Debian Codegraph @

lynx_help/lynx_url_support.html

raw
1 <!-- $LynxId: lynx_url_support.html,v 1.38 2025/07/30 23:39:50 tom Exp $ -->
2 <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN">
3 <html>
4 <head>
5 <meta name="generator" content=
6 "HTML Tidy for HTML5 for Linux version 5.6.0">
7 <title>URL Schemes Supported in Lynx</title>
8 <link rel="author" href="mailto:[email protected]">
9 <meta http-equiv="Content-Type" content=
10 "text/html; charset=us-ascii">
11 <meta name="description" content=
12 "Enumerate, describe and provide examples of Lynx's URL support on Unix and VMS. Lynx supports both Web standards and extensions.">
13 </head>
14 <body>
15 <blockquote>
16 <p><em>[</em><a href="#http_url">http, https</a> <em>|</em>
17 <a href="#telnet_url">telnet, tn3270, rlogin</a> <em>|</em>
18 <a href="#gopher_url">gopher</a> <em>|</em> <a href=
19 "#file_url">file</a> <em>|</em> <a href="#ftp_url">ftp</a>
20 <em>|</em> <a href="#wais_url">wais</a> <em>|</em> <a href=
21 "#news_url">news, nntp, snews</a> <em>|</em> <a href=
22 "#newspost_url">newspost, newsreply, snewspost, snewsreply</a>
23 <em>|</em> <a href="#mailto_url">mailto</a> <em>|</em> <a href=
24 "#finger_url">finger</a> <em>|</em> <a href="#cso_url">cso</a>
25 <em>|</em> <a href="#bibp_url">bibp</a> <em>|</em> <a href=
26 "#exec_url">lynxexec, lynxprog</a> <em>|</em> <a href=
27 "#cgi_url">lynxcgi</a><em>|</em> <a href="#ncftp_url">NcFTP</a>
28 <em>|</em> <a href="#internal_url">internal</a><em>]</em></p>
29 </blockquote>
30
31 <h1><em>URL Schemes Supported in Lynx</em>
32 </h1>
33
34 <p><strong>Lynx</strong> handles a number of URL types, that are
35 enumerated below. For more details about URLs (Uniform Resource
36 Locators) see <em>RFC1738</em>:</p>
37
38 <ul>
39 <li><a href=
40 "http://www.w3.org/Addressing/rfc1738.txt">http://www.w3.org/Addressing/rfc1738.txt</a></li>
41
42 <li><a href=
43 "ftp://ftp.rfc-editor.org/in-notes/rfc1738.txt">ftp://ftp.rfc-editor.org/in-notes/rfc1738.txt</a></li>
44 </ul>
45
46 <p><strong>Lynx</strong> resolves partial or relative URLs in
47 documents with respect to the BASE if one was specified,
48 otherwise with respect to the document's absolute URL, using the
49 rules described in <em>RFC1808</em>:</p>
50
51 <ul>
52 <li><a href=
53 "http://www.w3.org/Addressing/rfc1808.txt">http://www.w3.org/Addressing/rfc1808.txt</a></li>
54
55 <li><a href=
56 "ftp://ftp.rfc-editor.org/in-notes/rfc1808.txt">ftp://ftp.rfc-editor.org/in-notes/rfc1808.txt</a></li>
57 </ul>
58
59 <p>and in subsequent drafts of the <em>IETF</em>:</p>
60
61 <ul>
62 <li><a href=
63 "https://web.archive.org/web/20130116065936/http://ftp.ics.uci.edu/pub/ietf/uri/">
64 Uniform Resource Identifiers (URI) Working Group</a></li>
65 </ul>
66
67 <p>When entering a URL on the command line to be used as the
68 <em>startfile</em>, or at the prompt for a
69 &ldquo;<em>g</em>&rdquo;oto entry, a partial host field can be
70 used and the scheme field can be omitted if the scheme and fully
71 qualified domain name can be constructed internally by using the
72 URL_DOMAIN_PREFIXES and URL_DOMAIN_SUFFIXES definitions in the
73 <strong>Lynx</strong> configuration file. See the explanation of
74 those definitions and their use in your <em>lynx.cfg</em>.</p>
75
76 <p>For example, <em>wfbr</em> will be treated as
77 <em>http://www.wfbr.edu/</em>, and <em>wfbr/dir/lynx</em> will be
78 treated as <em>http://www.wfbr.edu/dir/lynx</em>, but
79 <em>gopher.wfbr.edu/11/_fileserv/_lynx</em> will be treated as
80 <em>gopher://gopher.wfbr.edu/11/_fileserv/_lynx</em>.</p>
81
82 <p>For files or directories on the local host, a tilde
83 (<em>~</em>) is expanded to the path of the account's login
84 directory, e.g., <em>~/foo</em> will be expanded to
85 <em>file://localhost/your/login/directory/foo</em>. The tilde
86 expansion is done homologously on Unix and VMS.</p>
87
88 <p>On VMS, <strong>Lynx</strong> also will expand any file or
89 directory spec recognizable to DCL into a valid URL, e.g.,
90 <em>[]</em> will be expanded to
