1.\"- 2.\" Copyright (c) 1998-2011 Dag-Erling Smørgrav 3.\" All rights reserved. 4.\" 5.\" Redistribution and use in source and binary forms, with or without 6.\" modification, are permitted provided that the following conditions 7.\" are met: 8.\" 1. Redistributions of source code must retain the above copyright 9.\" notice, this list of conditions and the following disclaimer. 10.\" 2. Redistributions in binary form must reproduce the above copyright 11.\" notice, this list of conditions and the following disclaimer in the 12.\" documentation and/or other materials provided with the distribution. 13.\" 14.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 15.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 16.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 17.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 18.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 19.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 20.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 21.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 22.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 23.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 24.\" SUCH DAMAGE. 25.\" 26.\" $FreeBSD$ 27.\" 28.Dd November 3, 2012 29.Dt FETCH 3 30.Os 31.Sh NAME 32.Nm fetchMakeURL , 33.Nm fetchParseURL , 34.Nm fetchFreeURL , 35.Nm fetchXGetURL , 36.Nm fetchGetURL , 37.Nm fetchPutURL , 38.Nm fetchStatURL , 39.Nm fetchListURL , 40.Nm fetchXGet , 41.Nm fetchGet , 42.Nm fetchPut , 43.Nm fetchStat , 44.Nm fetchList , 45.Nm fetchXGetFile , 46.Nm fetchGetFile , 47.Nm fetchPutFile , 48.Nm fetchStatFile , 49.Nm fetchListFile , 50.Nm fetchXGetHTTP , 51.Nm fetchGetHTTP , 52.Nm fetchPutHTTP , 53.Nm fetchStatHTTP , 54.Nm fetchListHTTP , 55.Nm fetchXGetFTP , 56.Nm fetchGetFTP , 57.Nm fetchPutFTP , 58.Nm fetchStatFTP , 59.Nm fetchListFTP 60.Nd file transfer functions 61.Sh LIBRARY 62.Lb libfetch 63.Sh SYNOPSIS 64.In sys/param.h 65.In stdio.h 66.In fetch.h 67.Ft struct url * 68.Fn fetchMakeURL "const char *scheme" "const char *host" "int port" "const char *doc" "const char *user" "const char *pwd" 69.Ft struct url * 70.Fn fetchParseURL "const char *URL" 71.Ft void 72.Fn fetchFreeURL "struct url *u" 73.Ft FILE * 74.Fn fetchXGetURL "const char *URL" "struct url_stat *us" "const char *flags" 75.Ft FILE * 76.Fn fetchGetURL "const char *URL" "const char *flags" 77.Ft FILE * 78.Fn fetchPutURL "const char *URL" "const char *flags" 79.Ft int 80.Fn fetchStatURL "const char *URL" "struct url_stat *us" "const char *flags" 81.Ft struct url_ent * 82.Fn fetchListURL "const char *URL" "const char *flags" 83.Ft FILE * 84.Fn fetchXGet "struct url *u" "struct url_stat *us" "const char *flags" 85.Ft FILE * 86.Fn fetchGet "struct url *u" "const char *flags" 87.Ft FILE * 88.Fn fetchPut "struct url *u" "const char *flags" 89.Ft int 90.Fn fetchStat "struct url *u" "struct url_stat *us" "const char *flags" 91.Ft struct url_ent * 92.Fn fetchList "struct url *u" "const char *flags" 93.Ft FILE * 94.Fn fetchXGetFile "struct url *u" "struct url_stat *us" "const char *flags" 95.Ft FILE * 96.Fn fetchGetFile "struct url *u" "const char *flags" 97.Ft FILE * 98.Fn fetchPutFile "struct url *u" "const char *flags" 99.Ft int 100.Fn fetchStatFile "struct url *u" "struct url_stat *us" "const char *flags" 101.Ft struct url_ent * 102.Fn fetchListFile "struct url *u" "const char *flags" 103.Ft