GNURL(1)
okay bestie, welcome to gnurl(1) from the General Commands Manual — yep, the manual name is folded in right here at the top instead of being copy-pasted as its own line. let’s get into it 💅
💖 NAME 💖
gnurl — transfer a URL. that’s the whole name section and honestly it ate.
✨ SYNOPSIS ✨
the synopsis is giving “every flag at once” energy, so here it is verbatim, no cap:
gnurl [--abstract-unix-socket <path>] [--alt-svc <file name>]
[--anyauth] [-a | --append] [--basic] [--cacert <file>]
[--capath <dir>] [--cert-status] [--cert-type <type>]
[-E <certificate[:password]> | --cert <certificate[:password]>]
[--ciphers <list of ciphers>] [--compressed-ssh] [--compressed]
[-K <file> | -config <file>] [--connect-timeout <seconds>]
[--connect-to <HOST1:PORT1:HOST2:PORT2>]
[-C <offset> | --continue-at <offset>]
[-c <filename> | --cookie-jar <filename>]
[-b <data|filename> | --cookie <data|filename>] [--create-dirs]
[--crlf] [--crlfile <file>] [--data-binary <data>]
[--data-raw <data>] [--data-urlencode <data>]
[-d data | --data data] [--delegation LEVEL] [--digest]
[--disable-eprt] [--disable-epsv] [-q | --disable]
[--dns-interface <interface>] [--dns-ipv4-addr <address>]
[--dns-ipv6-addr <address>] [--dns-servers <addresses>]
[--doh-url -<URL>] [-D <filename> | --dump-header <filename>]
[--egd-file <file>] [--engine <name>]
[--expect100-timeout <seconds>] [--fail-early] [-f | --fail]
[--false-start] [--form-string <name=string>]
[-F <name=content> | -form <name=content>] [--ftp-account <data>]
[--ftp-alternative-to-user <command>] [--ftp-create-dirs]
[--ftp-method <method>] [--ftp-pasv]
[-P <address> | -ftp-port <address>] [--ftp-pret]
[--ftp-skip-pasv-ip] [--ftp-ssl-ccc-mode <active/passive>]
[--ftp-ssl-ccc] [--ftp-ssl-control] [-G | --get] [-g | --globoff]
[--happy-eyeballs-timeout-ms milliseconds] [--haproxy-protocol]
[-I | --head] [-h | --help] [--hostpubmd5 <md5>] [--http0.9]
[-0 | --http1.0] [--http1.1] [--http2-prior-knowledge] [--http2]
[--ignore-content-length] [-i | --include] [-k | --insecure]
[--interface <name>] [-4 | --ipv4] [-6 | --ipv6]
[-j | --junk-session-cookies] [--key-type <type>] [--key <key>]
[--krb <level>] [--libcurl <file>] [--limit-rate <speed>]
[-l | --list-only] [--location-trusted] [-L | --location]
[--login-options <options>] [--mail-auth <address>]
[--mail-from <address>] [--mail-rcpt <address>] [-M | --manual]
[--max-filesize <bytes>] [--max-redirs <num>] [--metalink]
[--negotiate] [--netrc-file filename] [--netrc-optional]
[-n | --netrc] [-: | --next] [--no-alpn] [-N | --no-buffer]
[--no-keepalive] [--no-npn] [--no-sessionid]
[--noproxy no-proxy-list] [--ntlm-wb] [--ntlm]
[--oauth2-bearer token] [-o FILE | --output FILE] [--pass phrase]
[--path-as-is] [--pinnedpubkey <hashes>] [--post301] [--post302]
[--post303] [--preproxy [protocol://]host[:port]]
[-# | --progress-bar] [--proto-default <protocol>]
[--proto-redir <protocols>] [--proto <protocols>]
[--proxy-anyauth] [--proxy-basic] [--proxy-cacert <file>]
[--proxy-capath <dir>] [--proxy-cert-type <type>]
[--proxy-cert <cert[:passwd]>] [--proxy-ciphers <list>]
[--proxy-crlfile <file>] [--proxy-digest]
[--proxy-header <header/@file>] [--proxy-insecure]
[--proxy-key-type <type>] [--proxy-key <key>] [--proxy-negotiate]
[--proxy-ntlm] [--proxy-pass <phrase>]
[--proxy-pinnedpubkey <hashes>] [--proxy-service-name <name>]
[--proxy-ssl-allow-beast]
