1.\"- 2.\" Copyright (c) 1998-2004 Dag-Erling Co�dan Sm�rgrav 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 December 18, 2007 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 RFC1738. 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}; 169.Ed 170.Pp 171The pointer returned by 172.Fn fetchMakeURL 173or 174.Fn fetchParseURL 175should be freed using 176.Fn fetchFreeURL . 177.Pp 178.Fn fetchXGetURL , 179.Fn fetchGetURL , 180and 181.Fn fetchPutURL 182constitute the recommended interface to the 183.Nm fetch 184library. 185They examine the URL passed to them to determine the transfer 186method, and call the appropriate lower-level functions to perform the 187actual transfer. 188.Fn fetchXGetURL 189also returns the remote document's metadata in the 190.Vt url_stat 191structure pointed to by the 192.Fa us 193argument. 194.Pp 195The 196.Fa flags 197argument is a string of characters which specify transfer options. 198The 199meaning of the individual flags is scheme-dependent, and is detailed 200in the appropriate section below. 201.Pp 202.Fn fetchStatURL 203attempts to obtain the requested document's metadata and fill in the 204structure pointed to by its second argument. 205The 206.Vt url_stat 207structure is defined as follows in 208.In fetch.h : 209.Bd -literal 210struct url_stat { 211 off_t size; 212 time_t atime; 213 time_t mtime; 214}; 215.Ed 216.Pp 217If the size could not be obtained from the server, the 218.Fa size 219field is set to -1. 220If the modification time could not be obtained from the server, the 221.Fa mtime 222field is set to the epoch. 223If the access time could not be obtained from the server, the 224.Fa atime 225field is set to the modification time. 226.Pp 227.Fn fetchListURL 228attempts to list the contents of the directory pointed to by the URL 229provided. 230If successful, it returns a malloced array of 231.Vt url_ent 232structures. 233The 234.Vt url_ent 235structure is defined as follows in 236.In fetch.h : 237.Bd -literal 238struct url_ent { 239 char name[MAXPATHLEN]; 240 struct url_stat stat; 241}; 242.Ed 243.Pp 244The list is terminated by an entry with an empty name. 245.Pp 246The pointer returned by 247.Fn fetchListURL 248should be freed using 249.Fn free . 250.Pp 251.Fn fetchXGet , 252.Fn fetchGet , 253.Fn fetchPut 254and 255.Fn fetchStat 256are similar to 257.Fn fetchXGetURL , 258.Fn fetchGetURL , 259.Fn fetchPutURL 260and 261.Fn fetchStatURL , 262except that they expect a pre-parsed URL in the form of a pointer to 263a 264.Vt struct url 265rather than a string. 266.Pp 267All of the 268.Fn fetchXGetXXX , 269.Fn fetchGetXXX 270and 271.Fn fetchPutXXX 272functions return a pointer to a stream which can be used to read or 273write data from or to the requested document, respectively. 274Note that 275although the implementation details of the individual access methods 276vary, it can generally be assumed that a stream returned by one of the 277.Fn fetchXGetXXX 278or 279.Fn fetchGetXXX 280functions is read-only, and that a stream returned by one of the 281.Fn fetchPutXXX 282functions is write-only. 283.Sh FILE SCHEME 284.Fn fetchXGetFile , 285.Fn fetchGetFile 286and 287.Fn fetchPutFile 288provide access to documents which are files in a locally mounted file 289system. 290Only the <document> component of the URL is used. 291.Pp 292.Fn fetchXGetFile 293and 294.Fn fetchGetFile 295do not accept any flags. 