FILE * 104.Fn fetchXGetHTTP "struct url *u" "struct url_stat *us" "const char *flags" 105.Ft FILE * 106.Fn fetchGetHTTP "struct url *u" "const char *flags" 107.Ft FILE * 108.Fn fetchPutHTTP "struct url *u" "const char *flags" 109.Ft int 110.Fn fetchStatHTTP "struct url *u" "struct url_stat *us" "const char *flags" 111.Ft struct url_ent * 112.Fn fetchListHTTP "struct url *u" "const char *flags" 113.Ft FILE * 114.Fn fetchXGetFTP "struct url *u" "struct url_stat *us" "const char *flags" 115.Ft FILE * 116.Fn fetchGetFTP "struct url *u" "const char *flags" 117.Ft FILE * 118.Fn fetchPutFTP "struct url *u" "const char *flags" 119.Ft int 120.Fn fetchStatFTP "struct url *u" "struct url_stat *us" "const char *flags" 121.Ft struct url_ent * 122.Fn fetchListFTP "struct url *u" "const char *flags" 123.Sh DESCRIPTION 124These functions implement a high-level library for retrieving and 125uploading files using Uniform Resource Locators (URLs). 126.Pp 127.Fn fetchParseURL 128takes a URL in the form of a null-terminated string and splits it into 129its components function according to the Common Internet Scheme Syntax 130detailed in RFC 1738. 131A regular expression which produces this syntax is: 132.Bd -literal 133 <scheme>:(//(<user>(:<pwd>)?@)?<host>(:<port>)?)?/(<document>)? 134.Ed 135.Pp 136If the URL does not seem to begin with a scheme name, the following 137syntax is assumed: 138.Bd -literal 139 ((<user>(:<pwd>)?@)?<host>(:<port>)?)?/(<document>)? 140.Ed 141.Pp 142Note that some components of the URL are not necessarily relevant to 143all URL schemes. 144For instance, the file scheme only needs the <scheme> and <document> 145components. 146.Pp 147.Fn fetchMakeURL 148and 149.Fn fetchParseURL 150return a pointer to a 151.Vt url 152structure, which is defined as follows in 153.In fetch.h : 154.Bd -literal 155#define URL_SCHEMELEN 16 156#define URL_USERLEN 256 157#define URL_PWDLEN 256 158 159struct url { 160 char scheme[URL_SCHEMELEN+1]; 161 char user[URL_USERLEN+1]; 162 char pwd[URL_PWDLEN+1]; 163 char host[MAXHOSTNAMELEN+1]; 164 int port; 165 char *doc; 166 off_t offset; 167 size_t length; 168 time_t ims_time; 169}; 170.Ed 171.Pp 172The 173.Va ims_time 174field stores the time value for 175.Li If-Modified-Since 176HTTP requests. 177.Pp 178The pointer returned by 179.Fn fetchMakeURL 180or 181.Fn fetchParseURL 182should be freed using 183.Fn fetchFreeURL . 184.Pp 185.Fn fetchXGetURL , 186.Fn fetchGetURL , 187and 188.Fn fetchPutURL 189constitute the recommended interface to the 190.Nm fetch 191library. 192They examine the URL passed to them to determine the transfer 193method, and call the appropriate lower-level functions to perform the 194actual transfer. 195.Fn fetchXGetURL 196also returns the remote document's metadata in the 197.Vt url_stat 198structure pointed to by the 199.Fa us 200argument. 201.Pp 202The 203.Fa flags 204argument is a string of characters which specify transfer options. 205The 206meaning of the individual flags is scheme-dependent, and is detailed 207in the appropriate section below. 208.Pp 209.Fn fetchStatURL 210attempts to obtain the requested document's metadata and fill in the 211structure pointed to by its second argument. 212The 213.Vt url_stat 214structure is defined as follows in 215.In fetch.h : 216.Bd -literal 217struct url_stat { 218 off_t size; 219 time_t atime; 220 time_t mtime; 221}; 222.Ed 223.Pp 224If the size could not be obtained from the server, the 225.Fa size 226field is set to -1. 