1: _ _ ____ _
2: ___| | | | _ \| |
3: / __| | | | |_) | |
4: | (__| |_| | _ <| |___
5: \___|\___/|_| \_\_____|
6:
7: FAQ
8:
9: 1. Philosophy
10: 1.1 What is cURL?
11: 1.2 What is libcurl?
12: 1.3 What is curl not?
13: 1.4 When will you make curl do XXXX ?
14: 1.5 Who makes curl?
15: 1.6 What do you get for making curl?
16: 1.7 What about CURL from curl.com?
17: 1.8 I have a problem who do I mail?
18: 1.9 Where do I buy commercial support for curl?
19: 1.10 How many are using curl?
20: 1.11 Why don't you update ca-bundle.crt
21: 1.12 I have a problem who can I chat with?
22: 1.13 curl's ECCN number?
23: 1.14 How do I submit my patch?
24: 1.15 How do I port libcurl to my OS?
25:
26: 2. Install Related Problems
27: 2.1 configure doesn't find OpenSSL even when it is installed
28: 2.1.1 native linker doesn't find OpenSSL
29: 2.1.2 only the libssl lib is missing
30: 2.2 Does curl work/build with other SSL libraries?
31: 2.3 Where can I find a copy of LIBEAY32.DLL?
32: 2.4 Does curl support SOCKS (RFC 1928) ?
33:
34: 3. Usage Problems
35: 3.1 curl: (1) SSL is disabled, https: not supported
36: 3.2 How do I tell curl to resume a transfer?
37: 3.3 Why doesn't my posting using -F work?
38: 3.4 How do I tell curl to run custom FTP commands?
39: 3.5 How can I disable the Accept: */* header?
40: 3.6 Does curl support ASP, XML, XHTML or HTML version Y?
41: 3.7 Can I use curl to delete/rename a file through FTP?
42: 3.8 How do I tell curl to follow HTTP redirects?
43: 3.9 How do I use curl in my favorite programming language?
44: 3.10 What about SOAP, WebDAV, XML-RPC or similar protocols over HTTP?
45: 3.11 How do I POST with a different Content-Type?
46: 3.12 Why do FTP-specific features over HTTP proxy fail?
47: 3.13 Why do my single/double quotes fail?
48: 3.14 Does curl support Javascript or PAC (automated proxy config)?
49: 3.15 Can I do recursive fetches with curl?
50: 3.16 What certificates do I need when I use SSL?
51: 3.17 How do I list the root dir of an FTP server?
52: 3.18 Can I use curl to send a POST/PUT and not wait for a response?
53: 3.19 How do I get HTTP from a host using a specific IP address?
54: 3.20 How to SFTP from my user's home directory?
55: 3.21 Protocol xxx not supported or disabled in libcurl
56: 3.22 curl -X gives me HTTP problems
57:
58: 4. Running Problems
59: 4.1 Problems connecting to SSL servers.
60: 4.2 Why do I get problems when I use & or % in the URL?
61: 4.3 How can I use {, }, [ or ] to specify multiple URLs?
62: 4.4 Why do I get downloaded data even though the web page doesn't exist?
63: 4.5 Why do I get return code XXX from a HTTP server?
64: 4.5.1 "400 Bad Request"
65: 4.5.2 "401 Unauthorized"
66: 4.5.3 "403 Forbidden"
67: 4.5.4 "404 Not Found"
68: 4.5.5 "405 Method Not Allowed"
69: 4.5.6 "301 Moved Permanently"
70: 4.6 Can you tell me what error code 142 means?
71: 4.7 How do I keep user names and passwords secret in Curl command lines?
72: 4.8 I found a bug!
73: 4.9 Curl can't authenticate to the server that requires NTLM?
74: 4.10 My HTTP request using HEAD, PUT or DELETE doesn't work!
75: 4.11 Why do my HTTP range requests return the full document?
76: 4.12 Why do I get "certificate verify failed" ?
77: 4.13 Why is curl -R on Windows one hour off?
78: 4.14 Redirects work in browser but not with curl!
79: 4.15 FTPS doesn't work
80: 4.16 My HTTP POST or PUT requests are slow!
81: 4.17 Non-functional connect timeouts on Windows
82: 4.18 file:// URLs containing drive letters (Windows, NetWare)
83: 4.19 Why doesn't curl return an error when the network cable is unplugged?
84: 4.20 curl doesn't return error for HTTP non-200 responses!
85: 4.21 Why is there a HTTP/1.1 in my HTTP/2 request?
86:
87: 5. libcurl Issues
88: 5.1 Is libcurl thread-safe?
89: 5.2 How can I receive all data into a large memory chunk?
90: 5.3 How do I fetch multiple files with libcurl?
91: 5.4 Does libcurl do Winsock initing on win32 systems?
92: 5.5 Does CURLOPT_WRITEDATA and CURLOPT_READDATA work on win32 ?
93: 5.6 What about Keep-Alive or persistent connections?
94: 5.7 Link errors when building libcurl on Windows!
95: 5.8 libcurl.so.X: open failed: No such file or directory
96: 5.9 How does libcurl resolve host names?
97: 5.10 How do I prevent libcurl from writing the response to stdout?
98: 5.11 How do I make libcurl not receive the whole HTTP response?
99: 5.12 Can I make libcurl fake or hide my real IP address?
100: 5.13 How do I stop an ongoing transfer?
101: 5.14 Using C++ non-static functions for callbacks?
102: 5.15 How do I get an FTP directory listing?
103: 5.16 I want a different time-out!
104: 5.17 Can I write a server with libcurl?
105: 5.18 Does libcurl use threads?
106:
107: 6. License Issues
108: 6.1 I have a GPL program, can I use the libcurl library?
109: 6.2 I have a closed-source program, can I use the libcurl library?
110: 6.3 I have a BSD licensed program, can I use the libcurl library?
111: 6.4 I have a program that uses LGPL libraries, can I use libcurl?
112: 6.5 Can I modify curl/libcurl for my program and keep the changes secret?
113: 6.6 Can you please change the curl/libcurl license to XXXX?
114: 6.7 What are my obligations when using libcurl in my commercial apps?
115:
116: 7. PHP/CURL Issues
117: 7.1 What is PHP/CURL?
118: 7.2 Who wrote PHP/CURL?
119: 7.3 Can I perform multiple requests using the same handle?
120: 7.4 Does PHP/CURL have dependencies?
121:
122: ==============================================================================
123:
124: 1. Philosophy
125:
126: 1.1 What is cURL?
127:
128: cURL is the name of the project. The name is a play on 'Client for URLs',
129: originally with URL spelled in uppercase to make it obvious it deals with
130: URLs. The fact it can also be pronounced 'see URL' also helped, it works as
131: an abbreviation for "Client URL Request Library" or why not the recursive
132: version: "Curl URL Request Library".
133:
134: The cURL project produces two products:
135:
136: libcurl
137:
138: A free and easy-to-use client-side URL transfer library, supporting DICT,
139: FILE, FTP, FTPS, GOPHER, HTTP, HTTPS, IMAP, IMAPS, LDAP, LDAPS, POP3,
140: POP3S, RTMP, RTSP, SCP, SFTP, SMB, SMBS, SMTP, SMTPS, TELNET and TFTP.
141:
142: libcurl supports HTTPS certificates, HTTP POST, HTTP PUT, FTP uploading,
143: Kerberos, SPNEGO, HTTP form based upload, proxies, cookies, user+password
144: authentication, file transfer resume, http proxy tunneling and more!
145:
146: libcurl is highly portable, it builds and works identically on numerous
147: platforms, including Solaris, NetBSD, FreeBSD, OpenBSD, Darwin, HP-UX,
148: IRIX, AIX, Tru64, Linux, UnixWare, HURD, Windows, Amiga, OS/2, BeOS, Mac
149: OS X, Ultrix, QNX, OpenVMS, RISC OS, Novell NetWare, DOS, Symbian, OSF,
150: Android, Minix, IBM TPF and more...
151:
152: libcurl is free, thread-safe, IPv6 compatible, feature rich, well
153: supported and fast.
154:
155: curl
156:
157: A command line tool for getting or sending files using URL syntax.
158:
159: Since curl uses libcurl, curl supports the same wide range of common
160: Internet protocols that libcurl does.