[--proxy-tls13-ciphers <ciphersuite list>]
[--proxy-tlsauthtype <type>] [--proxy-tlspassword <string>]
[--proxy-tlsuser <name>] [--proxy-tlsv1]
[-U <user:password> | --proxy-user <user:password>]
[-x [protocol://]host[:port] | -proxy [protocol://]host[:port]]
[--proxy1.0 <host[:port]>] [-p | --proxytunnel] [--pubkey <key>]
[-Q | --quote] [--random-file <file>]
[-r <range> | --range <range>] [--raw] [-e | --referer <URL>]
[-J | --remote-header-name] [--remote-name-all]
[-O | --remote-name] [-R | --remote-time] [--request-target]
[-X command | -request <command>]
[--resolve <host:port:address[,address]...>] [--retry-connrefused]
[--retry-delay <seconds>] [--retry-max-time <seconds>]
[--retry <num>] [--sasl-ir] [--service-name <name>]
[-S | --show-error] [-s | --silent] [--socks4 <host[:port]>]
[--socks4a <host[:port]>] [--socks5-basic] [--socks5-gssapi-nec]
[--socks5-gssapi-service <name>] [--socks5-gssapi]
[--socks5-hostname <host[:port]>] [--socks5 <host[:port]>]
[-Y speed | --speed-limit <speed>]
[--y seconds | -speed-time <seconds>] [--ssl-allow-beast]
[--ssl-no-revoke] [--ssl-reqd] [--ssl] [-2 | --sslv2]
[-3 | --sslv3] [--stderr] [--styled-output]
[--suppress-connect-headers] [--tcp-fastopen] [--tcp-nodelay]
[-t opt=val | --telnet-option <opt=val>] [--tftp-blksize <value>]
[--tftp-no-options] [-z -time | --time-cond <time>]
[--tls-max <VERSION>]
[--tls13-ciphers <list of TLS 1.3 ciphersuites>]
[--tlsauthtype <type>] [--tlspassword] [--tlsuser <name>]
[--tlsv1.0] [--tlsv1.1] [--tlsv1.2] [--tlsv1.3] [-1 | --tlsv1]
[--tr-encoding] [--trace-ascii <file>] [--trace-time]
[--trace <file>] [-T file | --upload-file <file>] [--url <url>]
[-B | --use-ascii] [-A <name> | --user-agent <name>]
[-u <user:password> | --user <user:password>] [-v | --verbose]
[-V | --version] [-w <format> | --write-out <format>] [--xattr]
⟨URLs⟩
🦋 DESCRIPTION 🦋
okay so gnurl is a tool to transfer data from or to a server, using one of the supported protocols: DICT, FILE, FTP, FTPS, GOPHER, HTTP, HTTPS, IMAP, IMAPS, LDAP, LDAPS, POP3, POP3S, RTMP, RTSP, SCP, SFTP, SMB, SMBS, SMTP, SMTPS, TELNET and TFTP. the command is designed to work without user interaction. main character energy: it just does the thing.
gnurl offers a busload of useful tricks like proxy support, user authentication, FTP upload, HTTP post, SSL connections, cookies, file transfer resume, Metalink, and more. as you will see below, the number of features will make your head spin! literally, there are so many.
gnurl is powered by libgnurl for all transfer-related features. see libcurl(3) for details. 🫶
🌈 URL 🌈
the URL syntax is protocol-dependent. you’ll find a detailed description in RFC 3986.
you can specify multiple URLs or parts of URLs by writing part sets within braces as in:
http://site.{one,two,three}.com
or you can get sequences of alphanumeric series by using [] as in:
ftp://ftp.example.com/file[1-100].txt
ftp://ftp.example.com/file[001-100].txt (with leading zeros)
ftp://ftp.example.com/file[a-z].txt
nested sequences are not supported, but you can use several ones next to each other:
http://example.com/archive[1996-1999]/vol[1-4]/part{a,b,c}.html
you can specify any amount of URLs on the command line. they will be fetched in a sequential manner in the specified order. you can specify command line options and URLs mixed and in any order on the command line. iconic behavior, honestly.