91 <em>file://localhost/current/default/directory</em>.</p>
92
93 <p>These expansions are <em>SOLELY</em> for <em>startfile</em> or
94 &ldquo;<em>g</em>&rdquo;oto entries! Any partial or relative URLs
95 within HTML documents are resolved according to the rules
96 specified in RFC1808 and subsequent IETF drafts.</p>
97
98 <hr>
99
100 <h2><a name="http_url" id="http_url">The <em>http</em> and
101 <em>https</em> URLs:</a></h2>
102
103 <p><strong>Lynx</strong> handles http URLs exactly as specified
104 in RFC1738. The format is:</p>
105
106 <pre>
107 <em>http://host:port/path?searchpart#fragment</em>
108 </pre>
109 <p>where <em>:port</em> is optional and defaults to <em>:80</em>,
110 <em>/path</em> if present is a slash-separated series of symbolic
111 elements, and <em>?searchpart</em> if present is the query for an
112 ISINDEX search or the content of a FORM with METHOD="GET". The
113 <em>#fragment</em> field if present indicates a location in the
114 document to seek for display, based on a NAME-ed anchor or an ID
115 attribute within the document, and is technically an instruction
116 rather than part of the URL. <strong>Lynx</strong> will treat ID
117 attributes as NAME-ed anchors for all tags in the BODY of a
118 document which can correspond to positions in the rendering of
119 the document.</p>
120
121 <p>The https URL has the same format, but the default port is
122 <em>:443</em>.</p>
123
124 <p><strong>Lynx</strong> relies for https support on external
125 libraries (OpenSSL or GnuTLS) whose capabilities have evolved
126 over time. In turn, those libraries may depend upon external
127 resources for verifying SSL certificates. For instance,
128 certification revocation may be provided via the Online
129 Certificate Status Protocol (OCSP) which is an external service.
130 Without this facility, <strong>Lynx</strong> may not warn about
131 websites using revoked SSL certificates.</p>
132
133 <hr>
134
135 <h2><a name="telnet_url" id="telnet_url">The <em>telnet</em>,
136 <em>tn3270</em>, and <em>rlogin</em> URLs:</a></h2>
137
138 <p>A <em>telnet</em> URL generally results in
139 <strong>Lynx</strong> spawning a telnet session.
140 <strong>Lynx</strong> implements the complete telnet URL scheme,
141 i.e.:</p>
142
143 <pre>
144 <em>telnet://user:password@host:port</em>
145 </pre>
146 <p>The <em>user</em> and/or <em>:password</em> fields may be
147 omitted, and the <em>@</em> should be omitted if neither is
148 present. The port defaults to <em>:23</em> when omitted in the
149 URL.</p>
150
151 <p>A <em>tn3270</em> or <em>rlogin</em> URL is specified
152 equivalently, and similarly spawns a tn3270 or rlogin session.
153 The actual behavior is dependent on the TCP-IP software installed
154 on the local and target hosts.</p>
155
156 <p>It is unwise to include the <em>:password</em> field except
157 for URLs which point to anonymous or other public access
158 accounts, and for most TCP-IP software you will be prompted for a
159 password whether or not one was included in the URL.</p>
160
161 <hr>
162
163 <h2><a name="gopher_url" id="gopher_url">The <em>gopher</em>
164 URL:</a></h2>
165
166 <p>The gopher URL takes the form:</p>
167
168 <pre>
169 <em>gopher://host:port/gopher-path</em>
170 </pre>
171 <p>where <em>:port</em> is optional and defaults to <em>:70</em>,
172 and the <em>/gopher-path</em> is opaque (not fully equivalent to
173 the slash-separated series of symbolic elements of http paths) as
174 explained in RFC1738. Typically, the gopher-path consists of a
175 <a href=
176 "keystrokes/gopher_types_help.html"><em>gophertype</em></a>
177 indicating the file or service type (e.g., <em>0</em> or
178 <em>I</em> for plain text or an image, respectively, <em>7</em>
179 for a search, or <em>1</em> for a directory), followed by a
180 platform-specific <em>selector</em>. Any reserved characters in
181 the selector should be hex escaped (<em>%hh</em>), including
182 slashes, although hex escaping of slashes is not required by
183 <strong>Lynx</strong> in gopher URLs.</p>
184