161:
162: We pronounce curl with an initial k sound. It rhymes with words like girl
163: and earl. This is a short WAV file to help you:
164:
165: https://media.merriam-webster.com/soundc11/c/curl0001.wav
166:
167: There are numerous sub-projects and related projects that also use the word
168: curl in the project names in various combinations, but you should take
169: notice that this FAQ is directed at the command-line tool named curl (and
170: libcurl the library), and may therefore not be valid for other curl-related
171: projects. (There is however a small section for the PHP/CURL in this FAQ.)
172:
173: 1.2 What is libcurl?
174:
175: libcurl is a reliable and portable library which provides you with an easy
176: interface to a range of common Internet protocols.
177:
178: You can use libcurl for free in your application, be it open source,
179: commercial or closed-source.
180:
181: libcurl is most probably the most portable, most powerful and most often
182: used C-based multi-platform file transfer library on this planet - be it
183: open source or commercial.
184:
185: 1.3 What is curl not?
186:
187: Curl is not a wget clone. That is a common misconception. Never, during
188: curl's development, have we intended curl to replace wget or compete on its
189: market. Curl is targeted at single-shot file transfers.
190:
191: Curl is not a web site mirroring program. If you want to use curl to mirror
192: something: fine, go ahead and write a script that wraps around curl to make
193: it reality (like curlmirror.pl does).
194:
195: Curl is not an FTP site mirroring program. Sure, get and send FTP with curl
196: but if you want systematic and sequential behavior you should write a
197: script (or write a new program that interfaces libcurl) and do it.
198:
199: Curl is not a PHP tool, even though it works perfectly well when used from
200: or with PHP (when using the PHP/CURL module).
201:
202: Curl is not a program for a single operating system. Curl exists, compiles,
203: builds and runs under a wide range of operating systems, including all
204: modern Unixes (and a bunch of older ones too), Windows, Amiga, BeOS, OS/2,
205: OS X, QNX etc.
206:
207: 1.4 When will you make curl do XXXX ?
208:
209: We love suggestions of what to change in order to make curl and libcurl
210: better. We do however believe in a few rules when it comes to the future of
211: curl:
212:
213: Curl -- the command line tool -- is to remain a non-graphical command line
214: tool. If you want GUIs or fancy scripting capabilities, you should look for
215: another tool that uses libcurl.
216:
217: We do not add things to curl that other small and available tools already do
218: very well at the side. Curl's output can be piped into another program or
219: redirected to another file for the next program to interpret.
220:
221: We focus on protocol related issues and improvements. If you want to do more
222: magic with the supported protocols than curl currently does, chances are good
223: we will agree. If you want to add more protocols, we may very well agree.
224:
225: If you want someone else to do all the work while you wait for us to
226: implement it for you, that is not a very friendly attitude. We spend a
227: considerable time already on maintaining and developing curl. In order to
228: get more out of us, you should consider trading in some of your time and
229: effort in return. Simply go to the GitHub repo which resides at
230: https://github.com/curl/curl, fork the project, and create pull requests
231: with your proposed changes.
232:
233: If you write the code, chances are better that it will get into curl faster.
234:
235: 1.5 Who makes curl?
236:
237: curl and libcurl are not made by any single individual. Daniel Stenberg is
238: project leader and main developer, but other persons' submissions are
239: important and crucial. Anyone can contribute and post their changes and
240: improvements and have them inserted in the main sources (of course on the
241: condition that developers agree that the fixes are good).
242:
243: The full list of all contributors is found in the docs/THANKS file.
244:
245: curl is developed by a community, with Daniel at the wheel.
246:
247: 1.6 What do you get for making curl?
248:
249: Project cURL is entirely free and open. No person gets paid for developing
250: curl full time. We do this voluntarily, mostly in our spare time.
251: Occasionally companies pay individual developers to work on curl, but that's
252: up to each company and developer. This is not controlled by nor supervised in
253: any way by the project.
254:
255: We still get help from companies. Haxx provides web site, bandwidth, mailing
256: lists etc, GitHub hosts the primary git repository and other services like
257: the bug tracker at https://github.com/curl/curl. Also again, some companies
258: have sponsored certain parts of the development in the past and I hope some
259: will continue to do so in the future.
260:
261: If you want to support our project, consider a donation or a banner-program
262: or even better: by helping us with coding, documenting or testing etc.
263:
264: 1.7 What about CURL from curl.com?
265:
266: During the summer of 2001, curl.com was busy advertising their client-side
267: programming language for the web, named CURL.
268:
269: We are in no way associated with curl.com or their CURL programming
270: language.
271:
272: Our project name curl has been in effective use since 1998. We were not the
273: first computer related project to use the name "curl" and do not claim any
274: rights to the name.
275:
276: We recognize that we will be living in parallel with curl.com and wish them
277: every success.
278:
279: 1.8 I have a problem whom do I mail?
280:
281: Please do not mail any single individual unless you really need to. Keep
282: curl-related questions on a suitable mailing list. All available mailing
283: lists are listed in the MANUAL document and online at
284: https://curl.haxx.se/mail/
285:
286: Keeping curl-related questions and discussions on mailing lists allows
287: others to join in and help, to share their ideas, to contribute their
288: suggestions and to spread their wisdom. Keeping discussions on public mailing
289: lists also allows for others to learn from this (both current and future
290: users thanks to the web based archives of the mailing lists), thus saving us
291: from having to repeat ourselves even more. Thanks for respecting this.
292:
293: If you have found or simply suspect a security problem in curl or libcurl,
294: mail curl-security at haxx.se (closed list of receivers, mails are not
295: disclosed) and tell. Then we can produce a fix in a timely manner before the
296: flaw is announced to the world, thus lessen the impact the problem will have
297: on existing users.
298:
299: 1.9 Where do I buy commercial support for curl?
300:
301: curl is fully open source. It means you can hire any skilled engineer to fix
302: your curl-related problems.
303:
304: We list available alternatives on the curl web site:
305: https://curl.haxx.se/support.html
306:
307: 1.10 How many are using curl?
308:
309: It is impossible to tell.
310:
311: We don't know how many users that knowingly have installed and use curl.
312:
313: We don't know how many users that use curl without knowing that they are in
314: fact using it.
315:
316: We don't know how many users that downloaded or installed curl and then
317: never use it.
318:
319: In May 2012 Daniel did a counting game and came up with a number that may
320: be completely wrong or somewhat accurate. Over 500 million!
321:
322: See https://daniel.haxx.se/blog/2012/05/16/300m-users/
323:
324: 1.11 Why don't you update ca-bundle.crt
325:
326: The ca cert bundle that used to be shipped with curl was very outdated and
327: must be replaced with an up-to-date version by anyone who wants to verify
328: peers. It is no longer provided by curl. The last curl release that ever
329: shipped a ca cert bundle was curl 7.18.0.
330:
331: In the cURL project we've decided not to attempt to keep this file updated
332: (or even present anymore) since deciding what to add to a ca cert bundle is
333: an undertaking we've not been ready to accept, and the one we can get from
334: Mozilla is perfectly fine so there's no need to duplicate that work.
335:
336: Today, with many services performed over HTTPS, every operating system
337: should come with a default ca cert bundle that can be deemed somewhat
338: trustworthy and that collection (if reasonably updated) should be deemed to
339: be a lot better than a private curl version.
340:
341: If you want the most recent collection of ca certs that Mozilla Firefox
342: uses, we recommend that you extract the collection yourself from Mozilla
343: Firefox (by running 'make ca-bundle), or by using our online service setup
344: for this purpose: https://curl.haxx.se/docs/caextract.html
345:
346: 1.12 I have a problem who can I chat with?
347:
348: There's a bunch of friendly people hanging out in the #curl channel on the
349: IRC network irc.freenode.net. If you're polite and nice, chances are good
350: that you can get -- or provide -- help instantly.
351:
352: 1.13 curl's ECCN number?
353:
354: The US government restricts exports of software that contains or uses
355: cryptography. When doing so, the Export Control Classification Number (ECCN)
356: is used to identify the level of export control etc.
357:
358: Apache Software Foundation gives a good explanation of ECCNs at
359: https://www.apache.org/dev/crypto.html
360:
361: We believe curl's number might be ECCN 5D002, another possibility is
362: 5D992. It seems necessary to write them (the authority that administers ECCN
363: numbers), asking to confirm.
364:
365: Comprehensible explanations of the meaning of such numbers and how to obtain
366: them (resp.) are here
367:
368: https://www.bis.doc.gov/licensing/exportingbasics.htm
369: https://www.bis.doc.gov/licensing/do_i_needaneccn.html
370:
371: An incomprehensible description of the two numbers above is here
372: https://www.bis.doc.gov/index.php/documents/new-encryption/1653-ccl5-pt2-3
373:
374: 1.14 How do I submit my patch?