you can specify a step counter for the ranges to get every Nth number or letter:
http://example.com/file[1-100:10].txt
http://example.com/file[a-z:2].txt
when using [] or {} sequences when invoked from a command line prompt, you probably have to put the full URL within double quotes to avoid the shell from interfering with it. this also goes for other characters treated special, like for example '&', '?' and '*'.
provide the IPv6 zone index in the URL with an escaped percentage sign and the interface name. like in:
http://[fe80::3%25eth0]/
if you specify URL without protocol:// prefix, curl will attempt to guess what protocol you might want. it will then default to HTTP but try other protocols based on often-used host name prefixes. for example, for host names starting with "ftp." curl will assume you want to speak FTP.
curl will do its best to use what you pass to it as a URL. it is not trying to validate it as a syntactically correct URL by any means but is instead very liberal with what it accepts.
curl will attempt to re-use connections for multiple file transfers, so that getting many files from the same server will not do multiple connects / handshakes. this improves speed. of course this is only done on files specified on a single command line and cannot be used between separate curl invokes.
⏳ PROGRESS METER ⏳
gnurl normally displays a progress meter during operations, indicating the amount of transferred data, transfer speeds and estimated time left, etc. the progress meter displays number of bytes and the speeds are in bytes per second. the suffixes (k, M, G, T, P) are 1024 based. for example 1k is 1024 bytes. 1M is 1048576 bytes.
curl displays this data to the terminal by default, so if you invoke curl to do an operation and it is about to write data to the terminal, it disables the progress meter as otherwise it would mess up the output mixing progress meter and response data.
if you want a progress meter for HTTP POST or PUT requests, you need to redirect the response output to a file, using shell redirect (>), -o | --output or similar.
it is not the same case for FTP upload as that operation does not spit out any response data to the terminal.
if you prefer a progress “bar” instead of the regular meter, -# | --progress-bar is your friend. you can also disable the progress meter completely with the -s | --silent option.
✨ OPTIONS ✨
options start with one or two dashes. many of the options require an additional value next to them.
the short “single-dash” form of the options, -d for example, may be used with or without a space between it and its value, although a space is a recommended separator. the long “double-dash” form, -d | --data for example, requires a space between it and its value.
short version options that don’t need any additional values can be used immediately next to each other, like for example you can specify all the options -O, -L and -v at once as -OLv.
in general, all boolean options are enabled with --option and yet again disabled with --no-option. that is, you use the exact same option name but prefix it with "no-". however, in this list we mostly only list and show the --option version of them. (this concept with --no options was added in 7.19.0. previously most options were toggled on/off on repeated use of the same command line option.)
--abstract-unix-socket <path>— (HTTP) connect through an abstract Unix domain socket, instead of using the network. Note:netstatshows the path of an abstract socket prefixed with'@', however the<path>argument should not have this leading character.
Added in 7.53.0.
✨ OPTIONS ✨
--alt-svc <file name>— (HTTPS) ⚠️ okay so first thing: this option is experimental. Do not use in production. fr fr, not a drill.
It turns on the alt-svc parser in curl. If the file name you give points at an alt-svc cache file that already exists, curl uses it. Then after a transfer finishes, the cache gets saved back to that same file name — but only if it got modified. ✨
Wanna keep it all in memory and touch zero files? Hand it a "" file name (zero length, like literally empty) and curl skips loading and saving, just handling the cache in memory. 🧠
Use it a bunch of times and curl loads contents from every single file you listed — but the last one is the one used for saving.
Added in 7.64.1.
--anyauth— (HTTP) tells curl to figure out the authentication method all by its own self 👑 and then use the most secure one the remote site claims to support. The way it does this is: do a request first, check the response headers, and then pick — which means possibly an extra network round-trip. So it's the "instead of" option for setting a specific method yourself, which you'd do with--basic,--digest,--ntlm, and--negotiate.
🫶 Heads up bestie: --anyauth is not recommended if you do uploads from stdin. It may need the data sent twice, and then the client has to be able to rewind. If that need arises while uploading from stdin, the upload operation will fail.
Used together with -u | --user.
See also --proxy-anyauth, --basic and --digest.