296.Pp 297.Fn fetchPutFile 298accepts the 299.Ql a 300(append to file) flag. 301If that flag is specified, the data written to 302the stream returned by 303.Fn fetchPutFile 304will be appended to the previous contents of the file, instead of 305replacing them. 306.Sh FTP SCHEME 307.Fn fetchXGetFTP , 308.Fn fetchGetFTP 309and 310.Fn fetchPutFTP 311implement the FTP protocol as described in RFC959. 312.Pp 313If the 314.Ql p 315(passive) flag is specified, a passive (rather than active) connection 316will be attempted. 317.Pp 318If the 319.Ql l 320(low) flag is specified, data sockets will be allocated in the low (or 321default) port range instead of the high port range (see 322.Xr ip 4 ) . 323.Pp 324If the 325.Ql d 326(direct) flag is specified, 327.Fn fetchXGetFTP , 328.Fn fetchGetFTP 329and 330.Fn fetchPutFTP 331will use a direct connection even if a proxy server is defined. 332.Pp 333If no user name or password is given, the 334.Nm fetch 335library will attempt an anonymous login, with user name "anonymous" 336and password "anonymous@<hostname>". 337.Sh HTTP SCHEME 338The 339.Fn fetchXGetHTTP , 340.Fn fetchGetHTTP 341and 342.Fn fetchPutHTTP 343functions implement the HTTP/1.1 protocol. 344With a little luck, there is 345even a chance that they comply with RFC2616 and RFC2617. 346.Pp 347If the 348.Ql d 349(direct) flag is specified, 350.Fn fetchXGetHTTP , 351.Fn fetchGetHTTP 352and 353.Fn fetchPutHTTP 354will use a direct connection even if a proxy server is defined. 355.Pp 356Since there seems to be no good way of implementing the HTTP PUT 357method in a manner consistent with the rest of the 358.Nm fetch 359library, 360.Fn fetchPutHTTP 361is currently unimplemented. 362.Sh AUTHENTICATION 363Apart from setting the appropriate environment variables and 364specifying the user name and password in the URL or the 365.Vt struct url , 366the calling program has the option of defining an authentication 367function with the following prototype: 368.Pp 369.Ft int 370.Fn myAuthMethod "struct url *u" 371.Pp 372The callback function should fill in the 373.Fa user 374and 375.Fa pwd 376fields in the provided 377.Vt struct url 378and return 0 on success, or any other value to indicate failure. 379.Pp 380To register the authentication callback, simply set 381.Va fetchAuthMethod 382to point at it. 383The callback will be used whenever a site requires authentication and 384the appropriate environment variables are not set. 385.Pp 386This interface is experimental and may be subject to change. 387.Sh RETURN VALUES 388.Fn fetchParseURL 389returns a pointer to a 390.Vt struct url 391containing the individual components of the URL. 392If it is 393unable to allocate memory, or the URL is syntactically incorrect, 394.Fn fetchParseURL 395returns a NULL pointer. 396.Pp 397The 398.Fn fetchStat 399functions return 0 on success and -1 on failure. 400.Pp 401All other functions return a stream pointer which may be used to 402access the requested document, or NULL if an error occurred. 403.Pp 404The following error codes are defined in 405.In fetch.h : 406.Bl -tag -width 18n 407.It Bq Er FETCH_ABORT 408Operation aborted 409.It Bq Er FETCH_AUTH 410Authentication failed 411.It Bq Er FETCH_DOWN 412Service unavailable 413.It Bq Er FETCH_EXISTS 414File exists 415.It Bq Er FETCH_FULL 416File system full 417.It Bq Er FETCH_INFO 418Informational response 419.It Bq Er FETCH_MEMORY 420Insufficient memory 421.It Bq Er FETCH_MOVED 422File has moved 423.It Bq Er FETCH_NETWORK 424Network error 425.It Bq Er FETCH_OK 426No error 427.It Bq Er FETCH_PROTO 428Protocol error 429.It Bq Er FETCH_RESOLV 430Resolver error 431.It Bq Er FETCH_SERVER 432Server error 433.It Bq Er FETCH_TEMP 434Temporary error 435.It Bq Er FETCH_TIMEOUT 436Operation timed out 437.It Bq Er FETCH_UNAVAIL 438File is not available 439.It Bq Er FETCH_UNKNOWN 440Unknown error 441.It Bq Er FETCH_URL 442Invalid URL 443.El 444.Pp 445The accompanying error message includes a protocol-specific error code 446and message, e.g.