227If the modification time could not be obtained from the server, the 228.Fa mtime 229field is set to the epoch. 230If the access time could not be obtained from the server, the 231.Fa atime 232field is set to the modification time. 233.Pp 234.Fn fetchListURL 235attempts to list the contents of the directory pointed to by the URL 236provided. 237If successful, it returns a malloced array of 238.Vt url_ent 239structures. 240The 241.Vt url_ent 242structure is defined as follows in 243.In fetch.h : 244.Bd -literal 245struct url_ent { 246 char name[PATH_MAX]; 247 struct url_stat stat; 248}; 249.Ed 250.Pp 251The list is terminated by an entry with an empty name. 252.Pp 253The pointer returned by 254.Fn fetchListURL 255should be freed using 256.Fn free . 257.Pp 258.Fn fetchXGet , 259.Fn fetchGet , 260.Fn fetchPut 261and 262.Fn fetchStat 263are similar to 264.Fn fetchXGetURL , 265.Fn fetchGetURL , 266.Fn fetchPutURL 267and 268.Fn fetchStatURL , 269except that they expect a pre-parsed URL in the form of a pointer to 270a 271.Vt struct url 272rather than a string. 273.Pp 274All of the 275.Fn fetchXGetXXX , 276.Fn fetchGetXXX 277and 278.Fn fetchPutXXX 279functions return a pointer to a stream which can be used to read or 280write data from or to the requested document, respectively. 281Note that 282although the implementation details of the individual access methods 283vary, it can generally be assumed that a stream returned by one of the 284.Fn fetchXGetXXX 285or 286.Fn fetchGetXXX 287functions is read-only, and that a stream returned by one of the 288.Fn fetchPutXXX 289functions is write-only. 290.Sh FILE SCHEME 291.Fn fetchXGetFile , 292.Fn fetchGetFile 293and 294.Fn fetchPutFile 295provide access to documents which are files in a locally mounted file 296system. 297Only the <document> component of the URL is used. 298.Pp 299.Fn fetchXGetFile 300and 301.Fn fetchGetFile 302do not accept any flags. 303.Pp 304.Fn fetchPutFile 305accepts the 306.Ql a 307(append to file) flag. 308If that flag is specified, the data written to 309the stream returned by 310.Fn fetchPutFile 311will be appended to the previous contents of the file, instead of 312replacing them. 313.Sh FTP SCHEME 314.Fn fetchXGetFTP , 315.Fn fetchGetFTP 316and 317.Fn fetchPutFTP 318implement the FTP protocol as described in RFC 959. 319.Pp 320If the 321.Ql P 322(not passive) flag is specified, an active (rather than passive) 323connection will be attempted. 324.Pp 325The 326.Ql p 327flag is supported for compatibility with earlier versions where active 328connections were the default. 329It has precedence over the 330.Ql P 331flag, so if both are specified, 332.Nm 333will use a passive connection. 334.Pp 335If the 336.Ql l 337(low) flag is specified, data sockets will be allocated in the low (or 338default) port range instead of the high port range (see 339.Xr ip 4 ) . 340.Pp 341If the 342.Ql d 343(direct) flag is specified, 344.Fn fetchXGetFTP , 345.Fn fetchGetFTP 346and 347.Fn fetchPutFTP 348will use a direct connection even if a proxy server is defined. 349.Pp 350If no user name or password is given, the 351.Nm fetch 352library will attempt an anonymous login, with user name "anonymous" 353and password "anonymous@<hostname>". 354.Sh HTTP SCHEME 355The 356.Fn fetchXGetHTTP , 357.Fn fetchGetHTTP 358and 359.Fn fetchPutHTTP 360functions implement the HTTP/1.1 protocol. 