375:
376: When you have made a patch or a change of whatever sort, and want to submit
377: that to the project, there are a few different ways we prefer:
378:
379: o send a patch to the curl-library mailing list. We're many subscribers
380: there and there are lots of people who can review patches, comment on them
381: and "receive" them properly.
382:
383: o if your patch changes or fixes a bug, you can also opt to submit a bug
384: report in the bug tracker and attach your patch there. There are less
385: people involved there.
386:
387: Lots of more details are found in the CONTRIBUTE and INTERNALS docs.
388:
389: 1.15 How do I port libcurl to my OS?
390:
391: Here's a rough step-by-step:
392:
393: 1. copy a suitable lib/config-*.h file as a start to lib/config-[youros].h
394:
395: 2. edit lib/config-[youros].h to match your OS and setup
396:
397: 3. edit lib/curl_setup.h to include config-[youros].h when your OS is
398: detected by the preprocessor, in the style others already exist
399:
400: 4. compile lib/*.c and make them into a library
401:
402:
403: 2. Install Related Problems
404:
405: 2.1 configure doesn't find OpenSSL even when it is installed
406:
407: This may be because of several reasons.
408:
409: 2.1.1 native linker doesn't find openssl
410:
411: Affected platforms:
412: Solaris (native cc compiler)
413: HPUX (native cc compiler)
414: SGI IRIX (native cc compiler)
415: SCO UNIX (native cc compiler)
416:
417: When configuring curl, I specify --with-ssl. OpenSSL is installed in
418: /usr/local/ssl Configure reports SSL in /usr/local/ssl, but fails to find
419: CRYPTO_lock in -lcrypto
420:
421: Cause: The cc for this test places the -L/usr/local/ssl/lib AFTER
422: -lcrypto, so ld can't find the library. This is due to a bug in the GNU
423: autoconf tool.
424:
425: Workaround: Specifying "LDFLAGS=-L/usr/local/ssl/lib" in front of
426: ./configure places the -L/usr/local/ssl/lib early enough in the command
427: line to make things work
428:
429: 2.1.2 only the libssl lib is missing
430:
431: If all include files and the libcrypto lib is present, with only the
432: libssl being missing according to configure, this is most likely because
433: a few functions are left out from the libssl.
434:
435: If the function names missing include RSA or RSAREF you can be certain
436: that this is because libssl requires the RSA and RSAREF libs to build.
437:
438: See the INSTALL file section that explains how to add those libs to
439: configure. Make sure that you remove the config.cache file before you
440: rerun configure with the new flags.
441:
442: 2.2 Does curl work/build with other SSL libraries?
443:
444: Curl has been written to use a generic SSL function layer internally, and
445: that SSL functionality can then be provided by one out of many different SSL
446: backends.
447:
448: curl can be built to use one of the following SSL alternatives: OpenSSL,
449: libressl, BoringSSL, GnuTLS, wolfSSL, NSS, mbedTLS, MesaLink, Secure
450: Transport (native iOS/OS X), Schannel (native Windows), GSKit (native IBM
451: i), or BearSSL. They all have their pros and cons, and we try to maintain a
452: comparison of them here: https://curl.haxx.se/docs/ssl-compared.html
453:
454: 2.3 Where can I find a copy of LIBEAY32.DLL?
455:
456: That is an OpenSSL binary built for Windows.
457:
458: Curl can be built with OpenSSL to do the SSL stuff. The LIBEAY32.DLL is then
459: what curl needs on a windows machine to do https:// etc. Check out the curl
460: web site to find accurate and up-to-date pointers to recent OpenSSL DLLs and
461: other binary packages.
462:
463: 2.4 Does curl support SOCKS (RFC 1928) ?
464:
465: Yes, SOCKS 4 and 5 are supported.
466:
467:
468: 3. Usage problems
469:
470: 3.1 curl: (1) SSL is disabled, https: not supported
471:
472: If you get this output when trying to get anything from a https:// server,
473: it means that the instance of curl/libcurl that you're using was built
474: without support for this protocol.
475:
476: This could've happened if the configure script that was run at build time
477: couldn't find all libs and include files curl requires for SSL to work. If
478: the configure script fails to find them, curl is simply built without SSL
479: support.
480:
481: To get the https:// support into a curl that was previously built but that
482: reports that https:// is not supported, you should dig through the document
483: and logs and check out why the configure script doesn't find the SSL libs
484: and/or include files.
485:
486: Also, check out the other paragraph in this FAQ labeled "configure doesn't
487: find OpenSSL even when it is installed".
488:
489: 3.2 How do I tell curl to resume a transfer?
490:
491: Curl supports resumed transfers both ways on both FTP and HTTP.
492: Try the -C option.
493:
494: 3.3 Why doesn't my posting using -F work?
495:
496: You can't arbitrarily use -F or -d, the choice between -F or -d depends on the
497: HTTP operation you need curl to do and what the web server that will receive
498: your post expects.
499:
500: If the form you're trying to submit uses the type 'multipart/form-data', then
501: and only then you must use the -F type. In all the most common cases, you
502: should use -d which then causes a posting with the type
503: 'application/x-www-form-urlencoded'.
504:
505: This is described in some detail in the MANUAL and TheArtOfHttpScripting
506: documents, and if you don't understand it the first time, read it again
507: before you post questions about this to the mailing list. Also, try reading
508: through the mailing list archives for old postings and questions regarding
509: this.
510:
511: 3.4 How do I tell curl to run custom FTP commands?
512:
513: You can tell curl to perform optional commands both before and/or after a
514: file transfer. Study the -Q/--quote option.
515:
516: Since curl is used for file transfers, you don't normally use curl to
517: perform FTP commands without transferring anything. Therefore you must
518: always specify a URL to transfer to/from even when doing custom FTP
519: commands, or use -I which implies the "no body" option sent to libcurl.
520:
521: 3.5 How can I disable the Accept: */* header?
522:
523: You can change all internally generated headers by adding a replacement with
524: the -H/--header option. By adding a header with empty contents you safely
525: disable that one. Use -H "Accept:" to disable that specific header.
526:
527: 3.6 Does curl support ASP, XML, XHTML or HTML version Y?
528:
529: To curl, all contents are alike. It doesn't matter how the page was
530: generated. It may be ASP, PHP, Perl, shell-script, SSI or plain HTML
531: files. There's no difference to curl and it doesn't even know what kind of
532: language that generated the page.
533:
534: See also item 3.14 regarding javascript.
535:
536: 3.7 Can I use curl to delete/rename a file through FTP?
537:
538: Yes. You specify custom FTP commands with -Q/--quote.
539:
540: One example would be to delete a file after you have downloaded it:
541:
542: curl -O ftp://download.com/coolfile -Q '-DELE coolfile'
543:
544: or rename a file after upload:
545:
546: curl -T infile ftp://upload.com/dir/ -Q "-RNFR infile" -Q "-RNTO newname"
547:
548: 3.8 How do I tell curl to follow HTTP redirects?
549:
550: Curl does not follow so-called redirects by default. The Location: header
551: that informs the client about this is only interpreted if you're using the
552: -L/--location option. As in:
553:
554: curl -L http://redirector.com
555:
556: Not all redirects are HTTP ones, see 4.14
557:
558: 3.9 How do I use curl in my favorite programming language?
559:
560: Many programming languages have interfaces/bindings that allow you to use
561: curl without having to use the command line tool. If you are fluent in such
562: a language, you may prefer to use one of these interfaces instead.
563:
564: Find out more about which languages that support curl directly, and how to
565: install and use them, in the libcurl section of the curl web site:
566: https://curl.haxx.se/libcurl/
567:
568: All the various bindings to libcurl are made by other projects and people,
569: outside of the cURL project. The cURL project itself only produces libcurl
570: with its plain C API. If you don't find anywhere else to ask you can ask
571: about bindings on the curl-library list too, but be prepared that people on
572: that list may not know anything about bindings.
573:
574: In February 2019, there were interfaces available for the following
575: languages: Ada95, Basic, C, C++, Ch, Cocoa, D, Delphi, Dylan, Eiffel,
576: Euphoria, Falcon, Ferite, Gambas, glib/GTK+, Go, Guile, Harbour, Haskell,
577: Java, Julia, Lisp, Lua, Mono, .NET, node.js, Object-Pascal, OCaml, Pascal,
578: Perl, PHP, PostgreSQL, Python, R, Rexx, Ring, RPG, Ruby, Rust, Scheme,
579: Scilab, S-Lang, Smalltalk, SP-Forth, SPL, Tcl, Visual Basic, Visual FoxPro,
580: Q, wxwidgets, XBLite and Xoho. By the time you read this, additional ones
581: may have appeared!
