mh-chart (7)
meillo@97	1145 man page to identify the tools with apparently too many switches.
meillo@217	1146 Then I considered the benefit of each switch by examining
meillo@217	1147 the tool's man page and source code, aided by literature research
meillo@217	1148 and testing.
meillo@97	1149
meillo@58	1150
meillo@93	1151 .U3 "Draft Folder Facility
meillo@93	1152 .P
meillo@100	1153 A change early in the project was the complete transition from
meillo@217	1154 the single draft message to the draft folder facility
meillo@217	1155 .Ci 337338b404931f06f0db2119c9e145e8ca5a9860 .
meillo@164	1156 .\" XXX ref to section ...
meillo@109	1157 The draft folder facility was introduced in the mid-eighties, when
meillo@100	1158 Rose and Romine called it a ``relatively new feature''.
meillo@93	1159 .[
meillo@93	1160 rose romine real work
meillo@93	1161 .]
meillo@217	1162 Since then, the facility was included, inactive by default.
meillo@217	1163 By making it permanently active and by related rework of the tools, the
meillo@93	1164 .Sw -[no]draftfolder ,
meillo@93	1165 and
meillo@93	1166 .Sw -draftmessage
meillo@217	1167 switches could be removed from
meillo@93	1168 .Pn comp ,
meillo@93	1169 .Pn repl ,
meillo@93	1170 .Pn forw ,
meillo@93	1171 .Pn dist ,
meillo@93	1172 .Pn whatnow ,
meillo@93	1173 and
meillo@217	1174 .Pn send
meillo@217	1175 .Ci 337338b404931f06f0db2119c9e145e8ca5a9860 .
meillo@217	1176 The only flexibility lost with this change is having multiple
meillo@97	1177 draft folders within one profile.
meillo@97	1178 I consider this a theoretical problem only.
meillo@159	1179 At the same time, the
meillo@93	1180 .Sw -draft
meillo@93	1181 switch of
meillo@93	1182 .Pn anno ,
meillo@93	1183 .Pn refile ,
meillo@93	1184 and
meillo@93	1185 .Pn send
meillo@93	1186 was removed.
meillo@159	1187 The special treatment of \fIthe\fP draft message became irrelevant after
meillo@217	1188 the rework of the draft system
meillo@159	1189 (cf. Sec.
meillo@217	1190 .Cf draft-folder ).
meillo@164	1191 Furthermore,
meillo@95	1192 .Pn comp
meillo@164	1193 no longer needs a
meillo@95	1194 .Sw -file
meillo@164	1195 switch as the draft folder facility together with the
meillo@95	1196 .Sw -form
meillo@164	1197 switch are sufficient.
meillo@93	1198
meillo@95	1199
meillo@102	1200 .U3 "In Place Editing
meillo@93	1201 .P
meillo@93	1202 .Pn anno
meillo@93	1203 had the switches
meillo@93	1204 .Sw -[no]inplace
meillo@100	1205 to either annotate the message in place and thus preserve hard links,
meillo@217	1206 or annotate a copy to replace the original message.
meillo@217	1207 The latter approach broke hard links.
meillo@97	1208 Following the assumption that linked messages should truly be the
meillo@217	1209 same message and annotating it should not break the link, the
meillo@93	1210 .Sw -[no]inplace
meillo@93	1211 switches were removed and the previous default
meillo@93	1212 .Sw -inplace
meillo@217	1213 was made the definitive behavior
meillo@217	1214 .Ci c8195849d2e366c569271abb0f5f60f4ebf0b4d0 .
meillo@93	1215 The
meillo@93	1216 .Sw -[no]inplace
meillo@93	1217 switches of
meillo@93	1218 .Pn repl ,
meillo@93	1219 .Pn forw ,
meillo@93	1220 and
meillo@93	1221 .Pn dist
meillo@217	1222 could be removed, as well, as they were simply passed through to
meillo@93	1223 .Pn anno .
meillo@93	1224 .P
meillo@93	1225 .Pn burst
meillo@93	1226 also had
meillo@93	1227 .Sw -[no]inplace
meillo@217	1228 switches, but with a different meaning.
meillo@95	1229 With
meillo@95	1230 .Sw -inplace ,
meillo@95	1231 the digest had been replaced by the table of contents (i.e. the
meillo@110	1232 introduction text) and the burst messages were placed right
meillo@95	1233 after this message, renumbering all following messages.
meillo@95	1234 Also, any trailing text of the digest was lost, though,
meillo@95	1235 in practice, it usually consists of an end-of-digest marker only.
meillo@217	1236 Nonetheless, this behavior appeared less elegant than the
meillo@95	1237 .Sw -noinplace
meillo@95	1238 behavior, which already had been the default.
meillo@95	1239 Nmh's
meillo@95	1240 .Mp burst (1)
meillo@95	1241 man page reads:
meillo@98	1242 .QS
meillo@164	1243 If
meillo@164	1244 .Sw -noinplace
meillo@164	1245 is given, each digest is preserved, no table
meillo@93	1246 of contents is produced, and the messages contained within
meillo@93	1247 the digest are placed at the end of the folder. Other messages
meillo@93	1248 are not tampered with in any way.
meillo@98	1249 .QE
meillo@95	1250 .LP
meillo@93	1251 The decision to drop the
meillo@93	1252 .Sw -inplace
meillo@95	1253 behavior was supported by the code complexity and the possible data loss
meillo@95	1254 it caused.
meillo@93	1255 .Sw -noinplace
meillo@95	1256 was chosen to be the definitive behavior.
meillo@97	1257 .Ci 68a686adeb39223a5e1ad35e4a24890ec053679d
meillo@93	1258
meillo@95	1259
meillo@95	1260 .U3 "Forms and Format Strings
meillo@93	1261 .P
meillo@95	1262 Historically, the tools that had
meillo@95	1263 .Sw -form
meillo@95	1264 switches to supply a form file had
meillo@95	1265 .Sw -format
meillo@95	1266 switches as well to supply the contents of a form file as a string
meillo@95	1267 on the command line directly.
meillo@95	1268 In consequence, the following two lines equaled:
meillo@95	1269 .VS
meillo@95	1270 scan -form scan.mailx
meillo@217	1271 scan -format "`cat /path/to/scan.mailx`"
meillo@95	1272 VE
meillo@95	1273 The
meillo@95	1274 .Sw -format
meillo@95	1275 switches were dropped in favor for extending the
meillo@95	1276 .Sw -form
meillo@217	1277 switches
meillo@217	1278 .Ci f51956be123db66b00138f80464d06f030dbb88d .
meillo@217	1279 If their argument starts with an equal sign (`\fL=\fP'),
meillo@95	1280 then the rest of the argument is taken as a format string,
meillo@95	1281 otherwise the arguments is treated as the name of a format file.
meillo@95	1282 Thus, now the following two lines equal:
meillo@95	1283 .VS
meillo@95	1284 scan -form scan.mailx
meillo@217	1285 scan -form "=`cat /path/to/scan.mailx`"
meillo@95	1286 VE
meillo@95	1287 This rework removed the prefix collision between
meillo@95	1288 .Sw -form
meillo@95	1289 and
meillo@95	1290 .Sw -format .
meillo@217	1291 Typing `\fL-fo\fP' is sufficient to specify form file or format string.
meillo@95	1292 .P
meillo@95	1293 The different meaning of
meillo@95	1294 .Sw -format
meillo@95	1295 for
meillo@217	1296 .Pn forw
meillo@217	1297 and
meillo@95	1298 .Pn repl
meillo@95	1299 was removed in mmh.
meillo@95	1300 .Pn forw
meillo@95	1301 was completely switched to MIME-type forwarding, thus removing the
meillo@217	1302 .Sw -[no]format
meillo@217	1303 .Ci 6e271608b7b9c23771523f88d23a4d3593010cf1 .
meillo@95	1304 For
meillo@95	1305 .Pn repl ,
meillo@95	1306 the
meillo@95	1307 .Sw -[no]format
meillo@95	1308 switches were reworked to
meillo@95	1309 .Sw -[no]filter
meillo@217	1310 switches
meillo@217	1311 .Ci 67411b1f95d6ec987b4c732459e1ba8a8ac192c6 .
meillo@95	1312 The
meillo@95	1313 .Sw -format
meillo@95	1314 switches of
meillo@95	1315 .Pn send
meillo@95	1316 and
meillo@95	1317 .Pn post ,
meillo@217	1318 which had a third meaning, were removed likewise
meillo@217	1319 .Ci f3cb7cde0e6f10451b6848678d95860d512224b9 .
meillo@95	1320 Eventually, the ambiguity of the
meillo@95	1321 .Sw -format
meillo@217	1322 switches is resolved by not having such switches anymore in mmh.
meillo@95	1323
meillo@95	1324
meillo@95	1325 .U3 "MIME Tools
meillo@95	1326 .P
meillo@217	1327 The MIME tools, which once were part of
meillo@100	1328 .Pn mhn
meillo@164	1329 (whatever that stood for),
meillo@95	1330 had several switches that added little practical value to the programs.
meillo@95	1331 The
meillo@95	1332 .Sw -[no]realsize
meillo@95	1333 switches of
meillo@95	1334 .Pn mhbuild
meillo@95	1335 and
meillo@95	1336 .Pn mhlist
meillo@217	1337 were removed
meillo@217	1338 .Ci 8d8f1c3abc586c005c904e52c4adbfe694d2201c .
meillo@217	1339 Real size calculations are done always now because nmh's
meillo@159	1340 .Mp mhbuild (1)
meillo@217	1341 man page states that
meillo@217	1342 ``This provides an accurate count at the expense of a small delay''
meillo@217	1343 with the small delay not being noticable on modern systems.
meillo@95	1344 .P
meillo@95	1345 The
meillo@95	1346 .Sw -[no]check
meillo@95	1347 switches were removed together with the support for
meillo@95	1348 .Hd Content-MD5
meillo@217	1349 header fields [RFC\\|1864]
meillo@154	1350 (cf. Sec.
meillo@154	1351 .Cf content-md5 )
meillo@217	1352 .Ci 31dc797eb5178970d68962ca8939da3fd9a8efda .
meillo@95	1353 .P
meillo@95	1354 The
meillo@95	1355 .Sw -[no]ebcdicsafe
meillo@95	1356 and
meillo@95	1357 .Sw -[no]rfc934mode
meillo@95	1358 switches of
meillo@95	1359 .Pn mhbuild
meillo@217	1360 were removed because they are considered obsolete
meillo@97	1361 .Ci 01a3480928da485b4d6109d36d751dfa71799d58
meillo@217	1362 .Ci 3363e2624dce0eb8164cf8b3f1ab385c8ff72e88 .
meillo@95	1363 .P
meillo@95	1364 Content caching of external MIME parts, activated with the
meillo@95	1365 .Sw -rcache
meillo@95	1366 and
meillo@95	1367 .Sw -wcache
meillo@217	1368 switches was completely removed
meillo@217	1369 .Ci d1fefd9f614e4dc3cda16da6c69133c1b2005269 .
meillo@97	1370 External MIME parts are rare today, having a caching facility
meillo@159	1371 for them appears to be unnecessary.
meillo@95	1372 .P
meillo@95	1373 In pre-MIME times,
meillo@95	1374 .Pn mhl
meillo@95	1375 had covered many tasks that are part of MIME handling today.
meillo@95	1376 Therefore,
meillo@95	1377 .Pn mhl
meillo@95	1378 could be simplified to a large extend, reducing the number of its
meillo@217	1379 switches from 21 to 6
meillo@97	1380 .Ci 350ad6d3542a07639213cf2a4fe524e829c1e7b6
meillo@217	1381 .Ci 0e46503be3c855bddaeae3843e1b659279c35d70 .
meillo@95	1382
meillo@95	1383
meillo@95	1384
meillo@95	1385
meillo@95	1386 .U3 "Header Printing
meillo@95	1387 .P
meillo@95	1388 .Pn folder 's
meillo@95	1389 data output is self-explaining enough that
meillo@159	1390 displaying the header line makes little sense.
meillo@95	1391 Hence, the
meillo@95	1392 .Sw -[no]header
meillo@217	1393 switch was removed and headers are never printed
meillo@217	1394 .Ci 601cc73d1fa05ce96faa728f036d6c51b91701c7 .
meillo@95	1395 .P
meillo@95	1396 In
meillo@95	1397 .Pn mhlist ,
meillo@95	1398 the
meillo@95	1399 .Sw -[no]header
meillo@217	1400 switches were removed, as well
meillo@217	1401 .Ci b24f96523aaf60e44e04a3ffb1d22e69a13a602f .
meillo@217	1402 In this case, the headers are printed always because the output
meillo@217	1403 is not self-explaining.
meillo@95	1404 .P
meillo@95	1405 .Pn scan
meillo@95	1406 also had
meillo@95	1407 .Sw -[no]header
meillo@95	1408 switches.
meillo@217	1409 Printing this header had been sensible until the introduction of
meillo@217	1410 format strings made it impossible to display column headings.
meillo@95	1411 Only the folder name and the current date remained to be printed.
meillo@217	1412 As this information can be perfectly generated with
meillo@95	1413 .Pn folder
meillo@95	1414 and
meillo@95	1415 .Pn date ,
meillo@217	1416 the switches were removed
meillo@217	1417 .Ci c477dc5d1d03fa6d9a8ab3dd3508c63cbddc044e .
meillo@95	1418 .P
meillo@95	1419 By removing all
meillo@95	1420 .Sw -header
meillo@95	1421 switches, the collision with
meillo@95	1422 .Sw -help
meillo@95	1423 on the first two letters was resolved.
meillo@95	1424 Currently,
meillo@95	1425 .Sw -h
meillo@95	1426 evaluates to
meillo@95	1427 .Sw -help
meillo@95	1428 for all tools of mmh.
meillo@95	1429
meillo@95	1430
meillo@139	1431 .U3 "Suppressing Edits or the Invocation of the WhatNow Shell
meillo@95	1432 .P
meillo@95	1433 The
meillo@95	1434 .Sw -noedit
meillo@100	1435 switch of
meillo@95	1436 .Pn comp ,
meillo@95	1437 .Pn repl ,
meillo@95	1438 .Pn forw ,
meillo@95	1439 .Pn dist ,
meillo@95	1440 and
meillo@95	1441 .Pn whatnow
meillo@217	1442 was removed and replaced by specifying
meillo@95	1443 .Sw -editor
meillo@217	1444 with an empty argument
meillo@217	1445 .Ci 75fca31a5b9d5c1a99c74ab14c94438d8852fba9 .
meillo@95	1446 (Specifying
meillo@159	1447 .Cl "-editor /bin/true
meillo@217	1448 is nearly the same. It differs only in setting the previous editor.)
meillo@95	1449 .P
meillo@95	1450 The more important change is the removal of the
meillo@95	1451 .Sw -nowhatnowproc
meillo@217	1452 switch
meillo@217	1453 .Ci ee4f43cf2ef0084ec698e4e87159a94c01940622 .
meillo@217	1454 This switch had once introduced an awkward behavior,
meillo@217	1455 as explained in nmh's man page for
meillo@95	1456 .Mp comp (1):
meillo@98	1457 .QS
meillo@164	1458 The
meillo@164	1459 .Sw -editor
meillo@164	1460 .Ar editor
meillo@164	1461 switch indicates the editor to use for
meillo@164	1462 the initial edit. Upon exiting from the editor,
meillo@164	1463 .Pn comp
meillo@164	1464 will invoke the
meillo@164	1465 .Pn whatnow
meillo@164	1466 program. See
meillo@164	1467 .Mp whatnow (1)
meillo@164	1468 for a discussion of available options.
meillo@164	1469 The invocation of this program can be
meillo@164	1470 inhibited by using the
meillo@164	1471 .Sw -nowhatnowproc
meillo@164	1472 switch. (In truth of fact, it is the
meillo@164	1473 .Pn whatnow
meillo@164	1474 program which starts the initial edit.
meillo@164	1475 Hence,
meillo@164	1476 .Sw -nowhatnowproc
meillo@164	1477 will prevent any edit from occurring.)
meillo@98	1478 .QE
meillo@95	1479 .P
meillo@95	1480 Effectively, the
meillo@95	1481 .Sw -nowhatnowproc
meillo@217	1482 switch caused only only a draft message to be created.
meillo@95	1483 As
meillo@159	1484 .Cl "-whatnowproc /bin/true
meillo@217	1485 does the same, the
meillo@95	1486 .Sw -nowhatnowproc
meillo@95	1487 switch was removed for being redundant.
meillo@95	1488
meillo@95	1489
meillo@95	1490
meillo@95	1491 .U3 "Various
meillo@95	1492 .BU
meillo@139	1493 With the removal of MMDF maildrop format support,
meillo@139	1494 .Pn packf
meillo@139	1495 and
meillo@139	1496 .Pn rcvpack
meillo@139	1497 no longer needed their
meillo@139	1498 .Sw -mbox
meillo@139	1499 and
meillo@139	1500 .Sw -mmdf
meillo@139	1501 switches.
meillo@217	1502 The behavior of
meillo@139	1503 .Sw -mbox
meillo@217	1504 is the sole behavior now
meillo@217	1505 .Ci 3916ab66ad5d183705ac12357621ea8661afd3c0 .
meillo@171	1506 Further rework in both tools made the
meillo@139	1507 .Sw -file
meillo@217	1508 switch unnecessary
meillo@217	1509 .Ci ca1023716d4c2ab890696f3e41fa0d94267a940e .
meillo@139	1510
meillo@139	1511 .BU
meillo@217	1512 Mmh's tools do no longer clear the screen (\c
meillo@139	1513 .Pn scan 's
meillo@139	1514 and
meillo@139	1515 .Pn mhl 's
meillo@139	1516 .Sw -[no]clear
meillo@139	1517 switches
meillo@139	1518 .Ci e57b17343dcb3ff373ef4dd089fbe778f0c7c270
meillo@139	1519 .Ci 943765e7ac5693ae177fd8d2b5a2440e53ce816e ).
meillo@217	1520 Neither does
meillo@139	1521 .Pn mhl
meillo@139	1522 ring the bell (\c
meillo@139	1523 .Sw -[no]bell
meillo@139	1524 .Ci e11983f44e59d8de236affa5b0d0d3067c192e24 )
meillo@217	1525 nor does it page the output itself (\c
meillo@139	1526 .Sw -length
meillo@139	1527 .Ci 5b9d883db0318ed2b84bb82dee880d7381f99188 ).
meillo@159	1528 .\" XXX Ref
meillo@139	1529 Generally, the pager to use is no longer specified with the
meillo@139	1530 .Sw -[no]moreproc
meillo@139	1531 command line switches for
meillo@139	1532 .Pn mhl
meillo@139	1533 and
meillo@139	1534 .Pn show /\c
meillo@217	1535 .Pn mhshow
meillo@217	1536 .Ci 39e87a75b5c2d3572ec72e717720b44af291e88a .