\& "File is not available (404 Not Found)" 447.Sh ENVIRONMENT 448.Bl -tag -width ".Ev FETCH_BIND_ADDRESS" 449.It Ev FETCH_BIND_ADDRESS 450Specifies a hostname or IP address to which sockets used for outgoing 451connections will be bound. 452.It Ev FTP_LOGIN 453Default FTP login if none was provided in the URL. 454.It Ev FTP_PASSIVE_MODE 455If set to anything but 456.Ql no , 457forces the FTP code to use passive mode. 458.It Ev FTP_PASSWORD 459Default FTP password if the remote server requests one and none was 460provided in the URL. 461.It Ev FTP_PROXY 462URL of the proxy to use for FTP requests. 463The document part is ignored. 464FTP and HTTP proxies are supported; if no scheme is specified, FTP is 465assumed. 466If the proxy is an FTP proxy, 467.Nm libfetch 468will send 469.Ql user@host 470as user name to the proxy, where 471.Ql user 472is the real user name, and 473.Ql host 474is the name of the FTP server. 475.Pp 476If this variable is set to an empty string, no proxy will be used for 477FTP requests, even if the 478.Ev HTTP_PROXY 479variable is set. 480.It Ev ftp_proxy 481Same as 482.Ev FTP_PROXY , 483for compatibility. 484.It Ev HTTP_AUTH 485Specifies HTTP authorization parameters as a colon-separated list of 486items. 487The first and second item are the authorization scheme and realm 488respectively; further items are scheme-dependent. 489Currently, only basic authorization is supported. 490.Pp 491Basic authorization requires two parameters: the user name and 492password, in that order. 493.Pp 494This variable is only used if the server requires authorization and 495no user name or password was specified in the URL. 496.It Ev HTTP_PROXY 497URL of the proxy to use for HTTP requests. 498The document part is ignored. 499Only HTTP proxies are supported for HTTP requests. 500If no port number is specified, the default is 3128. 501.Pp 502Note that this proxy will also be used for FTP documents, unless the 503.Ev FTP_PROXY 504variable is set. 505.It Ev http_proxy 506Same as 507.Ev HTTP_PROXY , 508for compatibility. 509.It Ev HTTP_PROXY_AUTH 510Specifies authorization parameters for the HTTP proxy in the same 511format as the 512.Ev HTTP_AUTH 513variable. 514.Pp 515This variable is used if and only if connected to an HTTP proxy, and 516is ignored if a user and/or a password were specified in the proxy 517URL. 518.It Ev HTTP_REFERER 519Specifies the referrer URL to use for HTTP requests. 520If set to 521.Dq auto , 522the document URL will be used as referrer URL. 523.It Ev HTTP_USER_AGENT 524Specifies the User-Agent string to use for HTTP requests. 525This can be useful when working with HTTP origin or proxy servers that 526differentiate between user agents. 527.It Ev NETRC 528Specifies a file to use instead of 529.Pa ~/.netrc 530to look up login names and passwords for FTP sites. 531See 532.Xr ftp 1 533for a description of the file format. 534This feature is experimental. 535.It Ev NO_PROXY 536Either a single asterisk, which disables the use of proxies 537altogether, or a comma- or whitespace-separated list of hosts for 538which proxies should not be used. 539.It Ev no_proxy 540Same as 541.Ev NO_PROXY , 542for compatibility. 