582:
583: 3.10 What about SOAP, WebDAV, XML-RPC or similar protocols over HTTP?
584:
585: Curl adheres to the HTTP spec, which basically means you can play with *any*
586: protocol that is built on top of HTTP. Protocols such as SOAP, WEBDAV and
587: XML-RPC are all such ones. You can use -X to set custom requests and -H to
588: set custom headers (or replace internally generated ones).
589:
590: Using libcurl is of course just as good and you'd just use the proper
591: library options to do the same.
592:
593: 3.11 How do I POST with a different Content-Type?
594:
595: You can always replace the internally generated headers with -H/--header.
596: To make a simple HTTP POST with text/xml as content-type, do something like:
597:
598: curl -d "datatopost" -H "Content-Type: text/xml" [URL]
599:
600: 3.12 Why do FTP-specific features over HTTP proxy fail?
601:
602: Because when you use a HTTP proxy, the protocol spoken on the network will
603: be HTTP, even if you specify a FTP URL. This effectively means that you
604: normally can't use FTP-specific features such as FTP upload and FTP quote
605: etc.
606:
607: There is one exception to this rule, and that is if you can "tunnel through"
608: the given HTTP proxy. Proxy tunneling is enabled with a special option (-p)
609: and is generally not available as proxy admins usually disable tunneling to
610: ports other than 443 (which is used for HTTPS access through proxies).
611:
612: 3.13 Why do my single/double quotes fail?
613:
614: To specify a command line option that includes spaces, you might need to
615: put the entire option within quotes. Like in:
616:
617: curl -d " with spaces " url.com
618:
619: or perhaps
620:
621: curl -d ' with spaces ' url.com
622:
623: Exactly what kind of quotes and how to do this is entirely up to the shell
624: or command line interpreter that you are using. For most unix shells, you
625: can more or less pick either single (') or double (") quotes. For
626: Windows/DOS prompts I believe you're forced to use double (") quotes.
627:
628: Please study the documentation for your particular environment. Examples in
629: the curl docs will use a mix of both of these as shown above. You must
630: adjust them to work in your environment.
631:
632: Remember that curl works and runs on more operating systems than most single
633: individuals have ever tried.
634:
635: 3.14 Does curl support Javascript or PAC (automated proxy config)?
636:
637: Many web pages do magic stuff using embedded Javascript. Curl and libcurl
638: have no built-in support for that, so it will be treated just like any other
639: contents.
640:
641: .pac files are a netscape invention and are sometimes used by organizations
642: to allow them to differentiate which proxies to use. The .pac contents is
643: just a Javascript program that gets invoked by the browser and that returns
644: the name of the proxy to connect to. Since curl doesn't support Javascript,
645: it can't support .pac proxy configuration either.
646:
647: Some workarounds usually suggested to overcome this Javascript dependency:
648:
649: Depending on the Javascript complexity, write up a script that translates it
650: to another language and execute that.
651:
652: Read the Javascript code and rewrite the same logic in another language.
653:
654: Implement a Javascript interpreter, people have successfully used the
655: Mozilla Javascript engine in the past.
656:
657: Ask your admins to stop this, for a static proxy setup or similar.
658:
659: 3.15 Can I do recursive fetches with curl?
660:
661: No. curl itself has no code that performs recursive operations, such as
662: those performed by wget and similar tools.
663:
664: There exists wrapper scripts with that functionality (for example the
665: curlmirror perl script), and you can write programs based on libcurl to do
666: it, but the command line tool curl itself cannot.
667:
668: 3.16 What certificates do I need when I use SSL?
669:
670: There are three different kinds of "certificates" to keep track of when we
671: talk about using SSL-based protocols (HTTPS or FTPS) using curl or libcurl.
672:
673: CLIENT CERTIFICATE
674:
675: The server you communicate with may require that you can provide this in
676: order to prove that you actually are who you claim to be. If the server
677: doesn't require this, you don't need a client certificate.
678:
679: A client certificate is always used together with a private key, and the
680: private key has a pass phrase that protects it.
681:
682: SERVER CERTIFICATE
683:
684: The server you communicate with has a server certificate. You can and should
685: verify this certificate to make sure that you are truly talking to the real
686: server and not a server impersonating it.
687:
688: CERTIFICATE AUTHORITY CERTIFICATE ("CA cert")
689:
690: You often have several CA certs in a CA cert bundle that can be used to
691: verify a server certificate that was signed by one of the authorities in the
692: bundle. curl does not come with a CA cert bundle but most curl installs
693: provide one. You can also override the default.
694:
695: The server certificate verification process is made by using a Certificate
696: Authority certificate ("CA cert") that was used to sign the server
697: certificate. Server certificate verification is enabled by default in curl
698: and libcurl and is often the reason for problems as explained in FAQ entry
699: 4.12 and the SSLCERTS document
700: (https://curl.haxx.se/docs/sslcerts.html). Server certificates that are
701: "self-signed" or otherwise signed by a CA that you do not have a CA cert
702: for, cannot be verified. If the verification during a connect fails, you are
703: refused access. You then need to explicitly disable the verification to
704: connect to the server.
705:
706: 3.17 How do I list the root dir of an FTP server?
707:
708: There are two ways. The way defined in the RFC is to use an encoded slash
709: in the first path part. List the "/tmp" dir like this:
710:
711: curl ftp://ftp.sunet.se/%2ftmp/
712:
713: or the not-quite-kosher-but-more-readable way, by simply starting the path
714: section of the URL with a slash:
715:
716: curl ftp://ftp.sunet.se//tmp/
717:
718: 3.18 Can I use curl to send a POST/PUT and not wait for a response?
719:
720: No.
721:
722: But you could easily write your own program using libcurl to do such stunts.
723:
724: 3.19 How do I get HTTP from a host using a specific IP address?
725:
726: For example, you may be trying out a web site installation that isn't yet in
727: the DNS. Or you have a site using multiple IP addresses for a given host
728: name and you want to address a specific one out of the set.
729:
730: Set a custom Host: header that identifies the server name you want to reach
731: but use the target IP address in the URL:
732:
733: curl --header "Host: www.example.com" http://127.0.0.1/
734:
735: You can also opt to add faked host name entries to curl with the --resolve
736: option. That has the added benefit that things like redirects will also work
737: properly. The above operation would instead be done as:
738:
739: curl --resolve www.example.com:80:127.0.0.1 http://www.example.com/
740:
741: 3.20 How to SFTP from my user's home directory?
742:
743: Contrary to how FTP works, SFTP and SCP URLs specify the exact directory to
744: work with. It means that if you don't specify that you want the user's home
745: directory, you get the actual root directory.
746:
747: To specify a file in your user's home directory, you need to use the correct
748: URL syntax which for SFTP might look similar to:
749:
750: curl -O -u user:password sftp://example.com/~/file.txt
751:
752: and for SCP it is just a different protocol prefix:
753:
754: curl -O -u user:password scp://example.com/~/file.txt
755:
756: 3.21 Protocol xxx not supported or disabled in libcurl
757:
758: When passing on a URL to curl to use, it may respond that the particular
759: protocol is not supported or disabled. The particular way this error message
760: is phrased is because curl doesn't make a distinction internally of whether
761: a particular protocol is not supported (i.e. never got any code added that
762: knows how to speak that protocol) or if it was explicitly disabled. curl can
763: be built to only support a given set of protocols, and the rest would then
764: be disabled or not supported.
765:
766: Note that this error will also occur if you pass a wrongly spelled protocol
767: part as in "htpt://example.com" or as in the less evident case if you prefix
768: the protocol part with a space as in " http://example.com/".
769:
770: 3.22 curl -X gives me HTTP problems
771:
772: In normal circumstances, -X should hardly ever be used.
773:
774: By default you use curl without explicitly saying which request method to
775: use when the URL identifies a HTTP transfer. If you just pass in a URL like
776: "curl http://example.com" it will use GET. If you use -d or -F curl will use
777: POST, -I will cause a HEAD and -T will make it a PUT.
778:
779: If for whatever reason you're not happy with these default choices that curl
780: does for you, you can override those request methods by specifying -X
781: [WHATEVER]. This way you can for example send a DELETE by doing "curl -X
782: DELETE [URL]".