185 <p><strong>Lynx</strong> does not overtly support the gopher+
186 protocol, and does not represent itself as gopher+ capable when
187 communicating with gopher servers. <strong>Lynx</strong> might
188 transmit any (hex-escaped-tab-separated) extended gopher+ fields
189 in a URL if an author included them in a document, but is likely
190 to mishandle what the gopher server returns in such cases, and
191 would not generate and transmit them itself. For pre-formed URLs
192 to submit gopher searches, it may be better to use a <em>?</em>
193 rather than hex-escaped tab (<em>%09</em>) as the separator for
194 the <em>searchpart</em> in the <em>selector</em>, e.g.:<br>
195 <em>gopher://gopher.wfbr.edu/77/_shell/search.shell%20/_shell/walker?lynx*</em>
196 <strong>Lynx</strong> will handle the <em>%09</em> if you use
197 that instead of <em>?</em>, but other WWW clients may mishandle
198 it.</p>
199
200 <p>For the <em>gophertype</em> which signifies HTML (<em>h</em>),
201 if the <em>selector</em> begins with <em>GET%20/</em>
202 <strong>Lynx</strong> will convert the gopher URL to an http URL,
203 e.g.:<br></p>
204
205 <pre>
206 <em>gopher://www.wfbr.edu:80/hGET%20/</em>
207 </pre>
208 <p>will become:<br></p>
209
210 <pre>
211 <em>http://www.wfbr.edu/</em>
212 </pre>
213 <p>The port field will be retained if it is not <em>:80</em>, and
214 will default to <em>:70</em> if it was defaulted originally.
215 These conventions were adopted during development of the
216 University of Minnesota gopher software to facilitate the
217 offering of links to MIME-capable http servers in the listings
218 returned by gopher servers, but should be considered Lynxisms and
219 UMN Gopherisms.</p>
220
221 <hr>
222
223 <h2><a name="file_url" id="file_url">The <em>file</em> URL:</a></h2>
224
225 <p>The file URL is used to retrieve files or generate a directory
226 listing on the local host. The host field can be
227 <em>localhost</em> or a domain name for the local host:<br></p>
228
229 <pre>
230 <em>file://localhost/path</em>
231 </pre>
232 <p>If you do not use <em>localhost</em> or a domain name for the
233 local host, <strong>Lynx</strong> will substitute <em>ftp://</em>
234 for <em>file://</em> and treat it as an ftp URL.</p>
235
236 <p>The <em>/path</em> is treated as originating at the root,
237 unless you include a tilde (<em>~</em>), e.g.:</p>
238
239 <pre>
240 <em>file://localhost/~/foo</em> will be converted to:
241 <em>file://localhost/your/login/directory/foo</em>
242 </pre>
243 <p>The latter feature is a Lynxism, is done homologously on Unix
244 and VMS, and should be used ONLY in local documents intended for
245 <strong>Lynx</strong>.</p>
246
247 <p>On VMS, the first element of the path, if not a tilde, is
248 assumed to be a device, e.g.:</p>
249
250 <pre>
251 <em>file://localhost/www_root/directory/filename.suffix</em>
252 </pre>
253 <p>should be used for:
254 <em>www_root:[directory]filename.suffix</em><br>
255 If you are unsure how to specify a file URL in local documents on
256 VMS, invoke <strong>Lynx</strong> with the desired file or
257 directory as the <em>startfile</em> using any spec acceptable to
258 DCL, and then use the <em>showinfo</em> command (<em>=</em>) to
259 see the file URL which <strong>Lynx</strong> created for it.</p>
260
261 <hr>
262
263 <h2><a name="ftp_url" id="ftp_url">The <em>ftp</em> URL:</a></h2>
264
265 <p>The ftp URL has the general format:</p>
266
267 <pre>
268 <em>ftp://host:port/path;type=[D,I, or A]</em>
269 <em>ftp://username@host:port/path;type=[D,I, or A]</em>
270 </pre>
271 <p>The default port is <em>:21</em> and the default
272 <em>username</em> is <em>anonymous</em>. If <em>username</em> is
273 included, <strong>Lynx</strong> will prompt you for the password.
274 For anonymous ftp, <strong>Lynx</strong> uses your
275 <em>personal_mail_address</em> (user@host) as the
276 <em>password</em> if it has been defined via the
277 &ldquo;<em>o</em>&rdquo;ptions menu. Otherwise,
278 <strong>Lynx</strong> uses the dummy password <em>WWWUser</em>.