meillo@139	1537
meillo@139	1538 .BU
meillo@96	1539 In order to avoid prefix collisions among switch names, the
meillo@95	1540 .Sw -version
meillo@95	1541 switch was renamed to
meillo@95	1542 .Sw -Version
meillo@217	1543 (with capital `V')
meillo@217	1544 .Ci 32b2354dbaf4bf934936eb5b102a4a3d2fdd209a .
meillo@95	1545 Every program has the
meillo@95	1546 .Sw -version
meillo@95	1547 switch but its first three letters collided with the
meillo@95	1548 .Sw -verbose
meillo@95	1549 switch, present in many programs.
meillo@95	1550 The rename solved this problem once for all.
meillo@95	1551 Although this rename breaks a basic interface, having the
meillo@95	1552 .Sw -V
meillo@95	1553 abbreviation to display the version information, isn't all too bad.
meillo@139	1554
meillo@95	1555 .BU
meillo@95	1556 .Sw -[no]preserve
meillo@95	1557 of
meillo@95	1558 .Pn refile
meillo@168	1559 was removed
meillo@168	1560 .Ci 8edc5aaf86f9f77124664f6801bc6c6cdf258173
meillo@168	1561 because what use was it anyway?
meillo@168	1562 Quoting nmh's man page
meillo@165	1563 .Mp refile (1):
meillo@98	1564 .QS
meillo@95	1565 Normally when a message is refiled, for each destination
meillo@95	1566 folder it is assigned the number which is one above the current
meillo@95	1567 highest message number in that folder. Use of the
meillo@164	1568 .Sw -preserv
meillo@164	1569 [sic!] switch will override this message renaming, and try
meillo@95	1570 to preserve the number of the message. If a conflict for a
meillo@164	1571 particular folder occurs when using the
meillo@164	1572 .Sw -preserve
meillo@164	1573 switch, then
meillo@164	1574 .Pn refile
meillo@164	1575 will use the next available message number which
meillo@95	1576 is above the message number you wish to preserve.
meillo@98	1577 .QE
meillo@139	1578
meillo@95	1579 .BU
meillo@95	1580 The removal of the
meillo@95	1581 .Sw -[no]reverse
meillo@95	1582 switches of
meillo@95	1583 .Pn scan
meillo@97	1584 .Ci 8edc5aaf86f9f77124664f6801bc6c6cdf258173
meillo@217	1585 is a bug fix.
meillo@217	1586 This is supported by the comments
meillo@95	1587 ``\-[no]reverse under #ifdef BERK (I really HATE this)''
meillo@95	1588 by Rose and
meillo@95	1589 ``Lists messages in reverse order with the `\-reverse' switch.
meillo@217	1590 This should be considered a bug'' by Romine in the changelogs.
meillo@217	1591 The question remains why neither Rose nor Romine have fixed this
meillo@217	1592 bug in the eighties when they wrote these comments.
meillo@93	1593
meillo@93	1594
meillo@93	1595
meillo@102	1596
meillo@102	1597
meillo@95	1598
meillo@95	1599
meillo@133	1600 .\" --------------------------------------------------------------
meillo@74	1601 .H1 "Modernizing
meillo@102	1602 .P
meillo@164	1603 In the more than thirty years of MH's existence, its code base was
meillo@159	1604 increasingly extended.
meillo@118	1605 New features entered the project and became alternatives to the
meillo@118	1606 existing behavior.
meillo@217	1607 Relics from several decades have gathered in the code base
meillo@118	1608 but seldom obsolete features were dropped.
meillo@118	1609 This section describes the removing of old code
meillo@118	1610 and the modernizing of the default setup.
meillo@118	1611 It focuses on the functional aspect only;
meillo@154	1612 the non-functional aspects of code style are discussed in Sec.
meillo@154	1613 .Cf code-style .
meillo@58	1614
meillo@58	1615
meillo@212	1616 .H2 "Code Relics
meillo@0	1617 .P
meillo@217	1618 My position regarding the removal of obsolete code
meillo@159	1619 is much more revolutional than the nmh community appreciates.
meillo@217	1620 Working on an experimental version, I was able to quickly drop
meillo@217	1621 functionality that I considered ancient.
meillo@104	1622 The need for consensus with peers would have slowed this process down.
meillo@104	1623 Without the need to justify my decisions, I was able to rush forward.
meillo@217	1624 .P
meillo@110	1625 In December 2011, Paul Vixie motivated the nmh developers to just
meillo@104	1626 do the work:
meillo@104	1627 .[
meillo@104	1628 paul vixie edginess nmh-workers
meillo@104	1629 .]
meillo@104	1630 .QS
meillo@104	1631 let's stop walking on egg shells with this code base. there's no need to
meillo@104	1632 discuss whether to keep using vfork, just note in [sic!] passing, [...]
meillo@104	1633 we don't need a separate branch for removing vmh
meillo@104	1634 or ridding ourselves of #ifdef's or removing posix replacement functions
meillo@164	1635 or depending on pure ansi/posix ``libc''.
meillo@104	1636 .QP
meillo@164	1637 these things should each be a day or two of work and the ``main branch''
meillo@104	1638 should just be modern. [...]
meillo@104	1639 let's push forward, aggressively.
meillo@104	1640 .QE
meillo@104	1641 .LP
meillo@104	1642 I did so already in the months before.
meillo@104	1643 I pushed forward.
meillo@159	1644 .\" XXX semicolon ?
meillo@104	1645 I simply dropped the cruft.
meillo@104	1646 .P
meillo@104	1647 The decision to drop a feature was based on literature research and
meillo@159	1648 careful thinking, but whether having had contact with this particular
meillo@104	1649 feature within my own computer life served as a rule of thumb.
meillo@159	1650 I explained my reasons in the commit messages
meillo@109	1651 in the version control system.
meillo@104	1652 Hence, others can comprehend my view and argue for undoing the change
meillo@104	1653 if I have missed an important aspect.
meillo@109	1654 I was quick in dropping parts.
meillo@179	1655 I rather include falsely dropped parts again, than going at a slower pace.
meillo@179	1656 Mmh is experimental work; it requires tough decisions.
meillo@159	1657 .\" XXX ``exp. work'' schon oft gesagt
meillo@12	1658
meillo@102	1659
meillo@217	1660 .U3 "Process Forking
meillo@12	1661 .P
meillo@109	1662 Being a tool chest, MH creates many processes.
meillo@104	1663 In earlier times
meillo@104	1664 .Fu fork()
meillo@104	1665 had been an expensive system call, because the process's image needed
meillo@159	1666 to be completely duplicated at once.
meillo@200	1667 This expensive work was especially unnecessary in the commonly occurring
meillo@159	1668 case wherein the image is replaced by a call to
meillo@104	1669 .Fu exec()
meillo@104	1670 right after having forked the child process.
meillo@104	1671 The
meillo@104	1672 .Fu vfork()
meillo@104	1673 system call was invented to speed up this particular case.
meillo@104	1674 It completely omits the duplication of the image.
meillo@104	1675 On old systems this resulted in significant speed ups.
meillo@104	1676 Therefore MH used
meillo@104	1677 .Fu vfork()
meillo@104	1678 whenever possible.
meillo@12	1679 .P
meillo@104	1680 Modern memory management units support copy-on-write semantics, which make
meillo@104	1681 .Fu fork()
meillo@104	1682 almost as fast as
meillo@104	1683 .Fu vfork() .
meillo@104	1684 The man page of
meillo@104	1685 .Mp vfork (2)
meillo@104	1686 in FreeBSD 8.0 states:
meillo@104	1687 .QS
meillo@104	1688 This system call will be eliminated when proper system sharing mechanisms
meillo@104	1689 are implemented. Users should not depend on the memory sharing semantics
meillo@104	1690 of vfork() as it will, in that case, be made synonymous to fork(2).
meillo@104	1691 .QE
meillo@104	1692 .LP
meillo@104	1693 Vixie supports the removal with the note that ``the last
meillo@104	1694 system on which fork was so slow that an mh user would notice it, was
meillo@104	1695 Eunice. that was 1987''.
meillo@104	1696 .[
meillo@104	1697 nmh-workers vixie edginess
meillo@104	1698 .]
meillo@104	1699 I replaced all calls to
meillo@104	1700 .Fu vfork()
meillo@104	1701 with calls to
meillo@217	1702 .Fu fork()
meillo@217	1703 .Ci 40821f5c1316e9205a08375e7075909cc9968e7d .
meillo@104	1704 .P
meillo@104	1705 Related to the costs of
meillo@104	1706 .Fu fork()
meillo@104	1707 is the probability of its success.
meillo@109	1708 In the eighties, on heavy loaded systems, calls to
meillo@104	1709 .Fu fork()
meillo@104	1710 were prone to failure.
meillo@104	1711 Hence, many of the
meillo@104	1712 .Fu fork()
meillo@104	1713 calls in the code were wrapped into loops to retry the
meillo@104	1714 .Fu fork()
meillo@217	1715 several times, to increase the chances to succeed eventually.
meillo@109	1716 On modern systems, a failing
meillo@104	1717 .Fu fork()
meillo@109	1718 call is unusual.
meillo@104	1719 Hence, in the rare case when
meillo@104	1720 .Fu fork()
meillo@217	1721 fails, mmh programs simply abort
meillo@217	1722 .Ci 5fbf37ee68e018998ada61eeab73e035b26834b6 .
meillo@12	1723
meillo@12	1724
meillo@109	1725 .U3 "Header Fields
meillo@104	1726 .BU
meillo@84	1727 The
meillo@84	1728 .Hd Encrypted
meillo@104	1729 header field was introduced by RFC\\|822,
meillo@109	1730 but already marked as legacy in RFC\\|2822.
meillo@109	1731 Today, OpenPGP provides the basis for standardized exchange of encrypted
meillo@104	1732 messages [RFC\\|4880, RFC\\|3156].
meillo@109	1733 Hence, the support for
meillo@104	1734 .Hd Encrypted
meillo@217	1735 header fields is removed in mmh
meillo@217	1736 .Ci 064527f7b57ab050e5af13e15ad99aeeab125857 .
meillo@104	1737 .BU
meillo@159	1738 The native support for
meillo@84	1739 .Hd Face
meillo@217	1740 header fields has been removed, as well
meillo@217	1741 .Ci 8e5be81f784682822f5e868c1bf3c8624682bd23 .
meillo@104	1742 This feature is similar to the
meillo@84	1743 .Hd X-Face
meillo@84	1744 header field in its intent,
meillo@21	1745 but takes a different approach to store the image.
meillo@84	1746 Instead of encoding the image data directly into the header field,
meillo@109	1747 it contains the hostname and UDP port where the image
meillo@109	1748 date can be retrieved.
meillo@159	1749 There is even a third Face system,
meillo@109	1750 which is the successor of
meillo@109	1751 .Hd X-Face ,
meillo@109	1752 although it re-uses the
meillo@104	1753 .Hd Face
meillo@217	1754 header field name.
meillo@109	1755 It was invented in 2005 and supports colored PNG images.
meillo@104	1756 None of the Face systems described here is popular today.
meillo@104	1757 Hence, mmh has no direct support for them.
meillo@104	1758 .BU
meillo@154	1759 .Id content-md5
meillo@104	1760 The
meillo@104	1761 .Hd Content-MD5
meillo@104	1762 header field was introduced by RFC\\|1864.
meillo@104	1763 It provides detection of data corruption during the transfer.
meillo@104	1764 But it can not ensure verbatim end-to-end delivery of the contents
meillo@104	1765 [RFC\\|1864].
meillo@104	1766 The proper approach to verify content integrity in an
meillo@166	1767 end-to-end relationship is the use of digital signatures.
meillo@104	1768 .\" XXX (RFCs FIXME).
meillo@104	1769 On the other hand, transfer protocols should detect corruption during
meillo@109	1770 the transmission.
meillo@109	1771 The TCP includes a checksum field therefore.
meillo@104	1772 These two approaches in combinations render the
meillo@104	1773 .Hd Content-MD5
meillo@104	1774 header field superfluous.
meillo@109	1775 Not a single one out of 4\\|200 messages from two decades
meillo@217	1776 in the nmh-workers mailing list archive
meillo@217	1777 .[
meillo@217	1778 nmh-workers mailing list archive website
meillo@217	1779 .]
meillo@217	1780 contains a
meillo@104	1781 .Hd Content-MD5
meillo@104	1782 header field.
meillo@104	1783 Neither did any of the 60\\|000 messages in my personal mail storage.
meillo@217	1784 Removing the support for this header field
meillo@217	1785 .Ci 31dc797eb5178970d68962ca8939da3fd9a8efda ,
meillo@104	1786 removed the last place where MD5 computation was needed.
meillo@104	1787 Hence, the MD5 code could be removed as well.
meillo@104	1788 Over 500 lines of code vanished by this one change.
meillo@104	1789
meillo@104	1790
meillo@104	1791 .U3 "MMDF maildrop support
meillo@21	1792 .P
meillo@217	1793 This type of maildrop format is conceptionally similar to the mbox format,
meillo@139	1794 but uses a different message delimiter (`\fL\\1\\1\\1\\1\fP',
meillo@139	1795 commonly written as `\fL^A^A^A^A\fP', instead of `\fLFrom\0\fP').
meillo@104	1796 Mbox is the de-facto standard maildrop format on Unix,
meillo@159	1797 whereas the MMDF maildrop format is now forgotten.
meillo@217	1798 Mbox remains as the only packed mailbox format, supported in mmh.
meillo@104	1799 .P
meillo@109	1800 The simplifications within the code were moderate.
meillo@109	1801 Mainly, the reading and writing of MMDF mailbox files was removed.
meillo@109	1802 But also, switches of
meillo@109	1803 .Pn packf
meillo@104	1804 and
meillo@109	1805 .Pn rcvpack
meillo@217	1806 could be removed
meillo@217	1807 .Ci 3916ab66ad5d183705ac12357621ea8661afd3c0 .
meillo@109	1808 In the message parsing function
meillo@109	1809 .Fn sbr/m_getfld.c ,
meillo@217	1810 knowledge of MMDF packed mail boxes was removed
meillo@217	1811 .Ci 684ec30d81e1223a282764452f4902ed4ad1c754 .
meillo@109	1812 Further code structure simplifications may be possible there,
meillo@109	1813 because only one single packed mailbox format is left to be supported.
meillo@104	1814 I have not worked on them yet because
meillo@104	1815 .Fu m_getfld()
meillo@104	1816 is heavily optimized and thus dangerous to touch.
meillo@104	1817 The risk of damaging the intricate workings of the optimized code is
meillo@104	1818 too high.
meillo@104	1819
meillo@12	1820
meillo@101	1821 .U3 "Prompter's Control Keys
meillo@20	1822 .P
meillo@20	1823 The program
meillo@20	1824 .Pn prompter
meillo@104	1825 queries the user to fill in a message form.
meillo@217	1826 When used as
meillo@104	1827 .Cl "comp -editor prompter" ,
meillo@20	1828 the resulting behavior is similar to
meillo@20	1829 .Pn mailx .
meillo@51	1830 Apparently,
meillo@20	1831 .Pn prompter
meillo@173	1832 had not been touched lately.
meillo@104	1833 Otherwise it's hardly explainable why it
meillo@20	1834 still offered the switches
meillo@84	1835 .Sw -erase
meillo@84	1836 .Ar chr
meillo@20	1837 and
meillo@84	1838 .Sw -kill
meillo@84	1839 .Ar chr
meillo@20	1840 to name the characters for command line editing.
meillo@21	1841 The times when this had been necessary are long time gone.
meillo@20	1842 Today these things work out-of-the-box, and if not, are configured
meillo@20	1843 with the standard tool
meillo@20	1844 .Pn stty .
meillo@104	1845 The switches are removed now
meillo@104	1846 .Ci 0bd9750710cdbab80cfb4036dd87af20afe1552f .
meillo@20	1847
meillo@104	1848
meillo@109	1849 .U3 "Hardcopy Terminal Support
meillo@21	1850 .P
meillo@109	1851 More of a funny anecdote is a check for being connected to a
meillo@109	1852 hardcopy terminal.
meillo@159	1853 It remained in the code until spring 2012, when I finally removed it
meillo@104	1854 .Ci b7764c4a6b71d37918a97594d866258f154017ca .
meillo@21	1855 .P
meillo@109	1856 The check only prevented a pager to be placed between the printing
meillo@104	1857 program (\c
meillo@104	1858 .Pn mhl )
meillo@104	1859 and the terminal.
meillo@109	1860 In nmh, this could have been ensured statically with the
meillo@104	1861 .Sw -nomoreproc
meillo@109	1862 at the command line, too.
meillo@121	1863 In mmh, setting the profile entry
meillo@104	1864 .Pe Pager
meillo@104	1865 or the environment variable
meillo@104	1866 .Ev PAGER
meillo@104	1867 to
meillo@109	1868 .Pn cat
meillo@159	1869 is sufficient.
meillo@104	1870
meillo@104	1871
meillo@21	1872
meillo@12	1873
meillo@58	1874 .H2 "Attachments
meillo@22	1875 .P
meillo@101	1876 The mind model of email attachments is unrelated to MIME.
meillo@217	1877 Although the MIME RFCs [RFC\\|2045\(en2049] define the technical
meillo@217	1878 requirements for having attachments, they do not mention the term.
meillo@101	1879 Instead of attachments, MIME talks about ``multi-part message bodies''
meillo@101	1880 [RFC\\|2045], a more general concept.
meillo@101	1881 Multi-part messages are messages
meillo@101	1882 ``in which one or more different
meillo@101	1883 sets of data are combined in a single body''
meillo@101	1884 [RFC\\|2046].
meillo@101	1885 MIME keeps its descriptions generic;
meillo@101	1886 it does not imply specific usage models.
meillo@217	1887 Today, one usage model is prevalent: attachments.
meillo@101	1888 The idea is having a main text document with files of arbitrary kind
meillo@101	1889 attached to it.
meillo@101	1890 In MIME terms, this is a multi-part message having a text part first
meillo@110	1891 and parts of arbitrary type following.
meillo@101	1892 .P
meillo@101	1893 MH's MIME support is a direct implementation of the RFCs.
meillo@101	1894 The perception of the topic described in the RFCs is clearly visible
meillo@101	1895 in MH's implementation.
meillo@159	1896 .\" XXX rewrite ``no idea''.
meillo@159	1897 As a result,
meillo@159	1898 MH had all the MIME features but no idea of attachments.
meillo@173	1899 But users do not need all the MIME features,
meillo@109	1900 they want convenient attachment handling.
meillo@109	1901
meillo@102	1902
meillo@102	1903 .U3 "Composing MIME Messages
meillo@102	1904 .P
meillo@102	1905 In order to improve the situation on the message composing side,
meillo@217	1906 Jon Steinhart had added an attachment system to nmh in 2002
meillo@217	1907 .Ci 7480dbc14bc90f2d872d434205c0784704213252 .