783:
784: It is thus pointless to do "curl -XGET [URL]" as GET would be used
785: anyway. In the same vein it is pointless to do "curl -X POST -d data
786: [URL]"... But you can make a fun and somewhat rare request that sends a
787: request-body in a GET request with something like "curl -X GET -d data
788: [URL]"
789:
790: Note that -X doesn't actually change curl's behavior as it only modifies the
791: actual string sent in the request, but that may of course trigger a
792: different set of events.
793:
794: Accordingly, by using -XPOST on a command line that for example would follow
795: a 303 redirect, you will effectively prevent curl from behaving
796: correctly. Be aware.
797:
798:
799: 4. Running Problems
800:
801: 4.1 Problems connecting to SSL servers.
802:
803: It took a very long time before we could sort out why curl had problems to
804: connect to certain SSL servers when using SSLeay or OpenSSL v0.9+. The
805: error sometimes showed up similar to:
806:
807: 16570:error:1407D071:SSL routines:SSL2_READ:bad mac decode:s2_pkt.c:233:
808:
809: It turned out to be because many older SSL servers don't deal with SSLv3
810: requests properly. To correct this problem, tell curl to select SSLv2 from
811: the command line (-2/--sslv2).
812:
813: There have also been examples where the remote server didn't like the SSLv2
814: request and instead you had to force curl to use SSLv3 with -3/--sslv3.
815:
816: 4.2 Why do I get problems when I use & or % in the URL?
817:
818: In general unix shells, the & symbol is treated specially and when used, it
819: runs the specified command in the background. To safely send the & as a part
820: of a URL, you should quote the entire URL by using single (') or double (")
821: quotes around it. Similar problems can also occur on some shells with other
822: characters, including ?*!$~(){}<>\|;`. When in doubt, quote the URL.
823:
824: An example that would invoke a remote CGI that uses &-symbols could be:
825:
826: curl 'http://www.altavista.com/cgi-bin/query?text=yes&q=curl'
827:
828: In Windows, the standard DOS shell treats the percent sign specially and you
829: need to use TWO percent signs for each single one you want to use in the
830: URL.
831:
832: If you want a literal percent sign to be part of the data you pass in a POST
833: using -d/--data you must encode it as '%25' (which then also needs the
834: percent sign doubled on Windows machines).
835:
836: 4.3 How can I use {, }, [ or ] to specify multiple URLs?
837:
838: Because those letters have a special meaning to the shell, to be used in
839: a URL specified to curl you must quote them.
840:
841: An example that downloads two URLs (sequentially) would be:
842:
843: curl '{curl,www}.haxx.se'
844:
845: To be able to use those characters as actual parts of the URL (without using
846: them for the curl URL "globbing" system), use the -g/--globoff option:
847:
848: curl -g 'www.site.com/weirdname[].html'
849:
850: 4.4 Why do I get downloaded data even though the web page doesn't exist?
851:
852: Curl asks remote servers for the page you specify. If the page doesn't exist
853: at the server, the HTTP protocol defines how the server should respond and
854: that means that headers and a "page" will be returned. That's simply how
855: HTTP works.
856:
857: By using the --fail option you can tell curl explicitly to not get any data
858: if the HTTP return code doesn't say success.
859:
860: 4.5 Why do I get return code XXX from a HTTP server?
861:
862: RFC2616 clearly explains the return codes. This is a short transcript. Go
863: read the RFC for exact details:
864:
865: 4.5.1 "400 Bad Request"
866:
867: The request could not be understood by the server due to malformed
868: syntax. The client SHOULD NOT repeat the request without modifications.
869:
870: 4.5.2 "401 Unauthorized"
871:
872: The request requires user authentication.
873:
874: 4.5.3 "403 Forbidden"
875:
876: The server understood the request, but is refusing to fulfill it.
877: Authorization will not help and the request SHOULD NOT be repeated.
878:
879: 4.5.4 "404 Not Found"
880:
881: The server has not found anything matching the Request-URI. No indication
882: is given of whether the condition is temporary or permanent.
883:
884: 4.5.5 "405 Method Not Allowed"
885:
886: The method specified in the Request-Line is not allowed for the resource
887: identified by the Request-URI. The response MUST include an Allow header
888: containing a list of valid methods for the requested resource.
889:
890: 4.5.6 "301 Moved Permanently"
891:
892: If you get this return code and an HTML output similar to this:
893:
894: <H1>Moved Permanently</H1> The document has moved <A
895: HREF="http://same_url_now_with_a_trailing_slash/">here</A>.
896:
897: it might be because you requested a directory URL but without the trailing
898: slash. Try the same operation again _with_ the trailing URL, or use the
899: -L/--location option to follow the redirection.
900:
901: 4.6 Can you tell me what error code 142 means?
902:
903: All curl error codes are described at the end of the man page, in the
904: section called "EXIT CODES".
905:
906: Error codes that are larger than the highest documented error code means
907: that curl has exited due to a crash. This is a serious error, and we
908: appreciate a detailed bug report from you that describes how we could go
909: ahead and repeat this!
910:
911: 4.7 How do I keep user names and passwords secret in Curl command lines?
912:
913: This problem has two sides:
914:
915: The first part is to avoid having clear-text passwords in the command line
916: so that they don't appear in 'ps' outputs and similar. That is easily
917: avoided by using the "-K" option to tell curl to read parameters from a file
918: or stdin to which you can pass the secret info. curl itself will also
919: attempt to "hide" the given password by blanking out the option - this
920: doesn't work on all platforms.
921:
922: To keep the passwords in your account secret from the rest of the world is
923: not a task that curl addresses. You could of course encrypt them somehow to
924: at least hide them from being read by human eyes, but that is not what
925: anyone would call security.
926:
927: Also note that regular HTTP (using Basic authentication) and FTP passwords
928: are sent as cleartext across the network. All it takes for anyone to fetch
929: them is to listen on the network. Eavesdropping is very easy. Use more secure
930: authentication methods (like Digest, Negotiate or even NTLM) or consider the
931: SSL-based alternatives HTTPS and FTPS.
932:
933: 4.8 I found a bug!
934:
935: It is not a bug if the behavior is documented. Read the docs first.
936: Especially check out the KNOWN_BUGS file, it may be a documented bug!
937:
938: If it is a problem with a binary you've downloaded or a package for your
939: particular platform, try contacting the person who built the package/archive
940: you have.
941:
942: If there is a bug, read the BUGS document first. Then report it as described
943: in there.
944:
945: 4.9 Curl can't authenticate to the server that requires NTLM?
946:
947: NTLM support requires OpenSSL, GnuTLS, mbedTLS, NSS, Secure Transport, or
948: Microsoft Windows libraries at build-time to provide this functionality.
949:
950: NTLM is a Microsoft proprietary protocol. Proprietary formats are evil. You
951: should not use such ones.
952:
953: 4.10 My HTTP request using HEAD, PUT or DELETE doesn't work!
954:
955: Many web servers allow or demand that the administrator configures the
956: server properly for these requests to work on the web server.
957:
958: Some servers seem to support HEAD only on certain kinds of URLs.
959:
960: To fully grasp this, try the documentation for the particular server
961: software you're trying to interact with. This is not anything curl can do
962: anything about.
963:
964: 4.11 Why do my HTTP range requests return the full document?
965:
966: Because the range may not be supported by the server, or the server may
967: choose to ignore it and return the full document anyway.
968:
969: 4.12 Why do I get "certificate verify failed" ?
970:
971: You invoke curl 7.10 or later to communicate on a https:// URL and get an
972: error back looking something similar to this:
973:
974: curl: (35) SSL: error:14090086:SSL routines:
975: SSL3_GET_SERVER_CERTIFICATE:certificate verify failed
976:
977: Then it means that curl couldn't verify that the server's certificate was
978: good. Curl verifies the certificate using the CA cert bundle that comes with
979: the curl installation.
980:
981: To disable the verification (which makes it act like curl did before 7.10),
982: use -k. This does however enable man-in-the-middle attacks.
983:
984: If you get this failure but are having a CA cert bundle installed and used,
985: the server's certificate is not signed by one of the CA's in the bundle. It
986: might for example be self-signed. You then correct this problem by obtaining
987: a valid CA cert for the server. Or again, decrease the security by disabling
988: this check.
989:
990: Details are also in the SSLCERTS file in the release archives, found online
991: here: https://curl.haxx.se/docs/sslcerts.html
992:
993: 4.13 Why is curl -R on Windows one hour off?