279 (A password can also be embedded in the URL, by replacing
280 <em>username</em> with <em>username:password</em>. This is
281 strongly discouraged for &ldquo;real&rdquo; passwords that must
282 be kept secret, since URLs with the completely unencrypted
283 <em>password</em> may show up on the screen, in HISTORY and LIST
284 pages etc., and may even become visible to remote sites for
285 example through Referer headers.) Do not include the <em>@</em>
286 if neither <em>username</em> nor <em>:password</em> is
287 included.</p>
288
289 <p>The <em>;type=</em> parameter can be used with value
290 <em>D</em>, <em>I</em>, or <em>A</em> to force handling of the
291 URL as, respectively, a directory listing, binary file, or ASCII
292 file. The <strong>Lynx</strong> ftp gateway normally determines
293 this itself, but the parameter can be used if the internal
294 procedure draws an incorrect inference about the nature of the
295 ftp URL.</p>
296
297 <p>The <em>/path</em> is treated according to RFC1738 for VMS and
298 VM/CMS ftp servers. The lead slash (<em>/</em>) is treated purely
299 as a separator, not as a designator for the root, and the
300 <em>path</em> string if present is treated as in or under the
301 login directory. For VMS ftp servers, if you wish to have the
302 first element treated as a device rather than file or
303 subdirectory name, begin it with a hex-escaped slash
304 (<em>%2f</em>), e.g.:</p>
305
306 <pre>
307 <em>ftp://user@myhost/%2fsys$common/syshlp</em>
308 </pre>
309 <p>can be used for a listing of sys$common:[syshlp]<br>
310 Also, on VM/CMS ftp servers, if the <em>path</em> string begins
311 with <em>vmsysu%3a</em> it receives special handling as an SFS
312 path, e.g.:</p>
313
314 <pre>
315 <em>ftp://ubvm.cc.buffalo.edu/vmsysu%3alistserv.webshare</em>
316 </pre>
317 <p>For Unix and Unix-emulation ftp servers, RFC1738 is not
318 respected and the lead slash is treated as the root, i.e., the
319 <em>/path</em> is handled equivalently to that in file URLs. The
320 distinction is irrelevant for anonymous ftp, but matters when
321 using ftp for non-anonymous accounts. If you are using ftp with a
322 Unix server and do wish to get a listing of the login directory
323 or have the <em>path</em> string treated as a file or path under
324 the login directory, include a tilde (<em>~</em>) as for <a href=
325 "#file_url">file</a> URLs, e.g.:</p>
326
327 <pre>
328 <em>ftp://user@myhost/~</em>
329 </pre>
330
331 <hr>
332
333 <h2><a name="wais_url" id="wais_url">The <em>wais</em> URL:</a></h2>
334
335 <p>The wais URL is used to retrieve resources using the Wide Area
336 Information System protocol. The format is:</p>
337
338 <pre>
339 <em>wais://host:port/database</em>
340 <em>wais://host:port/database?wais_query</em>
341 <em>wais://host:port/database/wais_type/wais_path</em>
342 </pre>
343 <p>where <em>:port</em> defaults to <em>:210</em></p>
344
345 <p>Direct wais support is built into <strong>Lynx</strong> for
346 VMS, and can be compiled into <strong>Lynx</strong> on Unix.</p>
347
348 <p>If only a <em>database</em> is indicated in the URL,
349 <strong>Lynx</strong> returns an ISINDEX cover page for searching
350 that <em>database</em>, and will submit your search with the
351 <em>wais_query</em> appended. <strong>Lynx</strong> will convert
352 the server's reply into a hit list with URLs that include the
353 <em>wais_type</em> and <em>wais_path</em> for retrieving items
354 from the hit list.</p>
355
356 <hr>
357
358 <h2><a name="news_url" id="news_url">The <em>news</em>,
359 <em>nntp</em>, and <em>snews</em> URLs:</a></h2>
360
361 <p>The news and nntp URLs are handled by <strong>Lynx</strong> as
362 specified in RFC1738, but for compatibility with other clients,
363 <strong>Lynx</strong> allows inclusion of host and port fields in
364 news URLs, which properly should be used <em>only</em> in nntp
365 and snews URLs. If not included in news URLs,
366 <strong>Lynx</strong> will use the nntp server pointed to by the
367 NNTPSERVER environment variable or configuration symbol (see
368 lynx.cfg), with default port <em>:119</em>. A host field must be
369 included in nntp URLs, and the port field is optional with the