meillo@102	1908 In the file
meillo@102	1909 .Fn docs/README-ATTACHMENTS ,
meillo@217	1910 he described his motivation to do so:
meillo@101	1911 .QS
meillo@159	1912 Although nmh contains the necessary functionality for MIME message
meillo@159	1913 handing [sic!], the interface to this functionality is pretty obtuse.
meillo@101	1914 There's no way that I'm ever going to convince my partner to write
meillo@101	1915 .Pn mhbuild
meillo@101	1916 composition files!
meillo@101	1917 .QE
meillo@102	1918 .LP
meillo@102	1919 With this change, the mind model of attachments entered nmh.
meillo@102	1920 In the same document:
meillo@101	1921 .QS
meillo@101	1922 These changes simplify the task of managing attachments on draft files.
meillo@101	1923 They allow attachments to be added, listed, and deleted.
meillo@101	1924 MIME messages are automatically created when drafts with attachments
meillo@101	1925 are sent.
meillo@101	1926 .QE
meillo@102	1927 .LP
meillo@217	1928 Unfortunately, the attachment system, like every new facilities in nmh,
meillo@110	1929 was inactive by default.
meillo@101	1930 .P
meillo@217	1931 During my time in Argentina, I tried to improve the attachment system.
meillo@217	1932 But, after long discussions my patch died as a proposal on the
meillo@217	1933 mailing list because of great opposition in the nmh community.
meillo@101	1934 .[
meillo@101	1935 nmh-workers attachment proposal
meillo@101	1936 .]
meillo@217	1937 In January 2012, I extended the patch and applied it to mmh
meillo@217	1938 .Ci 8ff284ff9167eff8f5349481529332d59ed913b1 .
meillo@102	1939 In mmh, the attachment system is active by default.
meillo@102	1940 Instead of command line switches, the
meillo@102	1941 .Pe Attachment-Header
meillo@102	1942 profile entry is used to specify
meillo@102	1943 the name of the attachment header field.
meillo@102	1944 It is pre-defined to
meillo@102	1945 .Hd Attach .
meillo@101	1946 .P
meillo@159	1947 To add an attachment to a draft, a header line needs to be added:
meillo@101	1948 .VS
meillo@101	1949 To: bob
meillo@101	1950 Subject: The file you wanted
meillo@101	1951 Attach: /path/to/the/file-bob-wanted
meillo@101	1952 --------
meillo@101	1953 Here it is.
meillo@101	1954 VE
meillo@101	1955 The header field can be added to the draft manually in the editor,
meillo@102	1956 or by using the `attach' command at the WhatNow prompt, or
meillo@102	1957 non-interactively with
meillo@101	1958 .Pn anno :
meillo@101	1959 .VS
meillo@102	1960 anno -append -nodate -component Attach -text /path/to/attachment
meillo@101	1961 VE
meillo@102	1962 Drafts with attachment headers are converted to MIME automatically by
meillo@102	1963 .Pn send .
meillo@102	1964 The conversion to MIME is invisible to the user.
meillo@159	1965 The draft stored in the draft folder is always in source form with
meillo@101	1966 attachment headers.
meillo@179	1967 If the MIMEification fails (e.g. because the file to attach
meillo@179	1968 is not accessible) the original draft is not changed.
meillo@101	1969 .P
meillo@102	1970 The attachment system handles the forwarding of messages, too.
meillo@173	1971 If the attachment header value starts with a plus character (`\fL+\fP'),
meillo@101	1972 like in
meillo@101	1973 .Cl "Attach: +bob 30 42" ,
meillo@159	1974 the given messages in the specified folder will be attached.
meillo@101	1975 This allowed to simplify
meillo@217	1976 .Pn forw
meillo@217	1977 .Ci f41f04cf4ceca7355232cf7413e59afafccc9550 .
meillo@101	1978 .P
meillo@101	1979 Closely related to attachments is non-ASCII text content,
meillo@217	1980 because it requires MIME as well.
meillo@102	1981 In nmh, the user needed to call `mime' at the WhatNow prompt
meillo@101	1982 to have the draft converted to MIME.
meillo@102	1983 This was necessary whenever the draft contained non-ASCII characters.
meillo@101	1984 If the user did not call `mime', a broken message would be sent.
meillo@101	1985 Therefore, the
meillo@101	1986 .Pe automimeproc
meillo@101	1987 profile entry could be specified to have the `mime' command invoked
meillo@102	1988 automatically each time.
meillo@179	1989 Unfortunately, this approach conflicted with the attachment system
meillo@101	1990 because the draft would already be in MIME format at the time
meillo@101	1991 when the attachment system wanted to MIMEify it.
meillo@102	1992 To use nmh's attachment system, `mime' must not be called at the
meillo@102	1993 WhatNow prompt and
meillo@101	1994 .Pe automimeproc
meillo@102	1995 must not be set in the profile.
meillo@101	1996 But then the case of non-ASCII text without attachment headers was
meillo@101	1997 not caught.
meillo@102	1998 All in all, the solution was complex and irritating.
meillo@168	1999 My patch from December 2010
meillo@168	2000 .[
meillo@168	2001 nmh-workers attachment proposal
meillo@168	2002 .]
meillo@168	2003 would have simplified the situation.
meillo@102	2004 .P
meillo@101	2005 Mmh's current solution is even more elaborate.
meillo@101	2006 Any necessary MIMEification is done automatically.
meillo@101	2007 There is no `mime' command at the WhatNow prompt anymore.
meillo@102	2008 The draft will be converted automatically to MIME when either an
meillo@102	2009 attachment header or non-ASCII text is present.
meillo@173	2010 Furthermore, the hash character (`\fL#\fP') is not special any more
meillo@159	2011 at line beginnings in the draft message.
meillo@159	2012 .\" XXX REF ?
meillo@159	2013 Users need not concern themselves with the whole topic at all.
meillo@101	2014 .P
meillo@102	2015 Although the new approach does not anymore support arbitrary MIME
meillo@102	2016 compositions directly, the full power of
meillo@101	2017 .Pn mhbuild
meillo@101	2018 can still be accessed.
meillo@217	2019 Given no attachment headers are included, users can create
meillo@101	2020 .Pn mhbuild
meillo@102	2021 composition drafts like in nmh.
meillo@217	2022 Then, at the WhatNow prompt, they can invoke
meillo@101	2023 .Cl "edit mhbuild
meillo@217	2024 to convert the draft to MIME.
meillo@217	2025 Because the resulting draft neither contains non-ASCII characters
meillo@102	2026 nor has it attachment headers, the attachment system will not touch it.
meillo@101	2027 .P
meillo@159	2028 The approach taken in mmh is tailored towards today's most common case:
meillo@159	2029 a text part, possibly with attachments.
meillo@159	2030 This case was simplified.
meillo@102	2031
meillo@112	2032
meillo@102	2033 .U3 "MIME Type Guessing
meillo@102	2034 .P
meillo@159	2035 From the programmer's point of view, the use of
meillo@101	2036 .Pn mhbuild
meillo@159	2037 composition drafts had one notable advantage over attachment headers:
meillo@159	2038 The user provides the appropriate MIME types for files to include.
meillo@217	2039 The new attachment system needs to find out the correct MIME type itself.
meillo@217	2040 This is a difficult task.
meillo@102	2041 Determining the correct MIME type of content is partly mechanical,
meillo@102	2042 partly intelligent work.
meillo@102	2043 Forcing the user to find out the correct MIME type,
meillo@102	2044 forces him to do partly mechanical work.
meillo@179	2045 Letting the computer do the work can lead to bad choices for difficult
meillo@102	2046 content.
meillo@217	2047 For mmh, the latter option was chosen to spare the user the work
meillo@217	2048 .Ci 3baec236a39c5c89a9bda8dbd988d643a21decc6 .
meillo@102	2049 .P
meillo@102	2050 Determining the MIME type by the suffix of the file name is a dumb
meillo@102	2051 approach, yet it is simple to implement and provides good results
meillo@102	2052 for the common cases.
meillo@217	2053 If no MIME type can be determined, text content is sent as `text/plain',
meillo@217	2054 anything else under the generic fall-back type `application/octet-stream'.
meillo@102	2055 Mmh implements this approach in the
meillo@102	2056 .Pn print-mimetype
meillo@217	2057 script
meillo@217	2058 .Ci 4b5944268ea0da7bb30598a27857304758ea9b44 .
meillo@102	2059 .P
meillo@112	2060 A far better, though less portable, approach is the use of
meillo@102	2061 .Pn file .
meillo@102	2062 This standard tool tries to determine the type of files.
meillo@102	2063 Unfortunately, its capabilities and accuracy varies from system to system.
meillo@102	2064 Additionally, its output was only intended for human beings,
meillo@102	2065 but not to be used by programs.
meillo@102	2066 Nevertheless, modern versions of GNU
meillo@102	2067 .Pn file ,
meillo@217	2068 which are prevalent on the popular GNU/Linux systems,
meillo@159	2069 provide MIME type output in machine-readable form.
meillo@217	2070 Although this solution is system-dependent,
meillo@102	2071 it solves the difficult problem well.
meillo@102	2072 On systems where GNU
meillo@102	2073 .Pn file ,
meillo@102	2074 version 5.04 or higher, is available it should be used.
meillo@102	2075 One needs to specify the following profile entry to do so:
meillo@102	2076 .VS
meillo@102	2077 Mime-Type-Query: file -b --mime
meillo@102	2078 VE
meillo@102	2079 .LP
meillo@102	2080 Other versions of
meillo@102	2081 .Pn file
meillo@217	2082 might possibly be usable with wrapper scripts that reformat the output.
meillo@102	2083 The diversity among
meillo@102	2084 .Pn file
meillo@102	2085 implementations is great; one needs to check the local variant.
meillo@102	2086 .P
meillo@102	2087 It is not possible in mmh to override the automatic MIME type guessing
meillo@102	2088 for a specific file.
meillo@159	2089 To do so, either the user would need to know in advance for which file
meillo@217	2090 the automatic guessing fails or the system would require interaction.
meillo@102	2091 I consider both cases impractical.
meillo@102	2092 The existing solution should be sufficient.
meillo@102	2093 If not, the user may always fall back to
meillo@102	2094 .Pn mhbuild
meillo@217	2095 composition drafts and bypass the attachment system.
meillo@101	2096
meillo@102	2097
meillo@102	2098 .U3 "Storing Attachments
meillo@102	2099 .P
meillo@169	2100 Extracting MIME parts of a message and storing them to disk is performed by
meillo@108	2101 .Pn mhstore .
meillo@108	2102 The program has two operation modes,
meillo@108	2103 .Sw -auto
meillo@108	2104 and
meillo@108	2105 .Sw -noauto .
meillo@108	2106 With the former one, each part is stored under the filename given in the
meillo@108	2107 MIME part's meta information, if available.
meillo@108	2108 This naming information is usually available for modern attachments.
meillo@108	2109 If no filename is available, this MIME part is stored as if
meillo@108	2110 .Sw -noauto
meillo@108	2111 would have been specified.
meillo@108	2112 In the
meillo@108	2113 .Sw -noauto
meillo@108	2114 mode, the parts are processed according to rules, defined by
meillo@108	2115 .Pe mhstore-store-*
meillo@108	2116 profile entries.
meillo@108	2117 These rules define generic filename templates for storing
meillo@108	2118 or commands to post-process the contents in arbitrary ways.
meillo@108	2119 If no matching rule is available the part is stored under a generic
meillo@108	2120 filename, built from message number, MIME part number, and MIME type.
meillo@108	2121 .P
meillo@108	2122 The
meillo@108	2123 .Sw -noauto
meillo@108	2124 mode had been the default in nmh because it was considered safe,
meillo@108	2125 in contrast to the
meillo@108	2126 .Sw -auto
meillo@108	2127 mode.
meillo@108	2128 In mmh,
meillo@108	2129 .Sw -auto
meillo@108	2130 is not dangerous anymore.
meillo@108	2131 Two changes were necessary:
meillo@171	2132 .LI 1
meillo@108	2133 Any directory path is removed from the proposed filename.
meillo@108	2134 Thus, the files are always stored in the expected directory.
meillo@108	2135 .Ci 41b6eadbcecf63c9a66aa5e582011987494abefb
meillo@171	2136 .LI 2
meillo@108	2137 Tar files are not extracted automatically any more.
meillo@108	2138 Thus, the rest of the file system will not be touched.
meillo@108	2139 .Ci 94c80042eae3383c812d9552089953f9846b1bb6
meillo@217	2140 .P
meillo@217	2141 In mmh, the result of
meillo@108	2142 .Cl "mhstore -auto
meillo@110	2143 can be foreseen from the output of
meillo@108	2144 .Cl "mhlist -verbose" .
meillo@217	2145 Although the
meillo@108	2146 .Sw -noauto
meillo@217	2147 mode is considered to be more powerful, it is less convenient and
meillo@108	2148 .Sw -auto
meillo@217	2149 is safe now.
meillo@217	2150 Additionally, storing attachments under their original name
meillo@217	2151 is intuitive.
meillo@108	2152 Hence,
meillo@108	2153 .Sw -auto
meillo@217	2154 serves better as the default option
meillo@217	2155 .Ci 3410b680416c49a7617491af38bc1929855a331d .
meillo@108	2156 .P
meillo@108	2157 Files are stored into the directory given by the
meillo@108	2158 .Pe Nmh-Storage
meillo@108	2159 profile entry, if set, or
meillo@108	2160 into the current working directory, otherwise.
meillo@108	2161 Storing to different directories is only possible with
meillo@108	2162 .Pe mhstore-store-*
meillo@108	2163 profile entries.
meillo@108	2164 .P
meillo@217	2165 Still existing files get overwritten silently in both modes.
meillo@108	2166 This can be considered a bug.
meillo@108	2167 Yet, each other behavior has its draw-backs, too.
meillo@108	2168 Refusing to replace files requires adding a
meillo@108	2169 .Sw -force
meillo@217	2170 switch.
meillo@108	2171 Users will likely need to invoke
meillo@108	2172 .Pn mhstore
meillo@108	2173 a second time with
meillo@159	2174 .Sw -force .
meillo@159	2175 Eventually, only the user can decide in the specific case.
meillo@108	2176 This requires interaction, which I like to avoid if possible.
meillo@108	2177 Appending a unique suffix to the filename is another bad option.
meillo@108	2178 For now, the behavior remains as it is.
meillo@108	2179 .P
meillo@108	2180 In mmh, only MIME parts of type message are special in
meillo@108	2181 .Pn mhstore 's
meillo@108	2182 .Sw -auto
meillo@108	2183 mode.
meillo@108	2184 Instead of storing message/rfc822 parts as files to disk,
meillo@108	2185 they are stored as messages into the current mail folder.
meillo@159	2186 The same applies to message/partial, although the parts are
meillo@159	2187 automatically reassembled beforehand.
meillo@159	2188 MIME parts of type message/external-body are not automatically retrieved
meillo@159	2189 anymore.
meillo@159	2190 Instead, information on how to retrieve them is output.
meillo@217	2191 Not supporting this rare case saved nearly one thousand lines of code
meillo@217	2192 .Ci 55e1d8c654ee0f7c45b9361ce34617983b454c32 .
meillo@108	2193 .\" XXX mention somewhere else too: (The profile entry `nmh-access-ftp'
meillo@108	2194 .\" and sbr/ruserpass.c for reading ~/.netrc are gone now.)
meillo@217	2195 The MIME type `application/octet-stream; type=tar' is not special anymore.
meillo@217	2196 The automatically extracting of such MIME parts had been the
meillo@217	2197 dangerous part of the
meillo@108	2198 .Sw -auto
meillo@217	2199 mode
meillo@217	2200 .Ci 94c80042eae3383c812d9552089953f9846b1bb6 .
meillo@108	2201
meillo@102	2202
meillo@102	2203
meillo@102	2204 .U3 "Showing MIME Messages
meillo@102	2205 .P
meillo@114	2206 The program
meillo@114	2207 .Pn mhshow
meillo@217	2208 was written to display MIME messages.
meillo@114	2209 It implemented the conceptional view of the MIME RFCs.
meillo@114	2210 Nmh's
meillo@114	2211 .Pn mhshow
meillo@217	2212 handles each MIME part independently, presenting them separately
meillo@114	2213 to the user.
meillo@114	2214 This does not match today's understanding of email attachments,
meillo@114	2215 where displaying a message is seen to be a single, integrated operation.
meillo@114	2216 Today, email messages are expected to consist of a main text part
meillo@114	2217 plus possibly attachments.
meillo@217	2218 They are no more seen to be arbitrary MIME hierarchies with
meillo@114	2219 information on how to display the individual parts.
meillo@114	2220 I adjusted
meillo@114	2221 .Pn mhshow 's
meillo@114	2222 behavior to the modern view on the topic.
meillo@114	2223 .P
meillo@224	2224 One should note that this section completely ignores the original
meillo@114	2225 .Pn show
meillo@114	2226 program, because it was not capable to display MIME messages
meillo@114	2227 and is no longer part of mmh.
meillo@179	2228 .\" XXX ref to other section
meillo@114	2229 Although
meillo@114	2230 .Pn mhshow
meillo@114	2231 was renamed to
meillo@114	2232 .Pn show
meillo@114	2233 in mmh, this section uses the name
meillo@114	2234 .Pn mhshow ,
meillo@224	2235 in order to avoid confusion.
meillo@114	2236 .P
meillo@114	2237 In mmh, the basic idea is that
meillo@114	2238 .Pn mhshow
meillo@114	2239 should display a message in one single pager session.
meillo@114	2240 Therefore,
meillo@114	2241 .Pn mhshow
meillo@114	2242 invokes a pager session for all its output,
meillo@217	2243 whenever it prints to a terminal
meillo@217	2244 .Ci a4197ea6ffc5c1550e8b52d5a654bcaaaee04a4e .
meillo@114	2245 In consequence,
meillo@114	2246 .Pn mhl
meillo@217	2247 does no more invoke a pager
meillo@217	2248 .Ci 0e46503be3c855bddaeae3843e1b659279c35d70 .
meillo@114	2249 With
meillo@114	2250 .Pn mhshow
meillo@114	2251 replacing the original
meillo@114	2252 .Pn show ,
meillo@217	2253 the output of
meillo@114	2254 .Pn mhl
meillo@217	2255 no longer goes to the terminal directly, but through
meillo@114	2256 .Pn mhshow .
meillo@114	2257 Hence,
meillo@114	2258 .Pn mhl
meillo@114	2259 does not need to invoke a pager.
meillo@114	2260 The one and only job of
meillo@114	2261 .Pn mhl
meillo@114	2262 is to format messages or parts of them.
meillo@114	2263 The only place in mmh, where a pager is invoked is
meillo@114	2264 .Pn mhshow .
meillo@114	2265 .P
meillo@217	2266 In the intended setup, only text content is be displayed,
meillo@217	2267 in a single pager session.