994:
995: Since curl 7.53.0 this issue should be fixed as long as curl was built with
996: any modern compiler that allows for a 64-bit curl_off_t type. For older
997: compilers or prior curl versions it may set a time that appears one hour off.
998: This happens due to a flaw in how Windows stores and uses file modification
999: times and it is not easily worked around. For more details read this:
1000: https://www.codeproject.com/Articles/1144/Beating-the-Daylight-Savings-Time-bug-and-getting
1001:
1002: 4.14 Redirects work in browser but not with curl!
1003:
1004: curl supports HTTP redirects well (see item 3.8). Browsers generally support
1005: at least two other ways to perform redirects that curl does not:
1006:
1007: Meta tags. You can write a HTML tag that will cause the browser to redirect
1008: to another given URL after a certain time.
1009:
1010: Javascript. You can write a Javascript program embedded in a HTML page that
1011: redirects the browser to another given URL.
1012:
1013: There is no way to make curl follow these redirects. You must either
1014: manually figure out what the page is set to do, or write a script that parses
1015: the results and fetches the new URL.
1016:
1017: 4.15 FTPS doesn't work
1018:
1019: curl supports FTPS (sometimes known as FTP-SSL) both implicit and explicit
1020: mode.
1021:
1022: When a URL is used that starts with FTPS://, curl assumes implicit SSL on
1023: the control connection and will therefore immediately connect and try to
1024: speak SSL. FTPS:// connections default to port 990.
1025:
1026: To use explicit FTPS, you use a FTP:// URL and the --ftp-ssl option (or one
1027: of its related flavors). This is the most common method, and the one
1028: mandated by RFC4217. This kind of connection will then of course use the
1029: standard FTP port 21 by default.
1030:
1031: 4.16 My HTTP POST or PUT requests are slow!
1032:
1033: libcurl makes all POST and PUT requests (except for POST requests with a
1034: very tiny request body) use the "Expect: 100-continue" header. This header
1035: allows the server to deny the operation early so that libcurl can bail out
1036: before having to send any data. This is useful in authentication
1037: cases and others.
1038:
1039: However, many servers don't implement the Expect: stuff properly and if the
1040: server doesn't respond (positively) within 1 second libcurl will continue
1041: and send off the data anyway.
1042:
1043: You can disable libcurl's use of the Expect: header the same way you disable
1044: any header, using -H / CURLOPT_HTTPHEADER, or by forcing it to use HTTP 1.0.
1045:
1046: 4.17 Non-functional connect timeouts
1047:
1048: In most Windows setups having a timeout longer than 21 seconds make no
1049: difference, as it will only send 3 TCP SYN packets and no more. The second
1050: packet sent three seconds after the first and the third six seconds after
1051: the second. No more than three packets are sent, no matter how long the
1052: timeout is set.
1053:
1054: See option TcpMaxConnectRetransmissions on this page:
1055: https://support.microsoft.com/en-us/kb/175523/en-us
1056:
1057: Also, even on non-Windows systems there may run a firewall or anti-virus
1058: software or similar that accepts the connection but does not actually do
1059: anything else. This will make (lib)curl to consider the connection connected
1060: and thus the connect timeout won't trigger.
1061:
1062: 4.18 file:// URLs containing drive letters (Windows, NetWare)
1063:
1064: When using curl to try to download a local file, one might use a URL
1065: in this format:
1066:
1067: file://D:/blah.txt
1068:
1069: You'll find that even if D:\blah.txt does exist, curl returns a 'file
1070: not found' error.
1071:
1072: According to RFC 1738 (https://www.ietf.org/rfc/rfc1738.txt),
1073: file:// URLs must contain a host component, but it is ignored by
1074: most implementations. In the above example, 'D:' is treated as the
1075: host component, and is taken away. Thus, curl tries to open '/blah.txt'.
1076: If your system is installed to drive C:, that will resolve to 'C:\blah.txt',
1077: and if that doesn't exist you will get the not found error.
1078:
1079: To fix this problem, use file:// URLs with *three* leading slashes:
1080:
1081: file:///D:/blah.txt
1082:
1083: Alternatively, if it makes more sense, specify 'localhost' as the host
1084: component:
1085:
1086: file://localhost/D:/blah.txt
1087:
1088: In either case, curl should now be looking for the correct file.
1089:
1090: 4.19 Why doesn't curl return an error when the network cable is unplugged?
1091:
1092: Unplugging a cable is not an error situation. The TCP/IP protocol stack
1093: was designed to be fault tolerant, so even though there may be a physical
1094: break somewhere the connection shouldn't be affected, just possibly
1095: delayed. Eventually, the physical break will be fixed or the data will be
1096: re-routed around the physical problem through another path.
1097:
1098: In such cases, the TCP/IP stack is responsible for detecting when the
1099: network connection is irrevocably lost. Since with some protocols it is
1100: perfectly legal for the client to wait indefinitely for data, the stack may
1101: never report a problem, and even when it does, it can take up to 20 minutes
1102: for it to detect an issue. The curl option --keepalive-time enables
1103: keep-alive support in the TCP/IP stack which makes it periodically probe the
1104: connection to make sure it is still available to send data. That should
1105: reliably detect any TCP/IP network failure.
1106:
1107: But even that won't detect the network going down before the TCP/IP
1108: connection is established (e.g. during a DNS lookup) or using protocols that
1109: don't use TCP. To handle those situations, curl offers a number of timeouts
1110: on its own. --speed-limit/--speed-time will abort if the data transfer rate
1111: falls too low, and --connect-timeout and --max-time can be used to put an
1112: overall timeout on the connection phase or the entire transfer.
1113:
1114: A libcurl-using application running in a known physical environment (e.g.
1115: an embedded device with only a single network connection) may want to act
1116: immediately if its lone network connection goes down. That can be achieved
1117: by having the application monitor the network connection on its own using an
1118: OS-specific mechanism, then signaling libcurl to abort (see also item 5.13).
1119:
1120: 4.20 curl doesn't return error for HTTP non-200 responses!
1121:
1122: Correct. Unless you use -f (--fail).
1123:
1124: When doing HTTP transfers, curl will perform exactly what you're asking it
1125: to do and if successful it will not return an error. You can use curl to
1126: test your web server's "file not found" page (that gets 404 back), you can
1127: use it to check your authentication protected web pages (that gets a 401
1128: back) and so on.
1129:
1130: The specific HTTP response code does not constitute a problem or error for
1131: curl. It simply sends and delivers HTTP as you asked and if that worked,
1132: everything is fine and dandy. The response code is generally providing more
1133: higher level error information that curl doesn't care about. The error was
1134: not in the HTTP transfer.
1135:
1136: If you want your command line to treat error codes in the 400 and up range
1137: as errors and thus return a non-zero value and possibly show an error
1138: message, curl has a dedicated option for that: -f (CURLOPT_FAILONERROR in
1139: libcurl speak).
1140:
1141: You can also use the -w option and the variable %{response_code} to extract
1142: the exact response code that was returned in the response.
1143:
1144: 4.21 Why is there a HTTP/1.1 in my HTTP/2 request?
1145:
1146: If you use verbose to see the HTTP request when you send off a HTTP/2
1147: request, it will still say 1.1.
1148:
1149: The reason for this is that we first generate the request to send using the
1150: old 1.1 style and show that request in the verbose output, and then we
1151: convert it over to the binary header-compressed HTTP/2 style. The actual
1152: "1.1" part from that request is then not actually used in the transfer.
1153: The binary HTTP/2 headers are not human readable.
1154:
1155: 5. libcurl Issues
1156:
1157: 5.1 Is libcurl thread-safe?
1158:
1159: Yes.
1160:
1161: We have written the libcurl code specifically adjusted for multi-threaded
1162: programs. libcurl will use thread-safe functions instead of non-safe ones if
1163: your system has such. Note that you must never share the same handle in
1164: multiple threads.
1165:
1166: There may be some exceptions to thread safety depending on how libcurl was
1167: built. Please review the guidelines for thread safety to learn more:
1168: https://curl.haxx.se/libcurl/c/threadsafe.html
1169:
1170: 5.2 How can I receive all data into a large memory chunk?
1171:
1172: [ See also the examples/getinmemory.c source ]
1173:
1174: You are in full control of the callback function that gets called every time
1175: there is data received from the remote server. You can make that callback do
1176: whatever you want. You do not have to write the received data to a file.