370 same default.</p>
371
372 <p>If the URL requires authentication, <strong>Lynx</strong> will
373 prompt you for the username and password. These are cached during
374 a session, for reuse on the same host. If $HOME/.newsauth exists,
375 <strong>Lynx</strong> initializes its cache from this file. The
376 .newsauth file contents are one line per entry: hostname,
377 password and username (in that order) separated by a space.</p>
378
379 <p>The formats are:<br></p>
380
381 <pre>
382 <em>news:newsgroup</em> (retrieves list of messages in newsgroup)
383 <em>news:messageID</em> (retrieves the message)
384 <em>news:*</em> (retrieves list of all available newsgroups)
385 <em>nntp://host:port/newsgroup</em>
386 <em>nntp://host:port/messageID</em>
387 <em>nntp://host:port/*</em>
388 </pre>
389 <p>(snews same as nntp, but the default port is
390 <em>:563</em>)</p>
391
392 <p>The <em>messageID</em> is the message's unique identifier,
393 consisting of an identification string and the host of origin for
394 the message (<em>ident_string@origin_host</em>).</p>
395
396 <p><strong>Lynx</strong> also supports wildcarding via an
397 asterisk for listings of news hierarchies or sub-hierarchies,
398 e.g.:</p>
399
400 <pre>
401 <em>news:comp.infosystems.*</em>
402 <em>nntp://host:port/comp.infosystems.*</em>
403 </pre>
404 <p>(snews same as nntp, but the default port is
405 <em>:563</em>)<br>
406 This is not in RFC1738 and may not be supported by all other
407 clients.</p>
408
409 <p><strong>Lynx</strong> allows you both to <em>reply</em> to the
410 author of a news message via email, and, if news posting has been
411 enabled, to send a <em>followup</em> message to the newsgroup
412 (see <a href="#newspost_url">newspost, newsreply, snewspost,
413 snewsreply</a>).</p>
414
415 <p><strong>Lynx</strong> converts any strings in news messages
416 which appear to be a URL with a supported scheme into a link for
417 accessing that URL.</p>
418
419 <p><strong>Lynx</strong> also supports the newsgroup and message
420 number URL scheme:<br></p>
421
422 <pre>
423 <em>news:newsgroup/startNo-endNo</em> (lists message range in newsgroup)
424 <em>news:newsgroup/messageNo</em> (retrieves the message by number)
425 <em>nntp://host:port/newsgroup/startNo-endNo</em>
426 <em>nntp://host:port/newsgroup/messageNo</em>
427 </pre>
428 <p>(snews same as nntp, but the default port is
429 <em>:563</em>)<br>
430 Use of this scheme is not recommended, because the message
431 numbers are specific to each nntp server, unlike the unique
432 identifiers for news messages.</p>
433
434 <hr>
435
436 <h2><a name="newspost_url" id="newspost_url">The
437 <em>newspost</em>, <em>newsreply</em>, <em>snewspost</em>, and
438 <em>snewsreply</em> URLs:</a></h2>
439
440 <p>When <strong>Lynx</strong> receives group listings or articles
441 via <em>news</em>, <em>nntp</em> or <em>snews</em> URLs, it also
442 checks whether the nntp server supports posting from the
443 <strong>Lynx</strong> user's site, and if so, includes links for
444 posting new messages to that server, or for posting followups
445 (replies) to previously posted messages. RFC1738, and IETF URL
446 drafts through this release of <strong>Lynx</strong>, do not
447 include any schemes for posting to news groups.
448 <strong>Lynx</strong> has long supported newspost and newreply
449 URL schemes for posting new messages or sending followups,
450 respectively, to standard nntp servers, with default port
451 <em>:119</em>. <strong>Lynx</strong> now also supports homologous
452 snewspost and snewsreply URLs for use with SSL capable nntp
453 servers.</p>
454
455 <p>The formats are:</p>
456
457 <pre>
458 <em>newspost://host:port/newsgroup(s)</em> (post a new message)
459 <em>newsreply://host:port/newsgroup(s)</em> (post a followup message)
460 </pre>
461 <p>(snewspost and snewsreply have the same formats, but the
462 default port is <em>:563</em>)</p>
463
464 <p>If the host field is omitted, it defaults to that pointed to
465 by the NNTPSERVER configuration or environmental variable.
466 Inclusion of at least one newsgroup in the URL is required, and
467 additional groups can be specified as a comma-separated list.