meillo@217	2268 Non-text content needs to be converted to text by appropriate
meillo@217	2269 .Pe mhshow-show-*
meillo@217	2270 profile entries before, if this is possible and wanted.
meillo@217	2271 A common example for this are PDF files.
meillo@217	2272 .ig \"XXX
meillo@114	2273 .Pe mhshow-show-*
meillo@114	2274 profile entries can be used to display MIME parts in a specific way.
meillo@114	2275 to display them in the terminal.
meillo@217	2276 ..
meillo@217	2277 In mmh, MIME parts are always displayed serially.
meillo@114	2278 The request to display the MIME type `multipart/parallel' in parallel
meillo@114	2279 is ignored.
meillo@217	2280 It is simply treated as `multipart/mixed'
meillo@217	2281 .Ci d0581ba306a7299113a346f9b4c46ce97bc4cef6 .
meillo@217	2282 This was already possible to requested with the, now removed,
meillo@114	2283 .Sw -serialonly
meillo@114	2284 switch of
meillo@114	2285 .Pn mhshow .
meillo@179	2286 As MIME parts are always processed exclusively, i.e. serially,
meillo@217	2287 the `\fL%e\fP' escape in
meillo@114	2288 .Pe mhshow-show-*
meillo@217	2289 profile entries became useless and was thus removed
meillo@217	2290 .Ci a20d405db09b7ccca74d3e8c57550883da49e1ae .
meillo@114	2291 .P
meillo@114	2292 Other kinds of attachments are ignored.
meillo@114	2293 With
meillo@114	2294 .Pe mhshow-show-*
meillo@114	2295 profile entries for them, they can be displayed serially along
meillo@114	2296 the message.
meillo@114	2297 For parallel display, the attachments need to be stored to disk first.
meillo@114	2298 .P
meillo@114	2299 To display text content in foreign charsets, they need to be converted
meillo@114	2300 to the native charset.
meillo@114	2301 Therefore,
meillo@114	2302 .Pe mhshow-charset-*
meillo@217	2303 profile entries were needed.
meillo@169	2304 In mmh, the conversion is performed automatically by piping the
meillo@169	2305 text through the
meillo@114	2306 .Pn iconv
meillo@217	2307 command, if necessary
meillo@217	2308 .Ci 2433122c20baccb10b70b49c04c6b0497b5b3b60 .
meillo@114	2309 Custom
meillo@114	2310 .Pe mhshow-show-*
meillo@114	2311 rules for textual content might need a
meillo@114	2312 .Cl "iconv -f %c %f \|
meillo@114	2313 prefix to have the text converted to the native charset.
meillo@114	2314 .P
meillo@121	2315 Although the conversion of foreign charsets to the native one
meillo@114	2316 has improved, it is not consistent enough.
meillo@114	2317 Further work needs to be done and
meillo@114	2318 the basic concepts in this field need to be re-thought.
meillo@114	2319 Though, the default setup of mmh displays message in foreign charsets
meillo@114	2320 correctly without the need to configure anything.
meillo@114	2321
meillo@114	2322
meillo@114	2323 .ig
meillo@114	2324
meillo@114	2325 .P
meillo@114	2326 mhshow/mhstore: Removed support for retrieving message/external-body parts.
meillo@173	2327 These tools will not download the contents automatically anymore. Instead,
meillo@114	2328 they print the information needed to get the contents. If someone should
meillo@114	2329 really receive one of those rare message/external-body messages, he can
meillo@114	2330 do the job manually. We save nearly a thousand lines of code. That's worth
meillo@114	2331 it!
meillo@114	2332 (The profile entry `nmh-access-ftp' and sbr/ruserpass.c for reading
meillo@114	2333 ~/.netrc are gone now.)
meillo@114	2334 .Ci 55e1d8c654ee0f7c45b9361ce34617983b454c32
meillo@114	2335
meillo@114	2336 ..
meillo@102	2337
meillo@58	2338
meillo@58	2339
meillo@166	2340 .H2 "Signing and Encrypting
meillo@22	2341 .P
meillo@166	2342 Nmh offers no direct support for digital signatures and message encryption.
meillo@157	2343 This functionality needed to be added through third-party software.
meillo@217	2344 In mmh, the functionality is included because it
meillo@217	2345 is a part of modern email and is likely wanted by users of mmh.
meillo@217	2346 A fresh mmh installation supports signing and encrypting
meillo@157	2347 out-of-the-box.
meillo@157	2348 Therefore, Neil Rickert's
meillo@157	2349 .Pn mhsign
meillo@157	2350 and
meillo@157	2351 .Pn mhpgp
meillo@157	2352 scripts
meillo@157	2353 .[
meillo@157	2354 neil rickert mhsign mhpgp
meillo@157	2355 .]
meillo@217	2356 were included
meillo@177	2357 .Ci f45cdc98117a84f071759462c7ae212f4bc5ab2e
meillo@177	2358 .Ci 58cf09aa36e9f7f352a127158bbf1c5678bc6ed8 .
meillo@177	2359 The scripts fit well because they are lightweight and
meillo@177	2360 similar of style to the existing tools.
meillo@217	2361 Additionally, no licensing difficulties appeared
meillo@157	2362 as they are part of the public domain.
meillo@112	2363 .P
meillo@157	2364 .Pn mhsign
meillo@157	2365 handles the signing and encrypting part.
meillo@157	2366 It comprises about 250 lines of shell code and interfaces between
meillo@223	2367 .Pn gnupg
meillo@217	2368 and the MH system.
meillo@177	2369 It was meant to be invoked manually at the WhatNow prompt, but in mmh,
meillo@157	2370 .Pn send
meillo@177	2371 invokes
meillo@217	2372 .Pn mhsign
meillo@177	2373 automatically
meillo@177	2374 .Ci c7b5e1df086bcc37ff40163ee67571f076cf6683 .
meillo@177	2375 Special header fields were introduced to request this action.
meillo@157	2376 If a draft contains the
meillo@157	2377 .Hd Sign
meillo@157	2378 header field,
meillo@157	2379 .Pn send
meillo@177	2380 will initiate the signing.
meillo@217	2381 The signing key is either chosen automatically or it is specified by the
meillo@157	2382 .Pe Pgpkey
meillo@157	2383 profile entry.
meillo@157	2384 .Pn send
meillo@217	2385 always create signatures using the PGP/MIME standard [RFC\\|4880],
meillo@217	2386 but by invoking
meillo@217	2387 .Pn mhsign
meillo@217	2388 manually, old-style non-MIME signatures can be created as well.
meillo@177	2389 To encrypt an outgoing message, the draft needs to contain an
meillo@157	2390 .Hd Enc
meillo@157	2391 header field.
meillo@177	2392 Public keys of all recipients are searched for in the gnupg keyring and
meillo@177	2393 in a file called
meillo@177	2394 .Fn pgpkeys ,
meillo@177	2395 which contains exceptions and overrides.
meillo@157	2396 Unless public keys are found for all recipients,
meillo@177	2397 .Pn mhsign
meillo@177	2398 will refuse to encrypt it.
meillo@157	2399 Currently, messages with hidden (BCC) recipients can not be encrypted.
meillo@171	2400 This work is pending because it requires a structurally more complex
meillo@171	2401 approach.
meillo@157	2402 .P
meillo@177	2403 .Pn mhpgp
meillo@177	2404 is the companion to
meillo@177	2405 .Pn mhsign .
meillo@177	2406 It verifies signatures and decrypts messages.
meillo@217	2407 Encrypted messages can be either temporarily decrypted and displayed
meillo@177	2408 or permanently decrypted and stored into the current folder.
meillo@177	2409 Currently,
meillo@177	2410 .Pn mhpgp
meillo@177	2411 needs to be invoked manually.
meillo@177	2412 The integration into
meillo@177	2413 .Pn show
meillo@177	2414 and
meillo@177	2415 .Pn mhstore
meillo@217	2416 to verify signatures and decrypt messages as needed
meillo@217	2417 is planned but not yet realized.
meillo@177	2418 .P
meillo@217	2419 Both scripts were written for nmh.
meillo@217	2420 Hence they needed to be adjust
meillo@177	2421 according to the differences between nmh and mmh.
meillo@177	2422 For instance, they use the backup prefix no longer.
meillo@181	2423 Furthermore, compatibility support for old PGP features was dropped.
meillo@177	2424 .P
meillo@157	2425 The integrated message signing and encrypting support is one of the
meillo@157	2426 most recent features in mmh.
meillo@217	2427 It has not had the time to mature.
meillo@177	2428 User feedback and personal experience need to be accumulated to
meillo@177	2429 direct the further development of the facility.
meillo@217	2430 Already it seems to be worthwhile to consider adding
meillo@157	2431 .Sw -[no]sign
meillo@157	2432 and
meillo@157	2433 .Sw -[no]enc
meillo@157	2434 switches to
meillo@157	2435 .Pn send ,
meillo@177	2436 to be able to override the corresponding header fields.
meillo@177	2437 A profile entry:
meillo@157	2438 .VS
meillo@157	2439 send: -sign
meillo@157	2440 VE
meillo@177	2441 would then activate signing for all outgoing messages.
meillo@177	2442 With the present approach, a
meillo@177	2443 .Hd Send
meillo@177	2444 header component needs to be added to each draft template
meillo@177	2445 to achieve the same result.
meillo@177	2446 Adding the switches would ease the work greatly and keep the
meillo@177	2447 template files clean.
meillo@157	2448
meillo@58	2449
meillo@58	2450
meillo@102	2451
meillo@133	2452 .H2 "Draft and Trash Folder
meillo@131	2453 .P
meillo@58	2454
meillo@131	2455 .U3 "Draft Folder
meillo@154	2456 .Id draft-folder
meillo@131	2457 .P
meillo@131	2458 In the beginning, MH had the concept of a draft message.
meillo@217	2459 This was a file named
meillo@131	2460 .Fn draft
meillo@217	2461 in the MH directory, which was treated special.
meillo@131	2462 On composing a message, this draft file was used.
meillo@131	2463 When starting to compose another message before the former one was sent,
meillo@131	2464 the user had to decide among:
meillo@171	2465 .LI 1
meillo@168	2466 Using the old draft to finish and send it before starting with a new one.
meillo@171	2467 .LI 2
meillo@168	2468 Discarding the old draft and replacing it with a new one.
meillo@171	2469 .LI 3
meillo@168	2470 Preserving the old draft by refiling it to a folder.
meillo@171	2471 .LP
meillo@217	2472 Working on multiple drafts was only possible in alternation.
meillo@217	2473 For that, the current draft needed to be refiled to a folder and
meillo@168	2474 another one re-used for editing.
meillo@131	2475 Working on multiple drafts at the same time was impossible.
meillo@131	2476 The usual approach of switching to a different MH context did not
meillo@168	2477 help anything.
meillo@131	2478 .P
meillo@131	2479 The draft folder facility exists to
meillo@131	2480 allow true parallel editing of drafts, in a straight forward way.
meillo@131	2481 It was introduced by Marshall T. Rose, already in 1984.
meillo@131	2482 Similar to other new features, the draft folder was inactive by default.
meillo@131	2483 Even in nmh, the highly useful draft folder was not available
meillo@131	2484 out-of-the-box.
meillo@131	2485 At least, Richard Coleman added the man page
meillo@131	2486 .Mp mh-draft (5)
meillo@131	2487 to better document the feature.
meillo@131	2488 .P
meillo@131	2489 Not using the draft folder facility has the single advantage of having
meillo@131	2490 the draft file at a static location.
meillo@131	2491 This is simple in simple cases but the concept does not scale for more
meillo@131	2492 complex cases.
meillo@217	2493 The concept of the draft message is too limited for the problem
meillo@217	2494 it tries to solve.
meillo@131	2495 Therefore the draft folder was introduced.
meillo@131	2496 It is the more powerful and more natural concept.
meillo@131	2497 The draft folder is a folder like any other folder in MH.
meillo@131	2498 Its messages can be listed like any other messages.
meillo@131	2499 A draft message is no longer a special case.
meillo@131	2500 Tools do not need special switches to work on the draft message.
meillo@171	2501 Hence corner cases were removed.
meillo@131	2502 .P
meillo@131	2503 The trivial part of the work was activating the draft folder with a
meillo@131	2504 default name.
meillo@131	2505 I chose the name
meillo@217	2506 .Fn +drafts ,
meillo@131	2507 for obvious reasons.
meillo@131	2508 In consequence, the command line switches
meillo@131	2509 .Sw -draftfolder
meillo@131	2510 and
meillo@131	2511 .Sw -draftmessage
meillo@131	2512 could be removed.
meillo@131	2513 More difficult but also more improving was updating the tools to the
meillo@131	2514 new concept.
meillo@131	2515 For nearly three decades, the tools needed to support two draft handling
meillo@131	2516 approaches.
meillo@217	2517 By fully switching to the draft folder, the tools could be
meillo@217	2518 simplified by dropping the awkward draft message handling code.
meillo@131	2519 .Sw -draft
meillo@131	2520 switches were removed because operating on a draft message is no longer
meillo@131	2521 special.
meillo@131	2522 It became indistinguishable to operating on any other message.
meillo@168	2523 .Ci 337338b404931f06f0db2119c9e145e8ca5a9860
meillo@168	2524 .P
meillo@168	2525 There is no more need to query the user for draft handling
meillo@168	2526 .Ci 2d48b455c303a807041c35e4248955f8bec59eeb .
meillo@131	2527 It is always possible to add another new draft.
meillo@131	2528 Refiling drafts is without difference to refiling other messages.
meillo@168	2529 All of these special cases are gone.
meillo@131	2530 Yet, one draft-related switch remained.
meillo@131	2531 .Pn comp
meillo@131	2532 still has
meillo@131	2533 .Sw -[no]use
meillo@131	2534 for switching between two modes:
meillo@171	2535 .LI 1
meillo@217	2536 Modifying an existing draft, with
meillo@217	2537 .Sw -use .
meillo@171	2538 .LI 2
meillo@217	2539 Composing a new draft, possibly taking some existing message as template,
meillo@217	2540 with
meillo@217	2541 .Sw -nouse ,
meillo@217	2542 the default.
meillo@171	2543 .LP
meillo@131	2544 In either case, the behavior of
meillo@131	2545 .Pn comp
meillo@131	2546 is deterministic.
meillo@131	2547 .P
meillo@131	2548 .Pn send
meillo@131	2549 now operates on the current message in the draft folder by default.
meillo@131	2550 As message and folder can both be overridden by specifying them on
meillo@131	2551 the command line, it is possible to send any message in the mail storage
meillo@131	2552 by simply specifying its number and folder.
meillo@131	2553 In contrast to the other tools,
meillo@131	2554 .Pn send
meillo@131	2555 takes the draft folder as its default folder.
meillo@131	2556 .P
meillo@131	2557 Dropping the draft message concept in favor for the draft folder concept,
meillo@217	2558 replaced special cases with regular cases.
meillo@131	2559 This simplified the source code of the tools, as well as the concepts.
meillo@131	2560 In mmh, draft management does not break with the MH concepts
meillo@131	2561 but applies them.
meillo@133	2562 .Cl "scan +drafts" ,
meillo@133	2563 for instance, is a truly natural request.
meillo@217	2564 .P
meillo@169	2565 Most of the work was already performed by Rose in the eighties.
meillo@133	2566 The original improvement of mmh is dropping the old draft message approach
meillo@217	2567 and thus simplifying the tools, the documentation,
meillo@217	2568 and the system as a whole.
meillo@131	2569 Although my part in the draft handling improvement was small,
meillo@217	2570 it was important.
meillo@131	2571
meillo@131	2572
meillo@131	2573 .U3 "Trash Folder
meillo@154	2574 .Id trash-folder
meillo@131	2575 .P
meillo@131	2576 Similar to the situation for drafts is the situation for removed messages.
meillo@131	2577 Historically, a message was ``deleted'' by prepending a specific
meillo@173	2578 \fIbackup prefix\fP, usually the comma character,
meillo@173	2579 to the file name.
meillo@164	2580 The specific file would then be ignored by MH because only files with
meillo@164	2581 names consisting of digits only are treated as messages.
meillo@131	2582 Although files remained in the file system,
meillo@168	2583 the messages were no longer visible in MH.
meillo@168	2584 To truly delete them, a maintenance job was needed.
meillo@168	2585 Usually a cron job was installed to delete them after a grace time.
meillo@131	2586 For instance:
meillo@131	2587 .VS
meillo@131	2588 find $HOME/Mail -type f -name ',*' -ctime +7 -delete
meillo@131	2589 VE
meillo@168	2590 In such a setup, the original message could be restored
meillo@131	2591 within the grace time interval by stripping the
meillo@217	2592 backup prefix from the file name \(en usually but not always.
meillo@168	2593 If the last message of a folder with six messages (\fL1-6\fP) was removed,
meillo@131	2594 message
meillo@131	2595 .Fn 6 ,
meillo@168	2596 became file
meillo@131	2597 .Fn ,6 .
meillo@168	2598 If then a new message entered the same folder, it would be named with
meillo@168	2599 the number one above the highest existing message number.
meillo@168	2600 In this case the message would be named
meillo@217	2601 .Fn 6 ,
meillo@217	2602 reusing the number.
meillo@168	2603 If this new message would be removed as well,
meillo@217	2604 then the backup of the former message becomes overwritten.
meillo@168	2605 Hence, the ability to restore removed messages did not only depend on
meillo@181	2606 the sweeping cron job but also on the removing of further messages.
meillo@131	2607 It is undesirable to have such obscure and complex mechanisms.
meillo@168	2608 The user should be given a small set of clear assertions, such as
meillo@131	2609 ``Removed files are restorable within a seven-day grace time.''
meillo@131	2610 With the addition ``... unless a message with the same name in the
meillo@131	2611 same folder is removed before.'' the statement becomes complex.
meillo@217	2612 A user will hardly be able to keep track of all removals to know
meillo@131	2613 if the assertion still holds true for a specific file.
meillo@164	2614 In practice, the real mechanism is unclear to the user.
meillo@131	2615 .P
meillo@217	2616 Furthermore, the backup files were scattered within the whole mail storage.
meillo@217	2617 This complicated managing them.
meillo@217	2618 It was possible with the help of
meillo@131	2619 .Pn find ,
meillo@217	2620 but everything is more convenient
meillo@217	2621 if the deleted messages are collected in one place.
meillo@131	2622 .P
meillo@131	2623 The profile entry
meillo@131	2624 .Pe rmmproc
meillo@131	2625 (previously named
meillo@131	2626 .Pe Delete-Prog )
meillo@131	2627 was introduced very early to improve the situation.
meillo@164	2628 It could be set to any command, which would be executed to remove
meillo@131	2629 the specified messages.
meillo@217	2630 This had overridden the default action, described above.
meillo@217	2631 Refiling the to-be-removed files to a trash folder was the usual example.