-a|--append— (FTP SFTP) in an upload, this makes curl append to the target file instead of overwriting it. Slay. If the remote file doesn't exist, it gets created. Just know that some SFTP servers ignore this flag entirely (including OpenSSH) 😭.--basic— (HTTP) tells curl to use HTTP Basic authentication with the remote host. This is the default, so honestly this option is usually pointless — unless you're using it to override a previously set option that set a different auth method (like--ntlm,--digest, or--negotiate). Then it's doing actual work 💅.
Used together with -u | --user.
See also --proxy-basic.
--cacert <file>— (TLS) tells curl to use the specified certificate file to verify the peer. That file can hold multiple CA certificates, and the cert(s) must be in PEM format. Normally curl is built to use a default file for this, so this option is typically used to alter that default file.
curl recognizes the environment variable named CURL_CA_BUNDLE if it's set, and uses the given path as a path to a CA cert bundle. This option overrides that variable 👑
The windows version of curl will automatically look for a CA certs file named curl-ca-bundle.crt, either in the same directory as curl.exe, or in the Current Working Directory, or in any folder along your PATH.
If curl is built against the NSS SSL library, the NSS PEM PKCS#11 module (libnsspem.so) needs to be available for this option to work properly.
(iOS and macOS only) If curl is built against Secure Transport, then this option is supported for backward compatibility with other SSL engines, but it should not be set. If the option is not set, then curl uses the certificates in the system and user Keychain to verify the peer, which is the preferred method of verifying the peer's certificate chain. 🍓
(Schannel only) This option is supported for Schannel in Windows 7 or later with libcurl 7.60 or later. It's supported for backward compatibility with other SSL engines; instead it's recommended to use Windows' store of root certificates (the default for Schannel).
If this option is used several times, the last one will be used.
--capath <dir>— (TLS) tells curl to use the specified certificate directory to verify the peer. Multiple paths can be provided by separating them with":"(e.g."path1:path2:path3"). The certificates must be in PEM format, and if curl is built against OpenSSL, the directory must have been processed using thec_rehashutility supplied with OpenSSL. Using--capathcan let OpenSSL-powered curl make SSL-connections much more efficiently than using--cacertwhen the--cacertfile contains a whole lot of CA certificates. efficiency icon 💅
If this option is set, the default capath value will be ignored, and if it's used several times, the last one will be used.
--cert-status— (TLS) tells curl to verify the status of the server certificate using the Certificate Status Request TLS extension (aka. OCSP stapling).
If this option is enabled and the server sends an invalid (e.g. expired) response, or the response suggests the server certificate has been revoked, or no response at all is received — the verification fails. 😭 no cap.
This is currently only implemented in the OpenSSL, GnuTLS and NSS backends.
Added in 7.41.0.
--cert-type <type>— (TLS) tells curl what type the provided client certificate is using.PEM,DER,ENGandP12are the recognized types. If not specified,PEMis assumed.
If this option is used several times, the last one will be used.
See also -E | --cert and --key and --key-type.
-E <certificate[:password]>|--cert <certificate[:password]>— (TLS) tells curl to use the specified client certificate file when getting a file with HTTPS, FTPS or another SSL-based protocol.
The certificate must be in PKCS#12 format if using Secure Transport, or PEM format if using any other engine. If the optional password isn't specified, it will be queried for on the terminal 🔐. Note that this option assumes a "certificate" file that is the private key and the client certificate concatenated! See -E, --cert and --key to specify them independently.
If curl is built against the NSS SSL library then this option can tell curl the nickname of the certificate to use within the NSS database defined by the environment variable SSL_DIR (or by default /etc/pki/nssdb). If the NSS PEM PKCS#11 module (libnsspem.so) is available then PEM files may be loaded. If you want to use a file from the current directory, please precede it with a "./" prefix, in order to avoid confusion with a nickname. If the nickname contains ":", it needs to be preceded by "\" so that it is not recognized as password delimiter. If the nickname contains "\", it needs to be escaped as "\\" so that it is not recognized as an escape character. so many slashes 😵💫
If curl is built against OpenSSL library, and the engine pkcs11 is available, then a PKCS#11 URI (RFC 7512) can be used to specify a certificate located in a PKCS#11 device. A string beginning with "pkcs11:" will be interpreted as a PKCS#11 URI. If a PKCS#11 URI is provided, then the --engine option will be set as "pkcs11" if none was provided and the --cert-type option will be set as "ENG" if none was provided ✨
(iOS and macOS only) If curl is built against Secure Transport, then the certificate string can either be the name of a certificate/private key in the system or user keychain, or the path to a PKCS#12-encoded certificate and private key. If you want to use a file from the current directory, please precede it with a "./" prefix, in order to avoid confusion with a nickname.