468 Wildcarding of newsgroup names is not supported for these URLs.
469 For newsreply and snewsreply URLs, if an external editor has been
470 defined via the <em>Options Menu</em>, the user is offered an
471 option to include the currently displayed document, which
472 presumably is a news article with a <em>followup</em> link that
473 was activated, and if confirmed, each line of that document is
474 prefixed with a right-angle-bracket. The user is expected to edit
475 such an inclusion so that only the passages relevant to the
476 followup message are retained.</p>
477
478 <p>These URLs can be used as command line startfiles (in which
479 case, <strong>Lynx</strong> will exit after posting the message,
480 and the newreply or snewsreply URLs degrade to newspost or
481 snewpost URLs, respectively). They also can be used as HREF
482 attribute values in any HTML document homologously to <a href=
483 "#mailto_url">mailto</a> URLs, with the qualification that they
484 presently are supported only by <strong>Lynx</strong>.</p>
485
486 <hr>
487
488 <h2><a name="mailto_url" id="mailto_url">The <em>mailto</em>
489 URL:</a></h2>
490
491 <p>The mailto URL is used to provide links that when activated
492 can be used to send a comment or the content of a FORM to an
493 Internet email address (user@host). The format is:</p>
494
495 <pre>
496 <em>mailto:user@host</em>
497 </pre>
498 <p>The description of the mailto URL in RFC1738 has been
499 interpreted by some as allowing only a single recipient, but
500 <strong>Lynx</strong> invented the mailto URL, has always
501 supported a series of user@host addresses as a comma-separated
502 list, and still does. For compatibility with Explorer,
503 <strong>Lynx</strong> also accepts a semi-colon-separated
504 list.</p>
505
506 <p>For compatibility with Netscape, <strong>Lynx</strong> parses
507 any <em>?subject=The%20Subject</em> appended to the URL, trims
508 the URL at the <em>?</em>, and uses the value as the default
509 Subject: for the message or FORM content mailing. This is not
510 recommended practice. The preferred way to indicate the default
511 Subject: for a LINK or Anchor with a mailto HREF, or a FORM with
512 a mailto ACTION, is via a TITLE attribute with the subject string
513 as its value, e.g.:</p>
514
515 <pre>
516 <em>&lt;LINK REL="author"
517 HREF="mailto:me@myhost,her@herhost" TITLE="The Subject"&gt;</em>
518
519 <em>&lt;A HREF="mailto:user@host" TITLE="The Subject"&gt;...&lt;/A&gt;</em>
520
521 <em>&lt;FORM METHOD="post" ENCTYPE="text/plain"
522 ACTION="mailto:WebMaster@host" TITLE="The Subject"&gt;
523 ...
524 &lt;/FORM&gt;</em>
525 </pre>
526 <p>Note that a TITLE attribute for FORM is now included in the
527 HTML specifications. Some clients use a SUBJECT attribute for
528 this purpose in FORM tags, and <strong>Lynx</strong> recognizes
529 that as a synonym for TITLE.</p>
530
531 <p><strong>Lynx</strong> also will process any
532 <em>to=address(es)</em>, <em>cc=address(es)</em>,
533 <em>keywords=word_list</em> and/or <em>body=message</em> fields
534 in <em>?searchpart</em> tack-ons to mailto URLs. The <em>to</em>
535 and/or <em>cc</em> values can be single addresses, or comma- or
536 semi-colon-separated lists of addresses. All addresses, and any
537 <em>body</em> values, will be offered for approval by the user
538 before proceeding with a mailing. Any other name=value pairs in
539 the <em>?searchpart</em> will be ignored. Also, if the mailto URL
540 is the ACTION for a FORM, any <em>body</em> in a
541 <em>?searchpart</em> tack-on will be ignored, because the body of
542 the mailing must be constructed solely from the the FORM's
543 content. <strong>Lynx</strong> expects multiple name=value pairs
544 in a <em>?searchpart</em> tack-on to be separated by ampersands,
545 as in the original Netscape implementation, and in an equally
546 ill-advised IETF draft of that implementation (<a href=
547 "ftp://ftp.isi.edu/internet-drafts/draft-hoffman-mailto-url-03.txt">draft-hoffman-mailto-url-03.txt</a>).