meillo@131	2632 Nmh's man page
meillo@131	2633 .Mp rmm (1)
meillo@131	2634 proposes to set the
meillo@131	2635 .Pe rmmproc
meillo@131	2636 to
meillo@131	2637 .Cl "refile +d
meillo@217	2638 to move messages to the trash folder
meillo@217	2639 .Fn +d
meillo@131	2640 instead of renaming them with the backup prefix.
meillo@217	2641 The man page additionally proposes the expunge command
meillo@131	2642 .Cl "rm `mhpath +d all`
meillo@168	2643 to empty the trash folder.
meillo@131	2644 .P
meillo@217	2645 Removing messages in such a way has advantages:
meillo@217	2646 .LI 1
meillo@131	2647 The mail storage is prevented from being cluttered with removed messages
meillo@131	2648 because they are all collected in one place.
meillo@131	2649 Existing and removed messages are thus separated more strictly.
meillo@217	2650 .LI 2
meillo@131	2651 No backup files are silently overwritten.
meillo@217	2652 .LI 3
meillo@217	2653 Most important, however, removed messages are kept in the MH domain.
meillo@131	2654 Messages in the trash folder can be listed like those in any other folder.
meillo@131	2655 Deleted messages can be displayed like any other messages.
meillo@169	2656 .Pn refile
meillo@169	2657 can restore deleted messages.
meillo@131	2658 All operations on deleted files are still covered by the MH tools.
meillo@131	2659 The trash folder is just like any other folder in the mail storage.
meillo@131	2660 .P
meillo@131	2661 Similar to the draft folder case, I dropped the old backup prefix approach
meillo@131	2662 in favor for replacing it by the better suiting trash folder system.
meillo@131	2663 Hence,
meillo@131	2664 .Pn rmm
meillo@131	2665 calls
meillo@131	2666 .Pn refile
meillo@131	2667 to move the to-be-removed message to the trash folder,
meillo@131	2668 .Fn +trash
meillo@131	2669 by default.
meillo@164	2670 To sweep it clean, the user can use
meillo@131	2671 .Cl "rmm -unlink +trash a" ,
meillo@131	2672 where the
meillo@131	2673 .Sw -unlink
meillo@131	2674 switch causes the files to be unlinked.
meillo@168	2675 .Ci 8edc5aaf86f9f77124664f6801bc6c6cdf258173
meillo@168	2676 .Ci ca0b3e830b86700d9e5e31b1784de2bdcaf58fc5
meillo@131	2677 .P
meillo@217	2678 Dropping the legacy approach and converting to the new approach
meillo@217	2679 completely, simplified the code base.
meillo@131	2680 The relationship between
meillo@131	2681 .Pn rmm
meillo@131	2682 and
meillo@131	2683 .Pn refile
meillo@131	2684 was inverted.
meillo@131	2685 In mmh,
meillo@131	2686 .Pn rmm
meillo@131	2687 invokes
meillo@217	2688 .Pn refile .
meillo@217	2689 That used to be the other way round.
meillo@131	2690 Yet, the relationship is simpler now.
meillo@164	2691 Loops, like described in nmh's man page for
meillo@131	2692 .Mp refile (1),
meillo@164	2693 can no longer occur:
meillo@131	2694 .QS
meillo@131	2695 Since
meillo@131	2696 .Pn refile
meillo@131	2697 uses your
meillo@131	2698 .Pe rmmproc
meillo@131	2699 to delete the message, the
meillo@131	2700 .Pe rmmproc
meillo@131	2701 must NOT call
meillo@131	2702 .Pn refile
meillo@131	2703 without specifying
meillo@131	2704 .Sw -normmproc
meillo@131	2705 or you will create an infinite loop.
meillo@131	2706 .QE
meillo@131	2707 .LP
meillo@131	2708 .Pn rmm
meillo@131	2709 either unlinks a message with
meillo@131	2710 .Fu unlink()
meillo@131	2711 or invokes
meillo@131	2712 .Pn refile
meillo@131	2713 to move it to the trash folder.
meillo@131	2714 .Pn refile
meillo@131	2715 does not invoke any tools.
meillo@131	2716 .P
meillo@136	2717 By generalizing the message removal in the way that it became covered
meillo@136	2718 by the MH concepts made the whole system more powerful.
meillo@131	2719
meillo@131	2720
meillo@131	2721
meillo@131	2722
meillo@131	2723
meillo@133	2724 .H2 "Modern Defaults
meillo@133	2725 .P
meillo@133	2726 Nmh has a bunch of convenience-improving features inactive by default,
meillo@217	2727 although one can expect every new user to want them active.
meillo@133	2728 The reason they are inactive by default is the wish to stay compatible
meillo@133	2729 with old versions.
meillo@217	2730 But what are old versions?
meillo@136	2731 Still, the highly useful draft folder facility has not been activated
meillo@136	2732 by default although it was introduced over twenty-five years ago.
meillo@133	2733 .[
meillo@133	2734 rose romine real work
meillo@133	2735 .]
meillo@136	2736 The community seems not to care.
meillo@217	2737 .P
meillo@217	2738 In nmh, new users are required to first build up
meillo@217	2739 a profile before they can access the modern features.
meillo@136	2740 Without an extensive profile, the setup is hardly usable
meillo@133	2741 for modern emailing.
meillo@133	2742 The point is not the customization of the setup,
meillo@136	2743 but the need to activate generally useful facilities.
meillo@217	2744 Yet, the real problem lies less in enabling the features,
meillo@217	2745 as this is straight forward as soon as one knows what he wants.
meillo@168	2746 The real problem is that new users need deep insight into the project
meillo@217	2747 to discover the available but inactive features.
meillo@133	2748 To give an example, I needed one year of using nmh
meillo@133	2749 before I became aware of the existence of the attachment system.
meillo@133	2750 One could argue that this fact disqualifies my reading of the
meillo@133	2751 documentation.
meillo@133	2752 If I would have installed nmh from source back then, I could agree.
meillo@217	2753 Yet, I had used a pre-packaged version and had expected that it would
meillo@133	2754 just work.
meillo@133	2755 Nevertheless, I had been convinced by the concepts of MH already
meillo@133	2756 and I am a software developer,
meillo@133	2757 still I required a lot of time to discover the cool features.
meillo@133	2758 How can we expect users to be even more advanced than me,
meillo@217	2759 just to enable them to use MH in a convenient and modern way?
meillo@133	2760 Unless they are strongly convinced of the concepts, they will fail.
meillo@133	2761 I have seen friends of me giving up disappointed
meillo@133	2762 before they truly used the system,
meillo@133	2763 although they had been motivated in the beginning.
meillo@217	2764 New users suffer hard enough to get used to the tool chest approach,
meillo@179	2765 we developers should spare them further inconveniences.
meillo@133	2766 .P
meillo@136	2767 Maintaining compatibility for its own sake is bad,
meillo@217	2768 because the code base will collect more and more compatibility code.
meillo@200	2769 Sticking to the compatibility code means remaining limited;
meillo@168	2770 whereas adjusting to the changes renders the compatibility unnecessary.
meillo@217	2771 Keeping unused alternatives in the code for longer than a short
meillo@217	2772 grace time is a bad choice as they likely
meillo@217	2773 gather bugs by not being constantly tested.
meillo@136	2774 Also, the increased code size and the greater number of conditions
meillo@136	2775 increase the maintenance costs.
meillo@133	2776 If any MH implementation would be the back-end of widespread
meillo@133	2777 email clients with large user bases, compatibility would be more
meillo@133	2778 important.
meillo@133	2779 Yet, it appears as if this is not the case.
meillo@133	2780 Hence, compatibility is hardly important for technical reasons.
meillo@217	2781 Its importance originates from personal reasons rather.
meillo@133	2782 Nmh's user base is small and old.
meillo@217	2783 Changing the interfaces causes inconvenience to long-term users of MH.
meillo@217	2784 It forces them to change their many years old MH configurations.
meillo@168	2785 I do understand this aspect, but by sticking to the old users,
meillo@217	2786 new users are kept from entering the world of MH.
meillo@217	2787 But the future lies in new users.
meillo@168	2788 In consequence, mmh invites new users by providing a convenient
meillo@168	2789 and modern setup, readily usable out-of-the-box.
meillo@133	2790 .P
meillo@136	2791 In mmh, all modern features are active by default and many previous
meillo@217	2792 approaches are removed or only accessible in a manual way.
meillo@136	2793 New default features include:
meillo@133	2794 .BU
meillo@133	2795 The attachment system (\c
meillo@217	2796 .Hd Attach )
meillo@217	2797 .Ci 8ff284ff9167eff8f5349481529332d59ed913b1 .
meillo@133	2798 .BU
meillo@133	2799 The draft folder facility (\c
meillo@217	2800 .Fn +drafts )
meillo@217	2801 .Ci 337338b404931f06f0db2119c9e145e8ca5a9860 .
meillo@133	2802 .BU
meillo@133	2803 The unseen sequence (`u')
meillo@133	2804 .Ci c2360569e1d8d3678e294eb7c1354cb8bf7501c1
meillo@217	2805 and the sequence negation prefix (`!')
meillo@217	2806 .Ci db74c2bd004b2dc9bf8086a6d8bf773ac051f3cc .
meillo@133	2807 .BU
meillo@217	2808 Quoting the original message in the reply
meillo@217	2809 .Ci 67411b1f95d6ec987b4c732459e1ba8a8ac192c6 .
meillo@133	2810 .BU
meillo@217	2811 Forwarding messages using MIME
meillo@217	2812 .Ci 6e271608b7b9c23771523f88d23a4d3593010cf1 .
meillo@171	2813 .LP
meillo@217	2814 An mmh setup with a profile that defines only the path to the
meillo@136	2815 mail storage, is already convenient to use.
meillo@217	2816 Again, Paul Vixie's supports the direction I took:
meillo@136	2817 ``the `main branch' should just be modern''.
meillo@136	2818 .[
meillo@136	2819 paul vixie edginess nmh-workers
meillo@136	2820 .]
meillo@131	2821
meillo@133	2822
meillo@133	2823
meillo@133	2824
meillo@133	2825
meillo@133	2826 .\" --------------------------------------------------------------
meillo@131	2827 .H1 "Styling
meillo@22	2828 .P
meillo@118	2829 Kernighan and Pike have emphasized the importance of style in the
meillo@219	2830 preface of \fPThe Practice of Programming\fP:
meillo@118	2831 .[ [
meillo@118	2832 kernighan pike practice of programming
meillo@118	2833 .], p. x]
meillo@118	2834 .QS
meillo@118	2835 Chapter 1 discusses programming style.
meillo@219	2836 Good style is so important to good programming that we have chosen
meillo@118	2837 to cover it first.
meillo@118	2838 .QE
meillo@168	2839 This section covers changes in mmh that were guided by the desire
meillo@118	2840 to improve on style.
meillo@219	2841 Many of them follow the advice given in the quoted book.
meillo@118	2842
meillo@118	2843
meillo@127	2844
meillo@127	2845
meillo@127	2846 .H2 "Code Style
meillo@154	2847 .Id code-style
meillo@118	2848 .P
meillo@118	2849 .U3 "Indentation Style
meillo@118	2850 .P
meillo@219	2851 Indentation styles are the holy cow of programming.
meillo@219	2852 Kernighan and Pike write:
meillo@118	2853 .[ [
meillo@118	2854 kernighan pike practice of programming
meillo@118	2855 .], p. 10]
meillo@118	2856 .QS
meillo@118	2857 Programmers have always argued about the layout of programs,
meillo@118	2858 but the specific style is much less important than its consistent
meillo@118	2859 application.
meillo@121	2860 Pick one style, preferably ours, use it consistently, and don't waste
meillo@118	2861 time arguing.
meillo@118	2862 .QE
meillo@118	2863 .P
meillo@118	2864 I agree that the constant application is most important,
meillo@118	2865 but I believe that some styles have advantages over others.
meillo@118	2866 For instance the indentation with tab characters only.
meillo@219	2867 The number of tabs corresponds to the nesting level \(en
meillo@118	2868 one tab, one level.
meillo@219	2869 Tab characters provide flexible visual appearance because developers
meillo@219	2870 can adjust their width as prefered.
meillo@219	2871 There is no more need to check for the correct mixture of
meillo@219	2872 tabs and spaces.
meillo@219	2873 Two simple rules ensure the integrity and flexibility of the visual
meillo@219	2874 appearance:
meillo@219	2875 .LI 1
meillo@219	2876 Leading whitespace must consist of tabs only.
meillo@219	2877 .LI 2
meillo@219	2878 All other whitespace should be spaces.
meillo@219	2879 .LP
meillo@121	2880 Although reformatting existing code should be avoided, I did it.
meillo@200	2881 I did not waste time arguing; I just reformatted the code.
meillo@118	2882 .Ci a485ed478abbd599d8c9aab48934e7a26733ecb1
meillo@118	2883
meillo@118	2884 .U3 "Comments
meillo@118	2885 .P
meillo@219	2886 Kernighan and Pike demand: ``Don't belabor the obvious''.
meillo@118	2887 .[ [
meillo@118	2888 kernighan pike practice of programming
meillo@118	2889 .], p. 23]
meillo@219	2890 Following the advice, I removed unnecessary comments.
meillo@219	2891 For instance, I removed all comments in the following code excerpt
meillo@219	2892 .Ci 426543622b377fc5d091455cba685e114b6df674 :
meillo@118	2893 .VS
meillo@120	2894 context_replace(curfolder, folder); /* update current folder */
meillo@120	2895 seq_setcur(mp, mp->lowsel); /* update current message */
meillo@120	2896 seq_save(mp); /* synchronize message sequences */
meillo@120	2897 folder_free(mp); /* free folder/message structure */
meillo@120	2898 context_save(); /* save the context file */
meillo@120	2899
meillo@120	2900 [...]
meillo@120	2901
meillo@120	2902 int c; /* current character */
meillo@120	2903 char cp; / miscellaneous character pointer */
meillo@120	2904
meillo@120	2905 [...]
meillo@120	2906
meillo@120	2907 /* NUL-terminate the field */
meillo@120	2908 *cp = '\0';
meillo@118	2909 VE
meillo@118	2910 .P
meillo@219	2911 The information in each of the comments was present in the code
meillo@219	2912 statements already, except for the NUL-termination, which became
meillo@219	2913 obvious from the context.
meillo@136	2914
meillo@118	2915
meillo@118	2916 .U3 "Names
meillo@118	2917 .P
meillo@219	2918 Regarding this topic, Kernighan and Pike suggest:
meillo@118	2919 ``Use active names for functions''.
meillo@118	2920 .[ [
meillo@118	2921 kernighan pike practice of programming
meillo@118	2922 .], p. 4]
meillo@118	2923 One application of this rule was the rename of
meillo@118	2924 .Fu check_charset()
meillo@118	2925 to
meillo@219	2926 .Fu is_native_charset()
meillo@219	2927 .Ci 8d77b48284c58c135a6b2787e721597346ab056d .
meillo@219	2928 The same change additionally fixed a violation of ``Be accurate'',
meillo@181	2929 .[ [
meillo@181	2930 kernighan pike practice of programming
meillo@181	2931 .], p. 4]
meillo@219	2932 as the code did not match the expectation the function suggested.
meillo@219	2933 It did not compare charset names but prefixes of them only.
meillo@219	2934 In case the native charset was `ISO-8859-1', then
meillo@219	2935 .VS
meillo@219	2936 check_charset("ISO-8859-11", strlen("ISO-8859-11"))
meillo@219	2937 VE
meillo@219	2938 had returned true although the upper halves of the code pages
meillo@219	2939 are different.
meillo@118	2940 .P
meillo@118	2941 More important than using active names is using descriptive names.
meillo@145	2942 .VS
meillo@145	2943 m_unknown(in); /* the MAGIC invocation... */
meillo@145	2944 VE
meillo@145	2945 Renaming the obscure
meillo@118	2946 .Fu m_unknown()
meillo@219	2947 function was a delightful event, although it made the code less funny
meillo@219	2948 .Ci 611d68d19204d7cbf5bd585391249cb5bafca846 .
meillo@118	2949 .P
meillo@118	2950 Magic numbers are generally considered bad style.
meillo@118	2951 Obviously, Kernighan and Pike agree:
meillo@118	2952 ``Give names to magic numbers''.
meillo@118	2953 .[ [
meillo@118	2954 kernighan pike practice of programming
meillo@118	2955 .], p. 19]
meillo@219	2956 .P
meillo@219	2957 The argument
meillo@219	2958 .CW outnum
meillo@219	2959 of the function
meillo@219	2960 .Fu scan()
meillo@219	2961 in
meillo@219	2962 .Fn uip/scansbr.c
meillo@219	2963 holds the number of the message to be created.
meillo@219	2964 As well it encodes program logic with negative numbers and zero.
meillo@219	2965 This led to obscure code.
meillo@219	2966 I clarified the code by introducing two variables that extracted
meillo@219	2967 the hidden information:
meillo@219	2968 .VS
meillo@219	2969 int incing = (outnum > 0);
meillo@219	2970 int ismbox = (outnum != 0);
meillo@219	2971 VE
meillo@219	2972 The readable names are thus used in conditions;
meillo@219	2973 the variable
meillo@219	2974 .CW outnum
meillo@219	2975 is used only to extract ordinary message numbers
meillo@219	2976 .Ci b8b075c77be7794f3ae9ff0e8cedb12b48fd139f .
meillo@219	2977 .P
meillo@219	2978 Through the clarity improvement of the change detours in the program
meillo@219	2979 logic of related code parts became apparent.
meillo@219	2980 The implementation was simplified.
meillo@219	2981 This possibility to improve had been invisible before
meillo@219	2982 .Ci aa60b0ab5e804f8befa890c0a6df0e3143ce0723 .
meillo@219	2983 .P
meillo@219	2984 The names just described were a first step, yet the situation
meillo@219	2985 was further improved by giving names to the magic values of
meillo@219	2986 .CW outnum :
meillo@118	2987 .VS
meillo@118	2988 #define SCN_MBOX (-1)
meillo@118	2989 #define SCN_FOLD 0
meillo@118	2990 VE
meillo@219	2991 The two variables were updated thereafter as well:
meillo@219	2992 .VS
meillo@219	2993 int incing = (outnum != SCN_MBOX && outnum != SCN_FOLD);
meillo@219	2994 int scanfolder = (outnum == SCN_FOLD);
meillo@219	2995 VE
meillo@219	2996 Furthermore,
meillo@219	2997 .CW ismbox
meillo@219	2998 was replaced by
meillo@219	2999 .CW scanfolder
meillo@219	3000 because that matched better to the program logic.
meillo@118	3001 .Ci 7ffb36d28e517a6f3a10272056fc127592ab1c19
meillo@219	3002
meillo@118	3003
meillo@133	3004
meillo@133	3005
meillo@133	3006 .H2 "Structural Rework
meillo@133	3007 .P
meillo@219	3008 Although the stylistic changes described already improve the
meillo@219	3009 readability of the source code, all of them were changes ``in the small''.