1177:
1178: One solution to this problem could be to have a pointer to a struct that you
1179: pass to the callback function. You set the pointer using the
1180: CURLOPT_WRITEDATA option. Then that pointer will be passed to the callback
1181: instead of a FILE * to a file:
1182:
1183: /* imaginary struct */
1184: struct MemoryStruct {
1185: char *memory;
1186: size_t size;
1187: };
1188:
1189: /* imaginary callback function */
1190: size_t
1191: WriteMemoryCallback(void *ptr, size_t size, size_t nmemb, void *data)
1192: {
1193: size_t realsize = size * nmemb;
1194: struct MemoryStruct *mem = (struct MemoryStruct *)data;
1195:
1196: mem->memory = (char *)realloc(mem->memory, mem->size + realsize + 1);
1197: if (mem->memory) {
1198: memcpy(&(mem->memory[mem->size]), ptr, realsize);
1199: mem->size += realsize;
1200: mem->memory[mem->size] = 0;
1201: }
1202: return realsize;
1203: }
1204:
1205: 5.3 How do I fetch multiple files with libcurl?
1206:
1207: libcurl has excellent support for transferring multiple files. You should
1208: just repeatedly set new URLs with curl_easy_setopt() and then transfer it
1209: with curl_easy_perform(). The handle you get from curl_easy_init() is not
1210: only reusable, but you're even encouraged to reuse it if you can, as that
1211: will enable libcurl to use persistent connections.
1212:
1213: 5.4 Does libcurl do Winsock initialization on win32 systems?
1214:
1215: Yes, if told to in the curl_global_init() call.
1216:
1217: 5.5 Does CURLOPT_WRITEDATA and CURLOPT_READDATA work on win32 ?
1218:
1219: Yes, but you cannot open a FILE * and pass the pointer to a DLL and have
1220: that DLL use the FILE * (as the DLL and the client application cannot access
1221: each others' variable memory areas). If you set CURLOPT_WRITEDATA you must
1222: also use CURLOPT_WRITEFUNCTION as well to set a function that writes the
1223: file, even if that simply writes the data to the specified FILE *.
1224: Similarly, if you use CURLOPT_READDATA you must also specify
1225: CURLOPT_READFUNCTION.
1226:
1227: 5.6 What about Keep-Alive or persistent connections?
1228:
1229: curl and libcurl have excellent support for persistent connections when
1230: transferring several files from the same server. Curl will attempt to reuse
1231: connections for all URLs specified on the same command line/config file, and
1232: libcurl will reuse connections for all transfers that are made using the
1233: same libcurl handle.
1234:
1235: When you use the easy interface the connection cache is kept within the easy
1236: handle. If you instead use the multi interface, the connection cache will be
1237: kept within the multi handle and will be shared among all the easy handles
1238: that are used within the same multi handle.
1239:
1240: 5.7 Link errors when building libcurl on Windows!
1241:
1242: You need to make sure that your project, and all the libraries (both static
1243: and dynamic) that it links against, are compiled/linked against the same run
1244: time library.
1245:
1246: This is determined by the /MD, /ML, /MT (and their corresponding /M?d)
1247: options to the command line compiler. /MD (linking against MSVCRT dll) seems
1248: to be the most commonly used option.
1249:
1250: When building an application that uses the static libcurl library, you must
1251: add -DCURL_STATICLIB to your CFLAGS. Otherwise the linker will look for
1252: dynamic import symbols. If you're using Visual Studio, you need to instead
1253: add CURL_STATICLIB in the "Preprocessor Definitions" section.
1254:
1255: If you get linker error like "unknown symbol __imp__curl_easy_init ..." you
1256: have linked against the wrong (static) library. If you want to use the
1257: libcurl.dll and import lib, you don't need any extra CFLAGS, but use one of
1258: the import libraries below. These are the libraries produced by the various
1259: lib/Makefile.* files:
1260:
1261: Target: static lib. import lib for libcurl*.dll.
1262: -----------------------------------------------------------
1263: MingW: libcurl.a libcurldll.a
1264: MSVC (release): libcurl.lib libcurl_imp.lib
1265: MSVC (debug): libcurld.lib libcurld_imp.lib
1266: Borland: libcurl.lib libcurl_imp.lib
1267:
1268: 5.8 libcurl.so.X: open failed: No such file or directory
1269:
1270: This is an error message you might get when you try to run a program linked
1271: with a shared version of libcurl and your run-time linker (ld.so) couldn't
1272: find the shared library named libcurl.so.X. (Where X is the number of the
1273: current libcurl ABI, typically 3 or 4).
1274:
1275: You need to make sure that ld.so finds libcurl.so.X. You can do that
1276: multiple ways, and it differs somewhat between different operating systems,
1277: but they are usually:
1278:
1279: * Add an option to the linker command line that specify the hard-coded path
1280: the run-time linker should check for the lib (usually -R)
1281:
1282: * Set an environment variable (LD_LIBRARY_PATH for example) where ld.so
1283: should check for libs
1284:
1285: * Adjust the system's config to check for libs in the directory where you've
1286: put the dir (like Linux's /etc/ld.so.conf)
1287:
1288: 'man ld.so' and 'man ld' will tell you more details
1289:
1290: 5.9 How does libcurl resolve host names?
1291:
1292: libcurl supports a large a number of different name resolve functions. One
1293: of them is picked at build-time and will be used unconditionally. Thus, if
1294: you want to change name resolver function you must rebuild libcurl and tell
1295: it to use a different function.
1296:
1297: - The non-IPv6 resolver that can use one of four different host name resolve
1298: calls (depending on what your system supports):
1299:
1300: A - gethostbyname()
1301: B - gethostbyname_r() with 3 arguments
1302: C - gethostbyname_r() with 5 arguments
1303: D - gethostbyname_r() with 6 arguments
1304:
1305: - The IPv6-resolver that uses getaddrinfo()
1306:
1307: - The c-ares based name resolver that uses the c-ares library for resolves.
1308: Using this offers asynchronous name resolves.
1309:
1310: - The threaded resolver (default option on Windows). It uses:
1311:
1312: A - gethostbyname() on plain IPv4 hosts
1313: B - getaddrinfo() on IPv6 enabled hosts
1314:
1315: Also note that libcurl never resolves or reverse-lookups addresses given as
1316: pure numbers, such as 127.0.0.1 or ::1.
1317:
1318: 5.10 How do I prevent libcurl from writing the response to stdout?
1319:
1320: libcurl provides a default built-in write function that writes received data
1321: to stdout. Set the CURLOPT_WRITEFUNCTION to receive the data, or possibly
1322: set CURLOPT_WRITEDATA to a different FILE * handle.
1323:
1324: 5.11 How do I make libcurl not receive the whole HTTP response?
1325:
1326: You make the write callback (or progress callback) return an error and
1327: libcurl will then abort the transfer.
1328:
1329: 5.12 Can I make libcurl fake or hide my real IP address?
1330:
1331: No. libcurl operates on a higher level. Besides, faking IP address would
1332: imply sending IP packets with a made-up source address, and then you normally
1333: get a problem with receiving the packet sent back as they would then not be
1334: routed to you!
1335:
1336: If you use a proxy to access remote sites, the sites will not see your local
1337: IP address but instead the address of the proxy.
1338:
1339: Also note that on many networks NATs or other IP-munging techniques are used
1340: that makes you see and use a different IP address locally than what the
1341: remote server will see you coming from. You may also consider using
1342: https://www.torproject.org/ .
1343:
1344: 5.13 How do I stop an ongoing transfer?
1345:
1346: With the easy interface you make sure to return the correct error code from
1347: one of the callbacks, but none of them are instant. There is no function you
1348: can call from another thread or similar that will stop it immediately.
1349: Instead, you need to make sure that one of the callbacks you use returns an
1350: appropriate value that will stop the transfer. Suitable callbacks that you
1351: can do this with include the progress callback, the read callback and the
1352: write callback.
1353:
1354: If you're using the multi interface, you can also stop a transfer by
1355: removing the particular easy handle from the multi stack at any moment you
1356: think the transfer is done or when you wish to abort the transfer.
1357:
1358: 5.14 Using C++ non-static functions for callbacks?
1359:
1360: libcurl is a C library, it doesn't know anything about C++ member functions.
1361:
1362: You can overcome this "limitation" with relative ease using a static
1363: member function that is passed a pointer to the class:
1364:
1365: // f is the pointer to your object.
1366: static size_t YourClass::func(void *buffer, size_t sz, size_t n, void *f)
1367: {
1368: // Call non-static member function.