543.El 544.Sh EXAMPLES 545To access a proxy server on 546.Pa proxy.example.com 547port 8080, set the 548.Ev HTTP_PROXY 549environment variable in a manner similar to this: 550.Pp 551.Dl HTTP_PROXY=http://proxy.example.com:8080 552.Pp 553If the proxy server requires authentication, there are 554two options available for passing the authentication data. 555The first method is by using the proxy URL: 556.Pp 557.Dl HTTP_PROXY=http://<user>:<pwd>@proxy.example.com:8080 558.Pp 559The second method is by using the 560.Ev HTTP_PROXY_AUTH 561environment variable: 562.Bd -literal -offset indent 563HTTP_PROXY=http://proxy.example.com:8080 564HTTP_PROXY_AUTH=basic:*:<user>:<pwd> 565.Ed 566.Pp 567To disable the use of a proxy for an HTTP server running on the local 568host, define 569.Ev NO_PROXY 570as follows: 571.Bd -literal -offset indent 572NO_PROXY=localhost,127.0.0.1 573.Ed 574.Sh SEE ALSO 575.Xr fetch 1 , 576.Xr ftpio 3 , 577.Xr ip 4 578.Rs 579.%A J. Postel 580.%A J. K. Reynolds 581.%D October 1985 582.%B File Transfer Protocol 583.%O RFC959 584.Re 585.Rs 586.%A P. Deutsch 587.%A A. Emtage 588.%A A. Marine. 589.%D May 1994 590.%T How to Use Anonymous FTP 591.%O RFC1635 592.Re 593.Rs 594.%A T. Berners-Lee 595.%A L. Masinter 596.%A M. McCahill 597.%D December 1994 598.%T Uniform Resource Locators (URL) 599.%O RFC1738 600.Re 601.Rs 602.%A R. Fielding 603.%A J. Gettys 604.%A J. Mogul 605.%A H. Frystyk 606.%A L. Masinter 607.%A P. Leach 608.%A T. Berners-Lee 609.%D January 1999 610.%B Hypertext Transfer Protocol -- HTTP/1.1 611.%O RFC2616 612.Re 613.Rs 614.%A J. Franks 615.%A P. Hallam-Baker 616.%A J. Hostetler 617.%A S. Lawrence 618.%A P. Leach 619.%A A. Luotonen 620.%A L. Stewart 621.%D June 1999 622.%B HTTP Authentication: Basic and Digest Access Authentication 623.%O RFC2617 624.Re 625.Sh HISTORY 626The 627.Nm fetch 628library first appeared in 629.Fx 3.0 . 630.Sh AUTHORS 631.An -nosplit 632The 633.Nm fetch 634library was mostly written by 635.An Dag-Erling Sm\(/orgrav Aq des@FreeBSD.org 636with numerous suggestions from 637.An Jordan K. Hubbard Aq jkh@FreeBSD.org , 638.An Eugene Skepner Aq eu@qub.com 639and other 640.Fx 641developers. 642It replaces the older 643.Nm ftpio 644library written by 645.An Poul-Henning Kamp Aq phk@FreeBSD.org 646and 647.An Jordan K. Hubbard Aq jkh@FreeBSD.org . 648.Pp 649This manual page was written by 650.An Dag-Erling Sm\(/orgrav Aq des@FreeBSD.org . 651.Sh BUGS 652Some parts of the library are not yet implemented. 653The most notable 654examples of this are 655.Fn fetchPutHTTP , 656.Fn fetchListHTTP , 657.Fn fetchListFTP 658and FTP proxy support. 659.Pp 660There is no way to select a proxy at run-time other than setting the 661.Ev HTTP_PROXY 662or 663.Ev FTP_PROXY 664environment variables as appropriate. 665.Pp 666.Nm libfetch 667does not understand or obey 305 (Use Proxy) replies. 668.Pp 669Error numbers are unique only within a certain context; the error 670codes used for FTP and HTTP overlap, as do those used for resolver and 671system errors. 672For instance, error code 202 means "Command not 673implemented, superfluous at this site" in an FTP context and 674"Accepted" in an HTTP context. 675.Pp 676.Fn fetchStatFTP 677does not check that the result of an MDTM command is a valid date. 678.Pp 679The man page is incomplete, poorly written and produces badly 680formatted text. 681.Pp 682The error reporting mechanism is unsatisfactory. 683.Pp 684Some parts of the code are not fully reentrant. 685