meillo@219	3010 Structural changes, in contrast, affect much larger code areas.
meillo@219	3011 They are more difficult to accomplish but lead to larger improvements,
meillo@219	3012 especially as they often influence the outer shape of the tools as well.
meillo@118	3013 .P
meillo@118	3014 At the end of their chapter on style,
meillo@118	3015 Kernighan and Pike ask: ``But why worry about style?''
meillo@181	3016 .[ [
meillo@181	3017 kernighan pike practice of programming
meillo@219	3018 .], p. 28].
meillo@219	3019 Following are two examples of structural rework that demonstrate
meillo@136	3020 why style is important in the first place.
meillo@136	3021
meillo@136	3022
meillo@136	3023 .U3 "Rework of \f(CWanno\fP
meillo@118	3024 .P
meillo@120	3025 Until 2002,
meillo@120	3026 .Pn anno
meillo@219	3027 had six functional command line switches:
meillo@120	3028 .Sw -component
meillo@120	3029 and
meillo@120	3030 .Sw -text ,
meillo@219	3031 each with an argument,
meillo@120	3032 and the two pairs of flags,
meillo@120	3033 .Sw -[no]date
meillo@120	3034 and
meillo@120	3035 .Sw -[no]inplace .
meillo@120	3036 Then Jon Steinhart introduced his attachment system.
meillo@120	3037 In need for more advanced annotation handling, he extended
meillo@120	3038 .Pn anno .
meillo@120	3039 He added five more switches:
meillo@120	3040 .Sw -draft ,
meillo@120	3041 .Sw -list ,
meillo@120	3042 .Sw -delete ,
meillo@120	3043 .Sw -append ,
meillo@120	3044 and
meillo@120	3045 .Sw -number ,
meillo@219	3046 the last one taking an argument
meillo@219	3047 .Ci 7480dbc14bc90f2d872d434205c0784704213252 .
meillo@120	3048 Later,
meillo@120	3049 .Sw -[no]preserve
meillo@219	3050 was added as well
meillo@219	3051 .Ci d9b1d57351d104d7ec1a5621f090657dcce8cb7f .
meillo@120	3052 Then, the Synopsis section of the man page
meillo@120	3053 .Mp anno (1)
meillo@120	3054 read:
meillo@120	3055 .VS
meillo@219	3056 anno [+folder] [msgs] [-component f(CIfieldfP] [-inplace \| -noinplace]
meillo@120	3057 [-date \| -nodate] [-draft] [-append] [-list] [-delete]
meillo@219	3058 [-number [f(CInumfP\|fPallfP]] [-preserve \| -nopreserve] [-version]
meillo@219	3059 [-help] [-text f(CIbodyfP]
meillo@120	3060 VE
meillo@120	3061 .LP
meillo@120	3062 The implementation followed the same structure.
meillo@120	3063 Problems became visible when
meillo@120	3064 .Cl "anno -list -number 42
meillo@219	3065 worked on the current message instead of on message number 42,
meillo@120	3066 and
meillo@120	3067 .Cl "anno -list -number l:5
meillo@124	3068 did not work on the last five messages but failed with the mysterious
meillo@120	3069 error message: ``anno: missing argument to -list''.
meillo@121	3070 Yet, the invocation matched the specification in the man page.
meillo@120	3071 There, the correct use of
meillo@120	3072 .Sw -number
meillo@120	3073 was defined as being
meillo@120	3074 .Cl "[-number [num\|all]]
meillo@120	3075 and the textual description for the combination with
meillo@120	3076 .Sw -list
meillo@120	3077 read:
meillo@120	3078 .QS
meillo@164	3079 The
meillo@164	3080 .Sw -list
meillo@164	3081 option produces a listing of the field bodies for
meillo@120	3082 header fields with names matching the specified component,
meillo@164	3083 one per line. The listing is numbered, starting at 1, if the
meillo@164	3084 .Sw -number
meillo@164	3085 option is also used.
meillo@120	3086 .QE
meillo@120	3087 .LP
meillo@120	3088 The problem was manifold.
meillo@120	3089 Semantically, the argument to the
meillo@120	3090 .Sw -number
meillo@120	3091 switch is only necessary in combination with
meillo@120	3092 .Sw -delete ,
meillo@120	3093 but not with
meillo@120	3094 .Sw -list .
meillo@219	3095 The code, however, required a numeric argument in any case.
meillo@219	3096 If the argument was missing or non-numeric,
meillo@219	3097 .Pn anno
meillo@219	3098 aborted with an error message that additionally had an off-by-one error.
meillo@219	3099 It printed the name of the switch one before the concerned one.
meillo@120	3100 .P
meillo@219	3101 Trying to fix these problems on the surface would not have solved them.
meillo@219	3102 They originate from a discrepance between the
meillo@120	3103 structure of the problem and the structure implemented in the program.
meillo@219	3104 Such structural differences can only be solved by adjusting the
meillo@219	3105 structure of the implementation to the structure of the problem.
meillo@120	3106 .P
meillo@219	3107 Steinhart had added the new
meillo@120	3108 .Sw -list
meillo@120	3109 and
meillo@120	3110 .Sw -delete
meillo@219	3111 switches in a style similar to the other switches though
meillo@219	3112 they are of structural different type.
meillo@120	3113 Semantically,
meillo@120	3114 .Sw -list
meillo@120	3115 and
meillo@120	3116 .Sw -delete
meillo@219	3117 introduce operation modes.
meillo@120	3118 Historically,
meillo@120	3119 .Pn anno
meillo@120	3120 had only one operation mode: adding header fields.
meillo@219	3121 With the extension, two more modes were added:
meillo@120	3122 listing and deleting header fields.
meillo@120	3123 The structure of the code changes did not pay respect to this
meillo@219	3124 fundamental change.
meillo@120	3125 Neither the implementation nor the documentation did clearly
meillo@219	3126 declare the exclusive operation modes as such.
meillo@120	3127 Having identified the problem, I solved it by putting structure into
meillo@120	3128 .Pn anno
meillo@219	3129 and its documentation
meillo@219	3130 .Ci d54c8db8bdf01e8381890f7729bc0ef4a055ea11 .
meillo@120	3131 .P
meillo@173	3132 The difference is visible in both the code and the documentation.
meillo@219	3133 For instance in the following code excerpt:
meillo@120	3134 .VS
meillo@120	3135 int delete = -2; /* delete header element if set */
meillo@120	3136 int list = 0; /* list header elements if set */
meillo@120	3137 [...]
meillo@121	3138 case DELETESW: /* delete annotations */
meillo@121	3139 delete = 0;
meillo@121	3140 continue;
meillo@121	3141 case LISTSW: /* produce a listing */
meillo@121	3142 list = 1;
meillo@121	3143 continue;
meillo@120	3144 VE
meillo@121	3145 .LP
meillo@219	3146 which was replaced by:
meillo@120	3147 .VS
meillo@120	3148 static enum { MODE_ADD, MODE_DEL, MODE_LIST } mode = MODE_ADD;
meillo@120	3149 [...]
meillo@121	3150 case DELETESW: /* delete annotations */
meillo@121	3151 mode = MODE_DEL;
meillo@121	3152 continue;
meillo@121	3153 case LISTSW: /* produce a listing */
meillo@121	3154 mode = MODE_LIST;
meillo@121	3155 continue;
meillo@120	3156 VE
meillo@120	3157 .LP
meillo@121	3158 The replacement code does not only reflect the problem's structure better,
meillo@121	3159 it is easier to understand as well.
meillo@121	3160 The same applies to the documentation.
meillo@120	3161 The man page was completely reorganized to propagate the same structure.
meillo@219	3162 This is already visible in the Synopsis section:
meillo@120	3163 .VS
meillo@219	3164 anno [+folder] [msgs] [-component f(CIfieldfP] [-text fPbodyfP]
meillo@120	3165 [-append] [-date \| -nodate] [-preserve \| -nopreserve]
meillo@120	3166 [-Version] [-help]
meillo@120	3167
meillo@219	3168 anno -delete [+folder] [msgs] [-component fPfieldfP] [-text
meillo@219	3169 fPbodyfP] [-number fPnum fP\| fPall fP] [-preserve \| -nopreserve]
meillo@120	3170 [-Version] [-help]
meillo@120	3171
meillo@219	3172 anno -list [+folder] [msgs] [-component fPfieldfP] [-number]
meillo@120	3173 [-Version] [-help]
meillo@120	3174 VE
meillo@118	3175
meillo@58	3176
meillo@58	3177
meillo@133	3178 .U3 "Path Conversion
meillo@133	3179 .P
meillo@134	3180 Four kinds of path names can appear in MH:
meillo@171	3181 .LI 1
meillo@134	3182 Absolute Unix directory paths, like
meillo@134	3183 .Fn /etc/passwd .
meillo@171	3184 .LI 2
meillo@134	3185 Relative Unix directory paths, like
meillo@134	3186 .Fn ./foo/bar .
meillo@171	3187 .LI 3
meillo@134	3188 Absolute MH folder paths, like
meillo@219	3189 .Fn +projects/mmh .
meillo@171	3190 .LI 4
meillo@134	3191 Relative MH folder paths, like
meillo@134	3192 .Fn @subfolder .
meillo@171	3193 .LP
meillo@219	3194 Relative MH folder paths, are hardly documented
meillo@219	3195 although they are useful for large mail storages.
meillo@134	3196 The current mail folder is specified as `\c
meillo@134	3197 .Fn @ ',
meillo@134	3198 just like the current directory is specified as `\c
meillo@134	3199 .Fn . '.
meillo@134	3200 .P
meillo@134	3201 To allow MH tools to understand all four notations,
meillo@219	3202 they need to be able to convert between them.
meillo@134	3203 In nmh, these path name conversion functions were located in the files
meillo@134	3204 .Fn sbr/path.c
meillo@134	3205 (``return a pathname'') and
meillo@134	3206 .Fn sbr/m_maildir.c
meillo@134	3207 (``get the path for the mail directory'').
meillo@134	3208 The seven functions in the two files were documented with no more
meillo@134	3209 than two comments, which described obvious information.
meillo@219	3210 The signatures of the four exported functions did not explain their
meillo@219	3211 semantics:
meillo@219	3212 .LI 1
meillo@219	3213 .CW "char path(char , int);
meillo@219	3214 .LI 2
meillo@219	3215 .CW "char pluspath(char );
meillo@219	3216 .LI 3
meillo@219	3217 .CW "char m_mailpath(char );
meillo@219	3218 .LI 4
meillo@219	3219 .CW "char m_maildir(char );
meillo@134	3220 .P
meillo@219	3221 My investigations provided the following descriptions:
meillo@171	3222 .LI 1
meillo@134	3223 The second parameter of
meillo@134	3224 .Fu path()
meillo@219	3225 defines the type as which the path given in the first parameter should
meillo@219	3226 be treated.
meillo@134	3227 Directory paths are converted to absolute directory paths.
meillo@134	3228 Folder paths are converted to absolute folder paths.
meillo@173	3229 Folder paths must not include a leading `\fL@\fP' character.
meillo@134	3230 Leading plus characters are preserved.
meillo@134	3231 The result is a pointer to newly allocated memory.
meillo@171	3232 .LI 2
meillo@134	3233 .Fu pluspath()
meillo@134	3234 is a convenience-wrapper to
meillo@134	3235 .Fu path() ,
meillo@134	3236 to convert folder paths only.
meillo@134	3237 This function can not be used for directory paths.
meillo@134	3238 An empty string parameter causes a buffer overflow.
meillo@171	3239 .LI 3
meillo@134	3240 .Fu m_mailpath()
meillo@134	3241 converts directory paths to absolute directory paths.
meillo@173	3242 The characters `\fL+\fP' or `\fL@\fP' at the beginning of the path name are
meillo@134	3243 treated literal, i.e. as the first character of a relative directory path.
meillo@134	3244 Hence, this function can not be used for folder paths.
meillo@219	3245 In any case, the result is an absolute directory path,
meillo@219	3246 returned as a pointer to newly allocated memory.
meillo@171	3247 .LI 4
meillo@134	3248 .Fu m_maildir()
meillo@134	3249 returns the parameter unchanged if it is an absolute directory path
meillo@173	3250 or begins with the entry `\fL.\fP' or `\fL..\fP'.
meillo@134	3251 All other strings are prepended with the current working directory.
meillo@219	3252 Hence, this function can not be used for folder paths.
meillo@134	3253 The result is either an absolute directory path or a relative
meillo@219	3254 directory path, starting with dot or dot-dot.
meillo@134	3255 In contrast to the other functions, the result is a pointer to
meillo@134	3256 static memory.
meillo@134	3257 .P
meillo@134	3258 The situation was obscure, irritating, error-prone, and non-orthogonal.
meillo@219	3259 Additionally, no clear terminology was used to name the different
meillo@219	3260 kinds of path names.
meillo@219	3261 Sometimes, the names were even misleading, much as the first argument of
meillo@134	3262 .Fu m_mailpath() ,
meillo@219	3263 which was named
meillo@219	3264 .CW folder ,
meillo@219	3265 although
meillo@134	3266 .Fu m_mailpath()
meillo@219	3267 could not be used with MH folder arguments.
meillo@134	3268 .P
meillo@219	3269 I clarified the path name conversion by complete rework.
meillo@134	3270 First of all, the terminology needed to be defined.
meillo@134	3271 A path name is either in the Unix domain, then it is called
meillo@219	3272 \fIdirectory path\fP (\fIdirpath\fP for short) or it is in the MH domain,
meillo@219	3273 then it is called \fIfolder path\fP (\fIfolpath\fP for short).
meillo@134	3274 The two terms need to be used with strict distinction.
meillo@219	3275 Often a clear terminology indicates that the problem is understood.
meillo@134	3276 Second, I exploited the concept of path type indicators.
meillo@219	3277 By requiring every path name to start with a distinct type identifier,
meillo@219	3278 the conversion between the types could be fully automated.
meillo@219	3279 This allows the tools to accept paths of any type from the user.
meillo@134	3280 Therefore, it was necessary to require relative directory paths to be
meillo@134	3281 prefixed with a dot character.
meillo@134	3282 In consequence, the dot character could no longer be an alias for the
meillo@134	3283 current message.
meillo@134	3284 .Ci cff0e16925e7edbd25b8b9d6d4fbdf03e0e60c01
meillo@134	3285 Third, I created three new functions to replace the previous mess:
meillo@171	3286 .LI 1
meillo@134	3287 .Fu expandfol()
meillo@219	3288 converts folder paths to absolute folder paths.
meillo@134	3289 Directory paths are simply passed through.
meillo@134	3290 This function is to be used for folder paths only, thus the name.
meillo@134	3291 The result is a pointer to static memory.
meillo@171	3292 .LI 2
meillo@134	3293 .Fu expanddir()
meillo@134	3294 converts directory paths to absolute directory paths.
meillo@134	3295 Folder paths are treated as relative directory paths.
meillo@134	3296 This function is to be used for directory paths only, thus the name.
meillo@134	3297 The result is a pointer to static memory.
meillo@171	3298 .LI 3
meillo@134	3299 .Fu toabsdir()
meillo@134	3300 converts any type of path to an absolute directory path.
meillo@134	3301 This is the function of choice for path conversion.
meillo@134	3302 Absolute directory paths are the most general representation of a
meillo@134	3303 path name.
meillo@134	3304 The result is a pointer to static memory.
meillo@134	3305 .P
meillo@180	3306 .\" XXX ueberfluessig?
meillo@134	3307 The new functions have names that indicate their use.
meillo@134	3308 Two of the functions convert relative to absolute path names of the
meillo@134	3309 same type.
meillo@134	3310 The third function converts any path name type to the most general one,
meillo@134	3311 the absolute directory path.
meillo@134	3312 All of the functions return pointers to static memory.
meillo@219	3313 The file
meillo@219	3314 .Fn sbr/path.c
meillo@219	3315 contains the implementation of the functions;
meillo@134	3316 .Fn sbr/m_maildir.c
meillo@219	3317 was removed.
meillo@168	3318 .Ci d39e2c447b0d163a5a63f480b23d06edb7a73aa0
meillo@134	3319 .P
meillo@134	3320 Along with the path conversion rework, I also replaced
meillo@134	3321 .Fu getfolder(FDEF)
meillo@134	3322 with
meillo@134	3323 .Fu getdeffol()
meillo@134	3324 and
meillo@134	3325 .Fu getfolder(FCUR)
meillo@134	3326 with
meillo@134	3327 .Fu getcurfol() ,
meillo@219	3328 which only wraps
meillo@219	3329 .Fu expandfol(""@"")
meillo@219	3330 for convenience.
meillo@134	3331 This code was moved from
meillo@134	3332 .Fn sbr/getfolder.c
meillo@219	3333 into
meillo@219	3334 .Fn sbr/path.c
meillo@219	3335 as well.
meillo@168	3336 .Ci d39e2c447b0d163a5a63f480b23d06edb7a73aa0
meillo@134	3337 .P
meillo@134	3338 The related function
meillo@134	3339 .Fu etcpath()
meillo@219	3340 is now included in
meillo@134	3341 .Fn sbr/path.c ,
meillo@168	3342 too
meillo@168	3343 .Ci b4c29794c12099556151d93a860ee51badae2e35 .
meillo@134	3344 Previously, it had been located in
meillo@219	3345 .Fn config/config.c .
meillo@134	3346 .P
meillo@219	3347 Now,
meillo@134	3348 .Fn sbr/path.c
meillo@219	3349 contains all path handling code.
meillo@219	3350 Besides being less code, its readability is highly improved.
meillo@219	3351 The functions follow a common style and are well documented.
meillo@133	3352
meillo@133	3353
meillo@133	3354
meillo@133	3355
meillo@133	3356 .H2 "Profile Reading
meillo@133	3357 .P
meillo@220	3358 The MH profile contains the configuration of a user-specific MH setup.
meillo@220	3359 MH tools read the profile right after starting up
meillo@220	3360 because it contains the location of the user's mail storage
meillo@138	3361 and similar settings that influence the whole setup.
meillo@220	3362 Furthermore, the profile contains the default switches for the tools
meillo@220	3363 as well.
meillo@220	3364 The context file is read along with the profile.
meillo@138	3365 .P
meillo@138	3366 For historic reasons, some MH tools did not read the profile and context.
meillo@138	3367 Among them were
meillo@138	3368 .Pn post /\c
meillo@138	3369 .Pn spost ,
meillo@138	3370 .Pn mhmail ,
meillo@138	3371 and
meillo@138	3372 .Pn slocal .
meillo@138	3373 The reason why these tools ignored the profile were not clearly stated.
meillo@220	3374 During a discussion on the nmh-workers mailing list,
meillo@181	3375 David Levine posted an explanation, quoting John Romine:
meillo@138	3376 .[
meillo@138	3377 nmh-workers levine post profile
meillo@138	3378 .]