361With a little luck, there is 362even a chance that they comply with RFC 2616 and RFC 2617. 363.Pp 364If the 365.Ql d 366(direct) flag is specified, 367.Fn fetchXGetHTTP , 368.Fn fetchGetHTTP 369and 370.Fn fetchPutHTTP 371will use a direct connection even if a proxy server is defined. 372.Pp 373If the 374.Ql i 375(if-modified-since) flag is specified, and 376the 377.Va ims_time 378field is set in 379.Vt "struct url" , 380then 381.Fn fetchXGetHTTP 382and 383.Fn fetchGetHTTP 384will send a conditional 385.Li If-Modified-Since 386HTTP header to only fetch the content if it is newer than 387.Va ims_time . 388.Pp 389Since there seems to be no good way of implementing the HTTP PUT 390method in a manner consistent with the rest of the 391.Nm fetch 392library, 393.Fn fetchPutHTTP 394is currently unimplemented. 395.Sh AUTHENTICATION 396Apart from setting the appropriate environment variables and 397specifying the user name and password in the URL or the 398.Vt struct url , 399the calling program has the option of defining an authentication 400function with the following prototype: 401.Pp 402.Ft int 403.Fn myAuthMethod "struct url *u" 404.Pp 405The callback function should fill in the 406.Fa user 407and 408.Fa pwd 409fields in the provided 410.Vt struct url 411and return 0 on success, or any other value to indicate failure. 412.Pp 413To register the authentication callback, simply set 414.Va fetchAuthMethod 415to point at it. 416The callback will be used whenever a site requires authentication and 417the appropriate environment variables are not set. 418.Pp 419This interface is experimental and may be subject to change. 420.Sh RETURN VALUES 421.Fn fetchParseURL 422returns a pointer to a 423.Vt struct url 424containing the individual components of the URL. 425If it is 426unable to allocate memory, or the URL is syntactically incorrect, 427.Fn fetchParseURL 428returns a NULL pointer. 429.Pp 430The 431.Fn fetchStat 432functions return 0 on success and -1 on failure. 433.Pp 434All other functions return a stream pointer which may be used to 435access the requested document, or NULL if an error occurred. 436.Pp 437The following error codes are defined in 438.In fetch.h : 439.Bl -tag -width 18n 440.It Bq Er FETCH_ABORT 441Operation aborted 442.It Bq Er FETCH_AUTH 443Authentication failed 444.It Bq Er FETCH_DOWN 445Service unavailable 446.It Bq Er FETCH_EXISTS 447File exists 448.It Bq Er FETCH_FULL 449File system full 450.It Bq Er FETCH_INFO 451Informational response 452.It Bq Er FETCH_MEMORY 453Insufficient memory 454.It Bq Er FETCH_MOVED 455File has moved 456.It Bq Er FETCH_NETWORK 457Network error 458.It Bq Er FETCH_OK 459No error 460.It Bq Er FETCH_PROTO 461Protocol error 462.It Bq Er FETCH_RESOLV 463Resolver error 464.It Bq Er FETCH_SERVER 465Server error 466.It Bq Er FETCH_TEMP 467Temporary error 468.It Bq Er FETCH_TIMEOUT 469Operation timed out 470.It Bq Er FETCH_UNAVAIL 471File is not available 472.It Bq Er FETCH_UNKNOWN 473Unknown error 474.It Bq Er FETCH_URL 475Invalid URL 476.El 477.Pp 478The accompanying error message includes a protocol-specific error code 479and message, e.g.\& "File is not available (404 Not Found)" 480.Sh ENVIRONMENT 481.Bl -tag -width ".Ev FETCH_BIND_ADDRESS" 482.It Ev FETCH_BIND_ADDRESS 483Specifies a hostname or IP address to which sockets used for outgoing 484connections will be bound. 