548 These should be represented as entities (<em>&amp;amp;</em>) in
549 the HTML markup. This functionality is generally desired, but the
550 IETF backward compatibility principal normally would lead to a
551 new scheme being used (e.g., <em>mail:</em>, or <em>smtp:</em>),
552 rather than breaking <em>mailto:</em> implementations.</p>
553
554 <p>If <em>ENCTYPE="text/plain"</em> is specified for a FORM with
555 a mailto ACTION, <strong>Lynx</strong> will not hex escape the
556 name=value pairs of the FORM's content, and will use physical
557 newlines instead of &ldquo;<em>&amp;</em>&rdquo; or
558 &ldquo;<em>;</em>&rdquo; to separate the pairs, so that the
559 content will be readable directly. Otherwise,
560 <strong>Lynx</strong> will mail the content with the default:</p>
561
562 <pre>
563 <em>ENCTYPE="application/x-www-form-urlencoded"</em> (&ldquo;<em>&amp;</em>&rdquo; separates pairs)
564 </pre>
565 <p>or:</p>
566
567 <pre>
568 <em>ENCTYPE="application/sgml-form-urlencoded"</em> (&ldquo;<em>;</em>&rdquo; separates pairs)
569 </pre>
570 <p>if the latter was indicated.</p>
571
572 <p>Note that when mailing FORM content <strong>Lynx</strong>
573 wraps any lines longer than 78 characters, to avoid buffer
574 overflows in mail software and to ensure reliable transmission
575 across gateways. If the ENCTYPE was not <em>text/plain</em>, any
576 script which decodes the mailed content should ignore the
577 physical newlines and recognize only hex escaped newline
578 characters as intended to be present in the decoded content.</p>
579
580 <p>If the mailto URL is not the ACTION for a FORM, and if an
581 external editor has been defined via the <em>Options Menu</em>,
582 the user is offered an option to include the currently displayed
583 document. If this option is accepted, each line of that document
584 is prefixed with a right-angle-bracket, and the prefixed
585 inclusion should be trimmed by the user to just those passages
586 relevant to the message which will be sent.</p>
587
588 <hr>
589
590 <h2><a name="finger_url" id="finger_url">The <em>finger</em>
591 URL:</a></h2>
592
593 <p><strong>Lynx</strong> has full support for the finger
594 protocol, but a format for finger URLs has not yet been adopted
595 by the IETF. The formats supported by <strong>Lynx</strong>
596 therefore include every possibility not inconsistent with
597 RFC1738, including:</p>
598
599 <pre>
600 finger://host finger://@host
601 finger://host/ finger://@host/
602 finger://host/%2fw finger://@host/w
603 finger://host/w finger://host/w/
604 finger://host/username[@host] finger://username@host
605 finger://host/username[@host]/ finger://username@host/
606 finger://host/w/username[@host] finger://username@host/w
607 finger://host/%2fw%20username[@host] finger://host/username[@host]/w
608 finger://host/w/username
609 </pre>
610 <p>Activating a finger URL will send a request to the finger
611 server via port 79 on the host specified. You can include
612 <em>:79</em> in the URL, but no other value is allowed. The
613 <em>/w</em> or <em>/%2fw</em> is used to request a full report
614 for finger servers which support it, and is not case sensitive
615 (i.e., can be <em>/W</em> or <em>/%2fW</em>). Any strings in the
616 report which appear to be a URL with a supported scheme will be
617 converted into a link for accessing that URL.</p>
618
619 <p>An alternative way to access finger servers is via gopher URLs
620 with port 79 and the plain text (<em>0</em>) <em>gophertype</em>
621 specified:<br>
622 <em>gopher://host:79/0</em><br>
623 <strong>Lynx</strong> will handle such URLs equivalently to overt
624 finger URLs, including creation of links for any strings which
625 appear to be supported URLs.</p>
626
627 <hr>
628
629 <h2><a name="cso_url" id="cso_url">The <em>cso</em> URL:</a></h2>
630
631 <p>The cso URL is intended to provide a gateway to CSO/PH (QI)
632 servers. The requests are made on port 105 by default
633 (<em>:105</em>), with the following overt cso URL format:<br></p>
634
635 <pre>
636 <em>cso://host</em>
637 </pre>
638 <p>You also can use a gopher URL format with port 105 and the CSO
639 (<em>2</em>) <em>gophertype</em> specified:</p>
640
641 <pre>
642 <em>gopher://host:105/2</em>
643 </pre>
644 <p><strong>Lynx</strong> will parse the stream returned by the
645 server for the above URLs and create a FORM for submitting
646 additional requests (searches) to the server. Any strings in the
647 reports returned for these requests (searches) which appear to be
648 a URL with a supported scheme will be converted into a link for
649 accessing that URL.</p>
650
651 <hr>
652
653 <h2><a name="bibp_url" id="bibp_url">The <em>bibp</em> URL:</a></h2>
654
655 <p><strong>Lynx</strong> provides built-in support for
656 bibliographic protocol (BibP). BibP links are links to published
657 works such as books or journal articles, without a predefined
658 server. BibP links are intended for resolution by a local bibhost