meillo@220	3379
meillo@138	3380 .QS
meillo@138	3381 I asked John Romine and here's what he had to say, which
meillo@138	3382 agrees and provides an example that convinces me:
meillo@138	3383 .QS
meillo@164	3384 My take on this is that
meillo@164	3385 .Pn post
meillo@164	3386 should not be called by users directly, and it doesn't read the
meillo@164	3387 .Fn .mh_profile
meillo@138	3388 (only front-end UI programs read the profile).
meillo@138	3389 .QP
meillo@164	3390 For example, there can be contexts where
meillo@164	3391 .Pn post
meillo@164	3392 is called by a helper program (like `\c
meillo@164	3393 .Pn mhmail ')
meillo@164	3394 which may be run by a non-MH user.
meillo@164	3395 We don't want this to prompt the user to create an MH profile, etc.
meillo@138	3396 .QP
meillo@164	3397 My suggestion would be to have
meillo@164	3398 .Pn send
meillo@164	3399 pass a (hidden) `\c
meillo@164	3400 .Sw -fileproc
meillo@164	3401 .Ar proc '
meillo@164	3402 option to
meillo@164	3403 .Pn post
meillo@164	3404 if needed.
meillo@164	3405 You could also
meillo@164	3406 use an environment variable (I think
meillo@164	3407 .Pn send /\c
meillo@164	3408 .Pn whatnow
meillo@164	3409 do this).
meillo@138	3410 .QE
meillo@220	3411 .sp \n(PDu
meillo@164	3412 I think that's the way to go.
meillo@164	3413 My personal preference is to use a command line option,
meillo@164	3414 not an environment variable.
meillo@138	3415 .QE
meillo@220	3416
meillo@138	3417 .P
meillo@220	3418 To solve the problem that
meillo@138	3419 .Pn post
meillo@220	3420 does not honor the
meillo@138	3421 .Pe fileproc
meillo@138	3422 profile entry,
meillo@138	3423 the community roughly agreed that a switch
meillo@138	3424 .Sw -fileproc
meillo@138	3425 should be added to
meillo@138	3426 .Pn post
meillo@138	3427 to be able to pass a different fileproc.
meillo@138	3428 I strongly disagree with this approach because it does not solve
meillo@138	3429 the problem; it only removes a single symptom.
meillo@220	3430 The actual problem is that
meillo@138	3431 .Pn post
meillo@220	3432 does not behave as expected,
meillo@220	3433 though all programs should behave as expected.
meillo@220	3434 Clear and general concepts are a precondition for this.
meillo@220	3435 Thus, there should be no separation into ``front-end UI programs''
meillo@220	3436 and ones that ``should not be called by users directly''.
meillo@220	3437 The real solution is having all MH tools read the profile.
meillo@138	3438 .P
meillo@220	3439 But the problem has a further aspect,
meillo@220	3440 which originates from
meillo@220	3441 .Pn mhmail
meillo@220	3442 mainly.
meillo@138	3443 .Pn mhmail
meillo@138	3444 was intended to be a replacement for
meillo@138	3445 .Pn mailx
meillo@138	3446 on systems with MH installations.
meillo@220	3447 In difference to
meillo@220	3448 .Pn mailx ,
meillo@138	3449 .Pn mhmail
meillo@220	3450 used MH's
meillo@138	3451 .Pn post
meillo@220	3452 to send the message.
meillo@220	3453 The idea was that using
meillo@138	3454 .Pn mhmail
meillo@220	3455 should not be influenced whether the user had
meillo@138	3456 MH set up for himself or not.
meillo@220	3457 Therefore
meillo@138	3458 .Pn mhmail
meillo@220	3459 had not read the profile.
meillo@138	3460 As
meillo@138	3461 .Pn mhmail
meillo@138	3462 used
meillo@138	3463 .Pn post ,
meillo@138	3464 .Pn post
meillo@220	3465 was not allowed to read the profile neither.
meillo@138	3466 This is the reason for the actual problem.
meillo@220	3467 Yet, this was not considered much of a problem because
meillo@138	3468 .Pn post
meillo@138	3469 was not intended to be used by users directly.
meillo@220	3470 To invoke
meillo@220	3471 .Pn post ,
meillo@138	3472 .Pn send
meillo@220	3473 was used an a front-end.
meillo@138	3474 .Pn send
meillo@138	3475 read the profile and passed all relevant values on the command line to
meillo@138	3476 .Pn post
meillo@138	3477 \(en an awkward solution.
meillo@138	3478 .P
meillo@138	3479 The important insight is that
meillo@138	3480 .Pn mhmail
meillo@220	3481 is a wolf in sheep's clothing.
meillo@220	3482 This alien tool broke the concepts because it was treated like
meillo@220	3483 a normal MH tool.
meillo@138	3484 Instead it should have been treated accordingly to its foreign style.
meillo@220	3485 .P
meillo@220	3486 The solution is not to prevent the tools from reading the profile but
meillo@220	3487 to instruct them to read a different profile.
meillo@138	3488 .Pn mhmail
meillo@220	3489 could have set up a well-defined profile and caused the following
meillo@138	3490 .Pn post
meillo@220	3491 to use this profile by exporting an environment variable.
meillo@220	3492 With this approach, no special cases would have been introduced
meillo@220	3493 and no surprises would have been caused.
meillo@220	3494 By writing a wrapper program to provide a clean temporary profile,
meillo@220	3495 the concept could have been generalized orthogonally to the whole
meillo@220	3496 MH tool chest.
meillo@220	3497 .P
meillo@220	3498 In mmh, the wish to have
meillo@220	3499 .Pn mhmail
meillo@220	3500 as a replacement for
meillo@220	3501 .Pn mailx
meillo@220	3502 is considered obsolete.
meillo@220	3503 Mmh's
meillo@220	3504 .Pn mhmail
meillo@220	3505 does no longer cover this use-case
meillo@220	3506 .Ci d36e56e695fe1c482c7920644bfbb6386ac9edb0 .
meillo@220	3507 Currently,
meillo@220	3508 .Pn mhmail
meillo@220	3509 is in a transition state
meillo@220	3510 .Ci 32d4f9daaa70519be3072479232ff7be0500d009 .
meillo@220	3511 It may become a front-end to
meillo@220	3512 .Pn comp ,
meillo@220	3513 which provides an alternative interface which can be more convenient
meillo@220	3514 in some cases.
meillo@220	3515 This would convert
meillo@220	3516 .Pn mhmail
meillo@220	3517 into an ordinary MH tool.
meillo@220	3518 If, however, this idea does not convince, then
meillo@220	3519 .Pn mhmail
meillo@220	3520 will be removed.
meillo@220	3521 .P
meillo@220	3522 In the mmh tool chest, every program reads the profile.
meillo@220	3523 (\c
meillo@220	3524 .Pn slocal
meillo@220	3525 is not considered part of the mmh tool chest (cf. Sec.
meillo@220	3526 .Cf slocal ).)
meillo@220	3527 Mmh has no
meillo@220	3528 .Pn post
meillo@220	3529 program, but it has
meillo@220	3530 .Pn spost ,
meillo@220	3531 which now does read the profile
meillo@220	3532 .Ci 3e017a7abbdf69bf0dff7a4073275961eda1ded8 .
meillo@220	3533 Following this change,
meillo@220	3534 .Pn send
meillo@220	3535 and
meillo@220	3536 .Pn spost
meillo@220	3537 can be considered for merging.
meillo@220	3538 Besides
meillo@220	3539 .Pn send ,
meillo@220	3540 .Pn spost
meillo@220	3541 is only invoked directly by the to-be-changed
meillo@220	3542 .Pn mhmail
meillo@220	3543 implementation and by
meillo@220	3544 .Pn rcvdist ,
meillo@220	3545 which requires rework anyway.
meillo@220	3546
meillo@220	3547 .P
meillo@220	3548 Jeffrey Honig quoted Marshall T. Rose explaining the decision that
meillo@220	3549 .Pn post
meillo@220	3550 ignores the profile:
meillo@138	3551 .[
meillo@197	3552 nmh-workers honig post profile
meillo@138	3553 .]
meillo@138	3554 .QS
meillo@138	3555 when you run mh commands in a script, you want all the defaults to be
meillo@138	3556 what the man page says.
meillo@138	3557 when you run a command by hand, then you want your own defaults...
meillo@138	3558 .QE
meillo@138	3559 .LP
meillo@220	3560 The explanation neither matches the problem concered exactly
meillo@220	3561 nor is the interpretation clear.
meillo@220	3562 If the described desire addresses the technical level,
meillo@220	3563 then it conflicts fundametally with the Unix philosophy,
meillo@220	3564 precisely because the indistinquishability of human and script
meillo@220	3565 input is the main reason for the huge software leverage in Unix.
meillo@220	3566 If, however, the described desire addresses the user's view,
meillo@220	3567 then different technical solutions are more appropriate.
meillo@220	3568 The two cases can be regarded simply as two different MH setups.
meillo@220	3569 Hence, mapping the problem of different behavior between interactive and
meillo@220	3570 automated use on the concept of switching between different profiles,
meillo@220	3571 marks it already solved.
meillo@133	3572
meillo@133	3573
meillo@127	3574
meillo@121	3575 .H2 "Standard Libraries
meillo@22	3576 .P
meillo@121	3577 MH is one decade older than the POSIX and ANSI C standards.
meillo@121	3578 Hence, MH included own implementations of functions
meillo@220	3579 that were neither standardized nor widely available, back then.
meillo@220	3580 Today, twenty years after POSIX and ANSI C were published,
meillo@220	3581 developers can expect that systems comply with these standards.
meillo@121	3582 In consequence, MH-specific replacements for standard functions
meillo@121	3583 can and should be dropped.
meillo@220	3584 Kernighan and Pike advise: ``Use standard libraries''.
meillo@121	3585 .[ [
meillo@121	3586 kernighan pike practice of programming
meillo@121	3587 .], p. 196]
meillo@121	3588 Actually, MH had followed this advice in history,
meillo@220	3589 but it had not adjusted to more recent changes in this field.
meillo@121	3590 The
meillo@121	3591 .Fu snprintf()
meillo@121	3592 function, for instance, was standardized with C99 and is available
meillo@121	3593 almost everywhere because of its high usefulness.
meillo@220	3594 Thus, the project's own implementation of
meillo@121	3595 .Fu snprintf()
meillo@123	3596 was dropped in March 2012 in favor for using the one of the
meillo@220	3597 standard library
meillo@220	3598 .Ci 0052f1024deb0a0a2fc2e5bacf93d45a5a9c9b32 .
meillo@123	3599 Such decisions limit the portability of mmh
meillo@173	3600 if systems do not support these standardized and widespread functions.
meillo@123	3601 This compromise is made because mmh focuses on the future.
meillo@121	3602 .P
meillo@180	3603 .\" XXX kuerzen und mit dem naechsten Absatz vereinen
meillo@220	3604 As I am still in my twenties, have no programming experience from
meillo@220	3605 past decades.
meillo@220	3606 I have not followed the evolution of C through time.
meillo@220	3607 I have not suffered from the the Unix wars.
meillo@121	3608 I have not longed for standardization.
meillo@121	3609 All my programming experience is from a time when ANSI C and POSIX
meillo@121	3610 were well established already.
meillo@220	3611 Thus, I needed to learn about the history in retrospective.
meillo@121	3612 I have only read a lot of books about the (good) old times.
meillo@220	3613 This put me in a difficult position when working with old code.
meillo@123	3614 I need to freshly acquire knowledge about old code constructs and ancient
meillo@123	3615 programming styles, whereas older programmers know these things by
meillo@123	3616 heart from their own experience.
meillo@123	3617 Being aware of the situation, I rather let people with more historic
meillo@220	3618 experience do the transition from ancient code constructs to
meillo@220	3619 standardized ones.
meillo@121	3620 Lyndon Nerenberg covered large parts of this task for the nmh project.
meillo@121	3621 He converted project-specific functions to POSIX replacements,
meillo@121	3622 also removing the conditionals compilation of now standardized features.
meillo@220	3623 Ken Hornstein and David Levine had their part in this work, as well.
meillo@220	3624 Often, I only pulled the changes over from nmh into mmh.
meillo@220	3625 These changes include many commits, among them:
meillo@121	3626 .Ci 768b5edd9623b7238e12ec8dfc409b82a1ed9e2d
meillo@121	3627 .Ci 0052f1024deb0a0a2fc2e5bacf93d45a5a9c9b32 .
meillo@102	3628 .P
meillo@220	3629 Nevertheless, I worked on the task as well, tidying up the
meillo@220	3630 \fIMH standard library\fP,
meillo@220	3631 .Fn libmh.a .
meillo@220	3632 It is located in the
meillo@123	3633 .Fn sbr
meillo@220	3634 (``subroutines'') directory in the source tree and
meillo@220	3635 includes functions that mmh tools usually need.
meillo@123	3636 Among them are MH-specific functions for profile, context, sequence,
meillo@123	3637 and folder handling, but as well
meillo@123	3638 MH-independent functions, such as auxiliary string functions,
meillo@123	3639 portability interfaces and error-checking wrappers for critical
meillo@123	3640 functions of the standard library.
meillo@220	3641 .BU
meillo@123	3642 I have replaced the
meillo@121	3643 .Fu atooi()
meillo@121	3644 function with calls to
meillo@220	3645 .Fu strtoul() ,
meillo@220	3646 setting the third parameter, the base, to eight.
meillo@123	3647 .Fu strtoul()
meillo@220	3648 is part of C89 and thus considered safe to use
meillo@220	3649 .Ci c490c51b3c0f8871b6953bd0c74551404f840a74 .
meillo@220	3650 .BU
meillo@121	3651 I did remove project-included fallback implementations of
meillo@121	3652 .Fu memmove()
meillo@121	3653 and
meillo@220	3654 .Fu strerror()
meillo@220	3655 .Ci b067ff5c465a5d243ce5a19e562085a9a1a97215 ,
meillo@121	3656 although Peter Maydell had re-included them into nmh in 2008
meillo@121	3657 to support SunOS 4.
meillo@121	3658 Nevertheless, these functions are part of ANSI C.
meillo@121	3659 Systems that do not even provide full ANSI C support should not
meillo@121	3660 put a load on mmh.
meillo@220	3661 .BU
meillo@121	3662 The
meillo@121	3663 .Fu copy()
meillo@180	3664 function copies the string in parameter one to the location in
meillo@180	3665 parameter two.
meillo@121	3666 In contrast to
meillo@121	3667 .Fu strcpy() ,
meillo@121	3668 it returns a pointer to the terminating null-byte in the destination area.
meillo@123	3669 The code was adjusted to replace
meillo@121	3670 .Fu copy()
meillo@123	3671 with
meillo@121	3672 .Fu strcpy() ,
meillo@121	3673 except within
meillo@121	3674 .Fu concat() ,
meillo@121	3675 where
meillo@121	3676 .Fu copy()
meillo@123	3677 was more convenient.
meillo@123	3678 Therefore, the definition of
meillo@121	3679 .Fu copy()
meillo@123	3680 was moved into the source file of
meillo@121	3681 .Fu concat()
meillo@220	3682 and its visibility it limited to that
meillo@220	3683 .Ci 552fd7253e5ee9e554c5c7a8248a6322aa4363bb .
meillo@220	3684 .BU
meillo@121	3685 The function
meillo@121	3686 .Fu r1bindex()
meillo@121	3687 had been a generalized version of
meillo@121	3688 .Fu basename()
meillo@121	3689 with minor differences.
meillo@121	3690 As all calls to
meillo@121	3691 .Fu r1bindex()
meillo@220	3692 had the slash (`\fL/\fP') as delimiter anyway,
meillo@121	3693 replacing
meillo@121	3694 .Fu r1bindex()
meillo@121	3695 with the more specific and better-named function
meillo@121	3696 .Fu basename()
meillo@121	3697 became desirable.
meillo@121	3698 Unfortunately, many of the 54 calls to
meillo@121	3699 .Fu r1bindex()
meillo@123	3700 depended on a special behavior,
meillo@121	3701 which differed from the POSIX specification for
meillo@121	3702 .Fu basename() .
meillo@121	3703 Hence,
meillo@121	3704 .Fu r1bindex()
meillo@121	3705 was kept but renamed to
meillo@123	3706 .Fu mhbasename() ,
meillo@220	3707 setting the delimiter to the slash
meillo@220	3708 .Ci 240013872c392fe644bd4f79382d9f5314b4ea60 .
meillo@121	3709 For possible uses of
meillo@121	3710 .Fu r1bindex()
meillo@121	3711 with a different delimiter,
meillo@121	3712 the ANSI C function
meillo@121	3713 .Fu strrchr()
meillo@121	3714 provides the core functionality.
meillo@220	3715 .BU
meillo@121	3716 The
meillo@121	3717 .Fu ssequal()
meillo@121	3718 function \(en apparently for ``substring equal'' \(en
meillo@121	3719 was renamed to
meillo@121	3720 .Fu isprefix() ,
meillo@220	3721 because this is what it actually checked
meillo@220	3722 .Ci c20b4fa14515c7ab388ce35411d89a7a92300711.
meillo@220	3723 Its source file had included both of the following comments, no joke.
meillo@220	3724 .in -\n(PIu
meillo@121	3725 .VS
meillo@121	3726 /*
meillo@121	3727 * THIS CODE DOES NOT WORK AS ADVERTISED.
meillo@121	3728 * It is actually checking if s1 is a PREFIX of s2.
meillo@121	3729 * All calls to this function need to be checked to see
meillo@121	3730 * if that needs to be changed. Prefix checking is cheaper, so
meillo@121	3731 * should be kept if it's sufficient.
meillo@121	3732 */
meillo@121	3733
meillo@121	3734 /*
meillo@121	3735 * Check if s1 is a substring of s2.
meillo@121	3736 * If yes, then return 1, else return 0.
meillo@121	3737 */
meillo@121	3738 VE
meillo@220	3739 .in +\n(PIu
meillo@220	3740 Eventually, the function was completely replaced with calls to
meillo@220	3741 .Fu strncmp()
meillo@220	3742 .Ci b0b1dd37ff515578cf7cba51625189eb34a196cb .
meillo@121	3743
meillo@102	3744
meillo@102	3745
meillo@102	3746
meillo@133	3747
meillo@133	3748 .H2 "User Data Locations
meillo@133	3749 .P
meillo@133	3750 In nmh, a personal setup consists of the MH profile and the MH directory.
meillo@133	3751 The profile is a file named
meillo@133	3752 .Fn \&.mh_profile
meillo@133	3753 in the user's home directory.
meillo@133	3754 It contains the static configuration.
meillo@133	3755 It also contains the location of the MH directory in the profile entry
meillo@133	3756 .Pe Path .
meillo@133	3757 The MH directory contains the mail storage and is the first
meillo@220	3758 place to search for form files, scan formats, and similar
meillo@133	3759 configuration files.