485.It Ev FTP_LOGIN 486Default FTP login if none was provided in the URL. 487.It Ev FTP_PASSIVE_MODE 488If set to 489.Ql no , 490forces the FTP code to use active mode. 491If set to any other value, forces passive mode even if the application 492requested active mode. 493.It Ev FTP_PASSWORD 494Default FTP password if the remote server requests one and none was 495provided in the URL. 496.It Ev FTP_PROXY 497URL of the proxy to use for FTP requests. 498The document part is ignored. 499FTP and HTTP proxies are supported; if no scheme is specified, FTP is 500assumed. 501If the proxy is an FTP proxy, 502.Nm libfetch 503will send 504.Ql user@host 505as user name to the proxy, where 506.Ql user 507is the real user name, and 508.Ql host 509is the name of the FTP server. 510.Pp 511If this variable is set to an empty string, no proxy will be used for 512FTP requests, even if the 513.Ev HTTP_PROXY 514variable is set. 515.It Ev ftp_proxy 516Same as 517.Ev FTP_PROXY , 518for compatibility. 519.It Ev HTTP_AUTH 520Specifies HTTP authorization parameters as a colon-separated list of 521items. 522The first and second item are the authorization scheme and realm 523respectively; further items are scheme-dependent. 524Currently, the 525.Dq basic 526and 527.Dq digest 528authorization methods are supported. 529.Pp 530Both methods require two parameters: the user name and 531password, in that order. 532.Pp 533This variable is only used if the server requires authorization and 534no user name or password was specified in the URL. 535.It Ev HTTP_PROXY 536URL of the proxy to use for HTTP requests. 537The document part is ignored. 538Only HTTP proxies are supported for HTTP requests. 539If no port number is specified, the default is 3128. 540.Pp 541Note that this proxy will also be used for FTP documents, unless the 542.Ev FTP_PROXY 543variable is set. 544.It Ev http_proxy 545Same as 546.Ev HTTP_PROXY , 547for compatibility. 548.It Ev HTTP_PROXY_AUTH 549Specifies authorization parameters for the HTTP proxy in the same 550format as the 551.Ev HTTP_AUTH 552variable. 553.Pp 554This variable is used if and only if connected to an HTTP proxy, and 555is ignored if a user and/or a password were specified in the proxy 556URL. 557.It Ev HTTP_REFERER 558Specifies the referrer URL to use for HTTP requests. 559If set to 560.Dq auto , 561the document URL will be used as referrer URL. 562.It Ev HTTP_USER_AGENT 563Specifies the User-Agent string to use for HTTP requests. 564This can be useful when working with HTTP origin or proxy servers that 565differentiate between user agents. 566.It Ev NETRC 567Specifies a file to use instead of 568.Pa ~/.netrc 569to look up login names and passwords for FTP sites. 570See 571.Xr ftp 1 572for a description of the file format. 573This feature is experimental. 574.It Ev NO_PROXY 575Either a single asterisk, which disables the use of proxies 576altogether, or a comma- or whitespace-separated list of hosts for 577which proxies should not be used. 578.It Ev no_proxy 579Same as 580.Ev NO_PROXY , 581for compatibility. 