1369: static_cast<YourClass*>(f)->nonStaticFunction();
1370: }
1371:
1372: // This is how you pass pointer to the static function:
1373: curl_easy_setopt(hcurl, CURLOPT_WRITEFUNCTION, YourClass::func);
1374: curl_easy_setopt(hcurl, CURLOPT_WRITEDATA, this);
1375:
1376: 5.15 How do I get an FTP directory listing?
1377:
1378: If you end the FTP URL you request with a slash, libcurl will provide you
1379: with a directory listing of that given directory. You can also set
1380: CURLOPT_CUSTOMREQUEST to alter what exact listing command libcurl would use
1381: to list the files.
1382:
1383: The follow-up question tends to be how is a program supposed to parse the
1384: directory listing. How does it know what's a file and what's a dir and what's
1385: a symlink etc. If the FTP server supports the MLSD command then it will
1386: return data in a machine-readable format that can be parsed for type. The
1387: types are specified by RFC3659 section 7.5.1. If MLSD is not supported then
1388: you have to work with what you're given. The LIST output format is entirely
1389: at the server's own liking and the NLST output doesn't reveal any types and
1390: in many cases doesn't even include all the directory entries. Also, both LIST
1391: and NLST tend to hide unix-style hidden files (those that start with a dot)
1392: by default so you need to do "LIST -a" or similar to see them.
1393:
1394: Example - List only directories.
1395: ftp.funet.fi supports MLSD and ftp.kernel.org does not:
1396:
1397: curl -s ftp.funet.fi/pub/ -X MLSD | \
1398: perl -lne 'print if s/(?:^|;)type=dir;[^ ]+ (.+)$/$1/'
1399:
1400: curl -s ftp.kernel.org/pub/linux/kernel/ | \
1401: perl -lne 'print if s/^d[-rwx]{9}(?: +[^ ]+){7} (.+)$/$1/'
1402:
1403: If you need to parse LIST output in libcurl one such existing
1404: list parser is available at https://cr.yp.to/ftpparse.html Versions of
1405: libcurl since 7.21.0 also provide the ability to specify a wildcard to
1406: download multiple files from one FTP directory.
1407:
1408: 5.16 I want a different time-out!
1409:
1410: Time and time again users realize that CURLOPT_TIMEOUT and
1411: CURLOPT_CONNECTIMEOUT are not sufficiently advanced or flexible to cover all
1412: the various use cases and scenarios applications end up with.
1413:
1414: libcurl offers many more ways to time-out operations. A common alternative
1415: is to use the CURLOPT_LOW_SPEED_LIMIT and CURLOPT_LOW_SPEED_TIME options to
1416: specify the lowest possible speed to accept before to consider the transfer
1417: timed out.
1418:
1419: The most flexible way is by writing your own time-out logic and using
1420: CURLOPT_XFERINFOFUNCTION (perhaps in combination with other callbacks) and
1421: use that to figure out exactly when the right condition is met when the
1422: transfer should get stopped.
1423:
1424: 5.17 Can I write a server with libcurl?
1425:
1426: No. libcurl offers no functions or building blocks to build any kind of
1427: internet protocol server. libcurl is only a client-side library. For server
1428: libraries, you need to continue your search elsewhere but there exist many
1429: good open source ones out there for most protocols you could possibly want a
1430: server for. And there are really good stand-alone ones that have been tested
1431: and proven for many years. There's no need for you to reinvent them!
1432:
1433: 5.18 Does libcurl use threads?
1434:
1435: Put simply: no, libcurl will execute in the same thread you call it in. All
1436: callbacks will be called in the same thread as the one you call libcurl in.
1437:
1438: If you want to avoid your thread to be blocked by the libcurl call, you make
1439: sure you use the non-blocking API which will do transfers asynchronously -
1440: but still in the same single thread.
1441:
1442: libcurl will potentially internally use threads for name resolving, if it
1443: was built to work like that, but in those cases it'll create the child
1444: threads by itself and they will only be used and then killed internally by
1445: libcurl and never exposed to the outside.
1446:
1447: 6. License Issues
1448:
1449: Curl and libcurl are released under a MIT/X derivate license. The license is
1450: very liberal and should not impose a problem for your project. This section
1451: is just a brief summary for the cases we get the most questions. (Parts of
1452: this section was much enhanced by Bjorn Reese.)
1453:
1454: We are not lawyers and this is not legal advice. You should probably consult
1455: one if you want true and accurate legal insights without our prejudice. Note
1456: especially that this section concerns the libcurl license only; compiling in
1457: features of libcurl that depend on other libraries (e.g. OpenSSL) may affect
1458: the licensing obligations of your application.
1459:
1460: 6.1 I have a GPL program, can I use the libcurl library?
1461:
1462: Yes!
1463:
1464: Since libcurl may be distributed under the MIT/X derivate license, it can be
1465: used together with GPL in any software.
1466:
1467: 6.2 I have a closed-source program, can I use the libcurl library?
1468:
1469: Yes!
1470:
1471: libcurl does not put any restrictions on the program that uses the library.
1472:
1473: 6.3 I have a BSD licensed program, can I use the libcurl library?
1474:
1475: Yes!
1476:
1477: libcurl does not put any restrictions on the program that uses the library.
1478:
1479: 6.4 I have a program that uses LGPL libraries, can I use libcurl?
1480:
1481: Yes!
1482:
1483: The LGPL license doesn't clash with other licenses.
1484:
1485: 6.5 Can I modify curl/libcurl for my program and keep the changes secret?
1486:
1487: Yes!
1488:
1489: The MIT/X derivate license practically allows you to do almost anything with
1490: the sources, on the condition that the copyright texts in the sources are
1491: left intact.
1492:
1493: 6.6 Can you please change the curl/libcurl license to XXXX?
1494:
1495: No.
1496:
1497: We have carefully picked this license after years of development and
1498: discussions and a large amount of people have contributed with source code
1499: knowing that this is the license we use. This license puts the restrictions
1500: we want on curl/libcurl and it does not spread to other programs or
1501: libraries that use it. It should be possible for everyone to use libcurl or
1502: curl in their projects, no matter what license they already have in use.
1503:
1504: 6.7 What are my obligations when using libcurl in my commercial apps?
1505:
1506: Next to none. All you need to adhere to is the MIT-style license (stated in
1507: the COPYING file) which basically says you have to include the copyright
1508: notice in "all copies" and that you may not use the copyright holder's name
1509: when promoting your software.
1510:
1511: You do not have to release any of your source code.
1512:
1513: You do not have to reveal or make public any changes to the libcurl source
1514: code.
1515:
1516: You do not have to broadcast to the world that you are using libcurl within
1517: your app.
1518:
1519: All we ask is that you disclose "the copyright notice and this permission
1520: notice" somewhere. Most probably like in the documentation or in the section
1521: where other third party dependencies already are mentioned and acknowledged.
1522:
1523: As can be seen here: https://curl.haxx.se/docs/companies.html and elsewhere,
1524: more and more companies are discovering the power of libcurl and take
1525: advantage of it even in commercial environments.
1526:
1527:
1528: 7. PHP/CURL Issues
1529:
1530: 7.1 What is PHP/CURL?
1531:
1532: The module for PHP that makes it possible for PHP programs to access curl-
1533: functions from within PHP.
1534:
1535: In the cURL project we call this module PHP/CURL to differentiate it from
1536: curl the command line tool and libcurl the library. The PHP team however
1537: does not refer to it like this (for unknown reasons). They call it plain
1538: CURL (often using all caps) or sometimes ext/curl, but both cause much
1539: confusion to users which in turn gives us a higher question load.
1540:
1541: 7.2 Who wrote PHP/CURL?
1542:
1543: PHP/CURL was initially written by Sterling Hughes.
1544:
1545: 7.3 Can I perform multiple requests using the same handle?
1546:
1547: Yes - at least in PHP version 4.3.8 and later (this has been known to not
1548: work in earlier versions, but the exact version when it started to work is
1549: unknown to me).
1550:
1551: After a transfer, you just set new options in the handle and make another
1552: transfer. This will make libcurl re-use the same connection if it can.
1553:
1554: 7.4 Does PHP/CURL have dependencies?
1555:
1556: PHP/CURL is a module that comes with the regular PHP package. It depends on
1557: and uses libcurl, so you need to have libcurl installed properly before
1558: PHP/CURL can be used.
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/cvsweb/back.gif){kind=icon}
Return to FAQ CVS log ![[TXT]](/icons/cvsweb/text.gif){kind=icon}
![[DIR]](/icons/cvsweb/dir.gif){kind=icon}
Up to [ELWIX - Embedded LightWeight unIX -] / embedaddon / curl / docs