659 server (http://bibhost/) if it exists. Otherwise, resolution is
660 performed by a document-specified server or a known global
661 server.</p>
662
663 <h2><a name="exec_url" id="exec_url">The <em>lynxexec</em> and
664 <em>lynxprog</em> URLs:</a></h2>
665
666 <p>If execution of spawned commands has been enabled in your
667 <strong>Lynx</strong> image, the lynxexec and lynxprog URLs can
668 be used to execute arbitrary system commands or invoke system
669 utilities. Any system command and associated switches or
670 qualifiers can be used, with the syntax appropriate for a shell
671 running <strong>Lynx</strong> on Unix, or for DCL on VMS,
672 e.g.:</p>
673
674 <pre>
675 <em>lynxexec:dir/date/size foo:[blah]</em> (VMS)
676 <em>lynxexec:ls -l /foo/blah</em> (Unix)
677 <em>lynxprog:news</em>
678 </pre>
679 <p>(Note, however, that restrictions on acceptable commands or
680 utilities may be imposed by the system administrator.)</p>
681
682 <p>You optionally can include <em>//localhost/</em> in the URL,
683 between the scheme field and the command, but that is always
684 implied. The lynxexec and lynxprog URLs differ only in that with
685 lynxexec you are prompted to enter <em>RETURN</em> before
686 <strong>Lynx</strong> clears the screen and restores the
687 previously displayed document, so that you can read any screen
688 output generated by the spawned command, whereas no such pause is
689 imposed upon exit from the utility invoked via lynxprog.</p>
690
691 <p>These are Lynxisms and should be used only in local documents
692 intended solely for <strong>Lynx</strong>.</p>
693
694 <hr>
695
696 <h2><a name="cgi_url" id="cgi_url">The <em>lynxcgi</em> URL:</a></h2>
697
698 <p>The lynxcgi URL is implemented only on Unix, can be used as
699 the ACTION for a FORM, and if enabled in your
700 <strong>Lynx</strong> image has the format:</p>
701
702 <pre>
703 <em>lynxcgi://localhost/path_to_CGI_script</em>
704 </pre>
705 <p>where <em>//localhost</em> is optional and always implied; the
706 full path should be specified, as &ldquo;~&rdquo; is not
707 recognized; if the script is in the directory
708 <strong>Lynx</strong> was started from, the simple file name is
709 adequate. The output of the script should be text/html and is
710 rendered and displayed by <strong>Lynx</strong>. Restrictions on
711 use of lynxcgi and on acceptable paths can be imposed in
712 <em>userdefs.h</em> and <em>lynx.cfg</em>, qv.</p>
713
714 <p>This is a Lynxism and should be used only in local documents
715 intended solely for <strong>Lynx</strong>, or for limited local
716 testing of CGI scripts without an http server.</p>
717
718 <hr>
719
720 <h2><a name="ncftp_url" id="ncftp_url">The <em>NcFTP</em>
721 URL:</a></h2>
722
723 <p><strong>Lynx</strong> recognizes the NcFTP-style ftp URL,
724 e.g.,</p>
725
726 <pre>
727 <cite>ftpHost</cite>:<cite>fileSpecification</cite>
728 </pre>
729 <p>for example</p>
730
731 <pre>
732 <code>
733 ftp.gnu.org:/pub/gnu
734 </code>
735 </pre>
736
737 <hr>
738
739 <h2><a name="internal_url" id="internal_url">The <em>LYNXfoo</em>
740 internal URLs:</a></h2>
741
742 <p><strong>Lynx</strong> uses a variety of private URL schemes
743 for communication among its internal modules. They start with
744 uppercase letters <code>LYNX</code> by convention, although, as
745 input, URL schemes are recognized in a case-insensitive
746 manner.</p>
747
748 <p>As you discover what they are, and are tempted to use them
749 externally in documents, you should <em>resist</em> that
750 temptation:</p>
751
752 <ul>
753 <li>There already is too much browser-specific markup
754 around...</li>
755
756 <li>The schemes, or their meanings, may change between
757 <strong>Lynx</strong> versions.</li>
758
759 <li>Even if a scheme stays the same, some aspect of its
760 behavior may be modified without notice, or the context in
761 which it is allowed may change.</li>
762
763 <li>If it does not work as expected when used outside of the
764 intended purpose, do not expect anyone to "fix" it.</li>
765 </ul>
766
767 <p>For example, tempting though it might be, do not use
768 these:</p>
769
770 <pre>
771 <em>Return to your &lt;A HREF="LYNXHIST:0"&gt;Startfile&lt;/A&gt;</em>
772 <em>Review your &lt;A HREF="LYNXKEYMAP:"&gt;Keymap&lt;/A&gt;</em>
773 </pre>
774 <p>(No, they will not do any harm. Yes, they work. But do not
775 rely on it.)</p>
776
777 <p>If you must try one, the second is OK from the command
778 line:<br></p>
779
780 <pre>
781 <em>lynx LYNXKEYMAP:</em>
782 </pre>
783 <p>But within <strong>Lynx</strong>, use the
784 &ldquo;<em>K</em>&rdquo; keystroke command. Sometimes it may be
785 convenient to use a private scheme with
786 &ldquo;<em>g</em>&rdquo;oto, as in:</p>
787
788 <pre>
789 <em>g LYNXMESSAGES:</em>
790 <em>g LYNXCOMPILEOPTS:</em>
791 <em>g LYNXCFG:</em>
792 </pre>
793 <p>But again, there usually is a way in which those special pages
794 are meant to be reached that is more convenient.</p>
795 </body>
796 </html>