582.El 583.Sh EXAMPLES 584To access a proxy server on 585.Pa proxy.example.com 586port 8080, set the 587.Ev HTTP_PROXY 588environment variable in a manner similar to this: 589.Pp 590.Dl HTTP_PROXY=http://proxy.example.com:8080 591.Pp 592If the proxy server requires authentication, there are 593two options available for passing the authentication data. 594The first method is by using the proxy URL: 595.Pp 596.Dl HTTP_PROXY=http://<user>:<pwd>@proxy.example.com:8080 597.Pp 598The second method is by using the 599.Ev HTTP_PROXY_AUTH 600environment variable: 601.Bd -literal -offset indent 602HTTP_PROXY=http://proxy.example.com:8080 603HTTP_PROXY_AUTH=basic:*:<user>:<pwd> 604.Ed 605.Pp 606To disable the use of a proxy for an HTTP server running on the local 607host, define 608.Ev NO_PROXY 609as follows: 610.Bd -literal -offset indent 611NO_PROXY=localhost,127.0.0.1 612.Ed 613.Sh SEE ALSO 614.Xr fetch 1 , 615.Xr ftpio 3 , 616.Xr ip 4 617.Rs 618.%A J. Postel 619.%A J. K. Reynolds 620.%D October 1985 621.%B File Transfer Protocol 622.%O RFC 959 623.Re 624.Rs 625.%A P. Deutsch 626.%A A. Emtage 627.%A A. Marine. 628.%D May 1994 629.%T How to Use Anonymous FTP 630.%O RFC 1635 631.Re 632.Rs 633.%A T. Berners-Lee 634.%A L. Masinter 635.%A M. McCahill 636.%D December 1994 637.%T Uniform Resource Locators (URL) 638.%O RFC 1738 639.Re 640.Rs 641.%A R. Fielding 642.%A J. Gettys 643.%A J. Mogul 644.%A H. Frystyk 645.%A L. Masinter 646.%A P. Leach 647.%A T. Berners-Lee 648.%D January 1999 649.%B Hypertext Transfer Protocol -- HTTP/1.1 650.%O RFC 2616 651.Re 652.Rs 653.%A J. Franks 654.%A P. Hallam-Baker 655.%A J. Hostetler 656.%A S. Lawrence 657.%A P. Leach 658.%A A. Luotonen 659.%A L. Stewart 660.%D June 1999 661.%B HTTP Authentication: Basic and Digest Access Authentication 662.%O RFC 2617 663.Re 664.Sh HISTORY 665The 666.Nm fetch 667library first appeared in 668.Fx 3.0 . 669.Sh AUTHORS 670.An -nosplit 671The 672.Nm fetch 673library was mostly written by 674.An Dag-Erling Sm\(/orgrav Aq Mt des@FreeBSD.org 675with numerous suggestions and contributions from 676.An Jordan K. Hubbard Aq Mt jkh@FreeBSD.org , 677.An Eugene Skepner Aq Mt eu@qub.com , 678.An Hajimu Umemoto Aq Mt ume@FreeBSD.org , 679.An Henry Whincup Aq Mt henry@techiebod.com , 680.An Jukka A. Ukkonen Aq Mt jau@iki.fi , 681.An Jean-Fran\(,cois Dockes Aq Mt jf@dockes.org 682and others. 683It replaces the older 684.Nm ftpio 685library written by 686.An Poul-Henning Kamp Aq Mt phk@FreeBSD.org 687and 688.An Jordan K. Hubbard Aq Mt jkh@FreeBSD.org . 689.Pp 690This manual page was written by 691.An Dag-Erling Sm\(/orgrav Aq Mt des@FreeBSD.org . 692.Sh BUGS 693Some parts of the library are not yet implemented. 694The most notable 695examples of this are 696.Fn fetchPutHTTP , 697.Fn fetchListHTTP , 698.Fn fetchListFTP 699and FTP proxy support. 700.Pp 701There is no way to select a proxy at run-time other than setting the 702.Ev HTTP_PROXY 703or 704.Ev FTP_PROXY 705environment variables as appropriate. 706.Pp 707.Nm libfetch 708does not understand or obey 305 (Use Proxy) replies. 709.Pp 710Error numbers are unique only within a certain context; the error 711codes used for FTP and HTTP overlap, as do those used for resolver and 712system errors. 713For instance, error code 202 means "Command not 714implemented, superfluous at this site" in an FTP context and 715"Accepted" in an HTTP context. 716.Pp 717.Fn fetchStatFTP 718does not check that the result of an MDTM command is a valid date. 719.Pp 720The man page is incomplete, poorly written and produces badly 721formatted text. 722.Pp 723The error reporting mechanism is unsatisfactory. 724.Pp 725Some parts of the code are not fully reentrant. 726