meillo@133	3760 The location of the MH directory can be chosen freely by the user.
meillo@220	3761 The usual name is a directory named
meillo@133	3762 .Fn Mail
meillo@220	3763 in the user's home directory.
meillo@133	3764 .P
meillo@200	3765 The way MH data is split between profile and MH directory is a legacy.
meillo@133	3766 It is only sensible in a situation where the profile is the only
meillo@133	3767 configuration file.
meillo@133	3768 Why else should the mail storage and the configuration files be intermixed?
meillo@220	3769 They are of different kind:
meillo@220	3770 One kind is the data to be operated on and the other kind is
meillo@220	3771 the configuration to change how tools operate.
meillo@133	3772 Splitting the configuration between the profile and the MH directory
meillo@220	3773 is inappropriate, as well.
meillo@220	3774 I improved the situation by breaking compatibility.
meillo@133	3775 .P
meillo@220	3776 In mmh, personal data is grouped by type.
meillo@220	3777 This results in two distinct parts:
meillo@171	3778 the mail storage and the configuration.
meillo@220	3779 The mail storage directory still contains all the messages,
meillo@133	3780 but, in exception of public sequences files, nothing else.
meillo@133	3781 In difference to nmh, the auxiliary configuration files are no longer
meillo@133	3782 located there.
meillo@133	3783 Therefore, the directory is no longer called the user's \fIMH directory\fP
meillo@220	3784 but the user's \fImail storage\fP.
meillo@133	3785 Its location is still user-chosen, with the default name
meillo@220	3786 .Fn Mail
meillo@133	3787 in the user's home directory.
meillo@220	3788 The configuration is grouped together in the hidden directory
meillo@133	3789 .Fn \&.mmh
meillo@133	3790 in the user's home directory.
meillo@133	3791 This \fImmh directory\fP contains the context file, personal forms,
meillo@133	3792 scan formats, and the like, but also the user's profile, now named
meillo@133	3793 .Fn profile .
meillo@220	3794 The path to the profile is no longer
meillo@133	3795 .Fn $HOME/.mh_profile
meillo@220	3796 but
meillo@133	3797 .Fn $HOME/.mmh/profile .
meillo@220	3798 (The alternative of having file
meillo@133	3799 .Fn $HOME/.mh_profile
meillo@220	3800 and a configuration directory
meillo@133	3801 .Fn $HOME/.mmh
meillo@220	3802 appeared to be inconsistent.)
meillo@220	3803 .P
meillo@133	3804 The approach chosen for mmh is consistent, simple, and familiar to
meillo@133	3805 Unix users.
meillo@220	3806 The main achievement of the change is the clear and sensible separation
meillo@220	3807 of the mail storage and the configuration.
meillo@168	3808 .Ci 7030d7edb099bff36ded7548bb5380f7acab4f9b
meillo@133	3809 .P
meillo@220	3810 As MH allows users to have multiple MH setups,
meillo@220	3811 it is necessary to switch the profile.
meillo@133	3812 The profile is the single entry point to access the rest of a
meillo@133	3813 personal MH setup.
meillo@133	3814 In nmh, the environment variable
meillo@133	3815 .Ev MH
meillo@220	3816 is used to specify a different profile.
meillo@220	3817 To operate in the same MH setup with a separate context, the
meillo@133	3818 .Ev MHCONTEXT
meillo@220	3819 environment variable is used.
meillo@220	3820 This allows having a separate current folder in each terminal at
meillo@220	3821 the same time, for instance.
meillo@220	3822 In mmh, three environment variables replace the two of nmh.
meillo@133	3823 .Ev MMH
meillo@133	3824 overrides the default location of the mmh directory (\c
meillo@133	3825 .Fn .mmh ).
meillo@133	3826 .Ev MMHP
meillo@133	3827 and
meillo@133	3828 .Ev MMHC
meillo@220	3829 override the paths to the profile and context file, respectively.
meillo@133	3830 This approach allows the set of personal configuration files to be chosen
meillo@220	3831 independently of the profile, context, and mail storage.
meillo@133	3832 The new approach has no functional disadvantages,
meillo@133	3833 as every setup I can imagine can be implemented with both approaches,
meillo@220	3834 possibly even easier with the new one.
meillo@220	3835 .Ci 7030d7edb099bff36ded7548bb5380f7acab4f9b
meillo@133	3836
meillo@133	3837
meillo@133	3838
meillo@133	3839
meillo@133	3840
meillo@118	3841 .H2 "Modularization
meillo@220	3842 .Id modularization
meillo@118	3843 .P
meillo@123	3844 The source code of the mmh tools is located in the
meillo@122	3845 .Fn uip
meillo@123	3846 (``user interface programs'') directory.
meillo@180	3847 Each tool has a source file with the name of the command.
meillo@122	3848 For example,
meillo@122	3849 .Pn rmm
meillo@122	3850 is built from
meillo@122	3851 .Fn uip/rmm.c .
meillo@123	3852 Some source files are used for multiple programs.
meillo@122	3853 For example
meillo@122	3854 .Fn uip/scansbr.c
meillo@173	3855 is used for both
meillo@122	3856 .Pn scan
meillo@122	3857 and
meillo@122	3858 .Pn inc .
meillo@122	3859 In nmh, 49 tools were built from 76 source files.
meillo@123	3860 This is a ratio of 1.6 source files per program.
meillo@123	3861 32 programs depended on multiple source files;
meillo@123	3862 17 programs depended on one source file only.
meillo@122	3863 In mmh, 39 tools are built from 51 source files.
meillo@123	3864 This is a ratio of 1.3 source files per program.
meillo@123	3865 18 programs depend on multiple source files;
meillo@123	3866 21 programs depend on one source file only.
meillo@123	3867 (These numbers and the ones in the following text ignore the MH library
meillo@123	3868 as well as shell scripts and multiple names for the same program.)
meillo@180	3869 .\" XXX graph
meillo@122	3870 .P
meillo@123	3871 Splitting the source code of a large program into multiple files can
meillo@220	3872 increase the readability of its source code,
meillo@220	3873 but most of the mmh tools are small and straight-forward programs.
meillo@220	3874 In exception of the MIME handling tools (i.e.
meillo@122	3875 .Pn mhbuild ,
meillo@122	3876 .Pn mhstore ,
meillo@122	3877 .Pn show ,
meillo@220	3878 etc.),
meillo@220	3879 .Pn pick
meillo@220	3880 is the only tool with more than one thousand lines of source code.
meillo@220	3881 Splitting programs with less than one thousand lines of code into
meillo@220	3882 multiple source files leads seldom to better readability.
meillo@220	3883 For such tools, splitting still makes sense
meillo@220	3884 when parts of the code are reused in other programs
meillo@179	3885 and the reused code fragment is (1) not general enough
meillo@179	3886 for including it in the MH library
meillo@179	3887 or (2) has dependencies on a library that only few programs need.
meillo@122	3888 .Fn uip/packsbr.c ,
meillo@122	3889 for instance, provides the core program logic for the
meillo@122	3890 .Pn packf
meillo@122	3891 and
meillo@122	3892 .Pn rcvpack
meillo@122	3893 programs.
meillo@122	3894 .Fn uip/packf.c
meillo@122	3895 and
meillo@122	3896 .Fn uip/rcvpack.c
meillo@122	3897 mainly wrap the core function appropriately.
meillo@122	3898 No other tools use the folder packing functions.
meillo@123	3899 As another example,
meillo@123	3900 .Fn uip/termsbr.c
meillo@220	3901 accesses terminal properties, which requires linking with the
meillo@220	3902 \fItermcap\fP or a \fIcurses\fP library.
meillo@220	3903 If
meillo@123	3904 .Fn uip/termsbr.c
meillo@220	3905 is included in the MH library, then every program needs to be linked
meillo@220	3906 with termcap or curses, although only few of the programs use
meillo@220	3907 the library.
meillo@122	3908 .P
meillo@122	3909 The task of MIME handling is complex enough that splitting its code
meillo@122	3910 into multiple source files improves the readability.
meillo@122	3911 The program
meillo@122	3912 .Pn mhstore ,
meillo@122	3913 for instance, is compiled out of seven source files with 2\\|500
meillo@122	3914 lines of code in summary.
meillo@122	3915 The main code file
meillo@122	3916 .Fn uip/mhstore.c
meillo@220	3917 consists of 800 lines; the other 1\\|700 lines are code reused in
meillo@123	3918 other MIME handling tools.
meillo@123	3919 It seems to be worthwhile to bundle the generic MIME handling code into
meillo@123	3920 a MH-MIME library, as a companion to the MH standard library.
meillo@220	3921 This is left to be done.
meillo@122	3922 .P
meillo@169	3923 The work already accomplished focussed on the non-MIME tools.
meillo@122	3924 The amount of code compiled into each program was reduced.
meillo@123	3925 This eases the understanding of the code base.
meillo@122	3926 In nmh,
meillo@122	3927 .Pn comp
meillo@122	3928 was built from six source files:
meillo@122	3929 .Fn comp.c ,
meillo@122	3930 .Fn whatnowproc.c ,
meillo@122	3931 .Fn whatnowsbr.c ,
meillo@122	3932 .Fn sendsbr.c ,
meillo@122	3933 .Fn annosbr.c ,
meillo@122	3934 and
meillo@122	3935 .Fn distsbr.c .
meillo@122	3936 In mmh, it builds from only two:
meillo@122	3937 .Fn comp.c
meillo@122	3938 and
meillo@122	3939 .Fn whatnowproc.c .
meillo@123	3940 In nmh's
meillo@123	3941 .Pn comp ,
meillo@123	3942 the core function of
meillo@122	3943 .Pn whatnow ,
meillo@122	3944 .Pn send ,
meillo@122	3945 and
meillo@122	3946 .Pn anno
meillo@220	3947 were all compiled into
meillo@122	3948 .Pn comp .
meillo@123	3949 This saved the need to execute these programs with
meillo@220	3950 the expensive system calls
meillo@122	3951 .Fu fork()
meillo@122	3952 and
meillo@220	3953 .Fu exec() .
meillo@171	3954 Whereas this approach improved the time performance,
meillo@171	3955 it interwove the source code.
meillo@122	3956 Core functionalities were not encapsulated into programs but into
meillo@122	3957 function, which were then wrapped by programs.
meillo@122	3958 For example,
meillo@122	3959 .Fn uip/annosbr.c
meillo@122	3960 included the function
meillo@122	3961 .Fu annotate() .
meillo@122	3962 Each program that wanted to annotate messages, included the source file
meillo@123	3963 .Fn uip/annosbr.c
meillo@123	3964 and called
meillo@123	3965 .Fu annotate() .
meillo@123	3966 Because the function
meillo@123	3967 .Fu annotate()
meillo@123	3968 was used like the tool
meillo@123	3969 .Pn anno ,
meillo@123	3970 it had seven parameters, reflecting the command line switches of the tool.
meillo@122	3971 When another pair of command line switches was added to
meillo@122	3972 .Pn anno ,
meillo@122	3973 a rather ugly hack was implemented to avoid adding another parameter
meillo@220	3974 to the function
meillo@220	3975 .Ci d9b1d57351d104d7ec1a5621f090657dcce8cb7f .
meillo@122	3976 .P
meillo@122	3977 In mmh, the relevant code of
meillo@122	3978 .Pn comp
meillo@122	3979 comprises the two files
meillo@122	3980 .Fn uip/comp.c
meillo@122	3981 and
meillo@122	3982 .Fn uip/whatnowproc.c ,
meillo@220	3983 together 210 lines of code,
meillo@220	3984 whereas in nmh,
meillo@122	3985 .Pn comp
meillo@122	3986 comprises six files with 2\\|450 lines.
meillo@220	3987 Not all of the code in these six files is actually used by
meillo@122	3988 .Pn comp ,
meillo@220	3989 but the reader needed to read it all to know which parts are relevant.
meillo@220	3990 Understanding nmh's
meillo@122	3991 .Pn comp ,
meillo@123	3992 required understanding the inner workings of
meillo@122	3993 .Fn uip/annosbr.c
meillo@122	3994 first.
meillo@123	3995 To be sure to fully understand a program, its whole source code needs
meillo@122	3996 to be examined.
meillo@123	3997 Not doing so is a leap of faith, assuming that the developers
meillo@122	3998 have avoided obscure programming techniques.
meillo@220	3999 Here, it should be recalled that information passed in obscure ways
meillo@220	4000 through the program's source base, due to the aforementioned hack
meillo@220	4001 to save an additional parameter in nmh's
meillo@220	4002 .Pn anno .
meillo@220	4003 .P
meillo@220	4004 In mmh, understanding
meillo@220	4005 .Pn comp
meillo@220	4006 requires to read only 210 lines of code to read, whereas the amount
meillo@220	4007 is ten times more for nmh's
meillo@220	4008 .Pn comp .
meillo@220	4009 .P
meillo@220	4010 By separating the tools on the program-level,
meillo@220	4011 the boundaries are clearly visible, as the interfaces are calls to
meillo@122	4012 .Fu exec()
meillo@122	4013 rather than arbitrary function calls.
meillo@220	4014 Additionally, this kind of separation is more strict because
meillo@220	4015 it is technically enforced by the operating system;
meillo@220	4016 it can not be simply bypassed with global variables.
meillo@220	4017 Good separation simplifies the understanding of program code
meillo@220	4018 because the area influenced by any particular statement is small.
meillo@220	4019 As I have read a lot in nmh's code base during the last two years,
meillo@220	4020 I have learned about the easy and the difficult parts.
meillo@220	4021 In my observation, the understanding of code is enormously eased
meillo@220	4022 if the influenced area is small and clearly bounded.
meillo@123	4023 .P
meillo@220	4024 Yet, the real problem is another:
meillo@123	4025 Nmh violates the golden ``one tool, one job'' rule of the Unix philosophy.
meillo@123	4026 Understanding
meillo@122	4027 .Pn comp
meillo@123	4028 requires understanding
meillo@123	4029 .Fn uip/annosbr.c
meillo@123	4030 and
meillo@123	4031 .Fn uip/sendsbr.c
meillo@123	4032 because
meillo@123	4033 .Pn comp
meillo@220	4034 annotates and sends messages.
meillo@220	4035 In nmh, there surely exist the tools
meillo@220	4036 .Pn anno
meillo@220	4037 and
meillo@122	4038 .Pn send ,
meillo@220	4039 which cover these jobs,
meillo@220	4040 but
meillo@122	4041 .Pn comp
meillo@123	4042 and
meillo@122	4043 .Pn repl
meillo@122	4044 and
meillo@122	4045 .Pn forw
meillo@122	4046 and
meillo@122	4047 .Pn dist
meillo@122	4048 and
meillo@122	4049 .Pn whatnow
meillo@122	4050 and
meillo@220	4051 .Pn viamail
meillo@220	4052 \(en they all (!) \(en
meillo@220	4053 have the same annotating and sending functions included, once more.
meillo@220	4054 As a result,
meillo@123	4055 .Pn comp
meillo@123	4056 sends messages without using
meillo@123	4057 .Pn send .
meillo@123	4058 The situation is the same as if
meillo@123	4059 .Pn grep
meillo@220	4060 would page its output without using
meillo@123	4061 .Pn more
meillo@123	4062 just because both programs are part of the same code base.
meillo@123	4063 .P
meillo@220	4064 The clear separation on the surface of nmh
meillo@220	4065 \(en the tool chest approach \(en
meillo@123	4066 is violated on the level below.
meillo@122	4067 This violation is for the sake of time performance.
meillo@220	4068 Decades ago, sacrificing readability and conceptional beauty
meillo@220	4069 for speed might have been necessary to prevent MH from being
meillo@220	4070 unusably slow, but today this is not the case anymore.
meillo@220	4071 No longer should speed improvements that became unnecessary be kept.
meillo@220	4072 No longer should readability or conceptional beauty be sacrificed.
meillo@220	4073 No longer should the Unix philosophy's ``one tool, one job''
meillo@220	4074 guideline be violated.
meillo@123	4075 Therefore, mmh's
meillo@123	4076 .Pn comp
meillo@220	4077 no longer sends messages.
meillo@220	4078 .P
meillo@123	4079 In mmh, different jobs are divided among separate programs that
meillo@122	4080 invoke each other as needed.
meillo@123	4081 In consequence,
meillo@123	4082 .Pn comp
meillo@123	4083 invokes
meillo@123	4084 .Pn whatnow
meillo@123	4085 which thereafter invokes
meillo@220	4086 .Pn send
meillo@168	4087 .Ci 3df5ab3c116e6d4a2fb4bb5cc9dfc5f781825815
meillo@220	4088 .Ci c73c00bfccd22ec77e9593f47462aeca4a8cd9c0 .
meillo@123	4089 The clear separation on the surface is maintained on the level below.
meillo@220	4090 Human users and other tools use the same interface \(en
meillo@123	4091 annotations, for example, are made by invoking
meillo@123	4092 .Pn anno ,
meillo@220	4093 no matter if requested by programs or by human beings
meillo@168	4094 .Ci 469a4163c2a1a43731d412eaa5d9cae7d670c48b
meillo@168	4095 .Ci aed384169af5204b8002d06e7a22f89197963d2d
meillo@220	4096 .Ci 3caf9e298a8861729ca8b8a84f57022b6f3ea742 .
meillo@123	4097 The decrease of tools built from multiple source files and thus
meillo@123	4098 the decrease of
meillo@123	4099 .Fn uip/*sbr.c
meillo@220	4100 files confirm the improvement
meillo@168	4101 .Ci 9e6d91313f01c96b4058d6bf419a8ca9a207bc33
meillo@168	4102 .ci 81744a46ac9f845d6c2b9908074d269275178d2e
meillo@168	4103 .Ci f0f858069d21111f0dbea510044593f89c9b0829
meillo@168	4104 .Ci 0503a6e9be34f24858b55b555a5c948182b9f24b
meillo@168	4105 .Ci 27826f9353e0f0b04590b7d0f8f83e60462b90f0
meillo@168	4106 .Ci d1da1f94ce62160aebb30df4063ccbc53768656b
meillo@220	4107 .Ci c42222869e318fff5dec395eca3e776db3075455 .
meillo@220	4108 This is also visible in the complexity of the build dependency graphs:
meillo@145	4109
meillo@220	4110 .sp
meillo@220	4111 Nmh:
meillo@220	4112 .BP input/deps-nmh.eps .5i
meillo@220	4113 .EP
meillo@220	4114 .sp
meillo@220	4115 Mmh:
meillo@220	4116 .BP input/deps-mmh.eps .8i
meillo@220	4117 .EP
meillo@145	4118
meillo@220	4119 The figures display all program to source file relationships
meillo@220	4120 that are not one-to-one,
meillo@220	4121 i.e. all programs that are built from multiple source files.
meillo@220	4122 The primary source file of each program is omited from the graph.