(Schannel only) Client certificates must be specified by a path expression to a certificate store. (Loading PFX is not supported; you can import it to a store first). You can use "<store location>\<store name>\<thumbprint>" to refer to a certificate in the system certificates store, for example, "CurrentUser\MY\934a7ac6f8a5d579285a74fa61e19f23ddfe8d7a". Thumbprint is usually a SHA-1 hex string which you can see in certificate details 🔍. Following store locations are supported: CurrentUser, LocalMachine, CurrentService, Services, CurrentUserGroupPolicy, LocalMachineGroupPolicy, LocalMachineEnterprise.
If this option is used several times, the last one will be used.
See also --cert-type and --key and --key-type.
--ciphers <list of ciphers>— (TLS) specifies which ciphers to use in the connection. The list of ciphers must specify valid ciphers. Read up on SSL cipher list details on this URL:
https://curl.haxx.se/docs/ssl-ciphers.html
If this option is used several times, the last one will be used.
--compressed-ssh— (SCP SFTP) enables built-in SSH compression 🗜️. Important vibe check though: this is a request, not an order — the server may or may not do it. iykyk.
Added in 7.56.0.
--compressed— (HTTP) requests a compressed response using one of the algorithms curl supports, and saves the uncompressed document. If this option is used and the server sends an unsupported encoding, curl will report an error 🚨.-K <file>|-config <file>— specify a text file to read curl arguments from. The command line arguments found in the text file will be used as if they were provided on the command line. it's giving "remote control but for flags" 🎀
Options and their parameters must be specified on the same line in the file, separated by whitespace, colon, or the equals sign. Long option names can optionally be given in the config file without the initial double dashes and if so, the colon or equals characters can be used as separators. If the option is specified with one or two dashes, there can be no colon or equals character between the option and its parameter.
If the parameter contains whitespace (or starts with : or =), the parameter must be enclosed within quotes. Within double quotes, the following escape sequences are available: \\, \", \t, \n, \r and \v. A backslash preceding any other letter is ignored. If the first column of a config line is a '#' character, the rest of the line will be treated as a comment. Only write one option per physical line in the config file.
Specify the filename to -K -|-config as '-' to make curl read the file from stdin 🫶
Like, if you wanna put a URL in the config file you have to do it with the --url option — you can't just yeet a bare URL onto its own line, okay? So it's giving something like this:
url = https://curl.haxx.se/docs/
When curl gets invoked, it checks for a default config file and uses it if it finds one — unless you're using -q | --disable, in which case it minds its own business. The default config file gets checked in these places, in this order:
- curl tries to find the "home dir": it first checks for the
CURL_HOMEand then theHOMEenvironment variables. Failing that, it usesgetpwuid()on Unix-like systems (which returns the home directory given the current user in your system). On Windows, it then checks for theAPPDATAvariable, or as a last resort the'%USERPROFILE%\Application Data'. - On windows, if there is no
_curlrcfile in the home dir, it checks for one in the same dir the curl executable is placed. On Unix-like systems, it will simply try to load.curlrcfrom the determined home dir.
Like, here's a whole example file so you can see the vibe:
# --- Example file ---
# this is a comment
url = "example.com"
output = "curlhere.html"
user-agent = "superagent/1.0"
# and fetch another URL too
url = "example.com/docs/manpage.html"
-O
referer = "http://nowhereatall.example.com/"
# --- End of example file ---
This option can be used multiple times to load multiple config files. Stack 'em up, bestie 💅
--connect-timeout<seconds>— the maximum time in seconds you're willing to let curl's connection take. This only limits the connection phase, so if curl connects within the given period it will continue — and if not, it will exit. Since version 7.32.0, this option accepts decimal values 🎀
If this option is used several times, the last one will be used.
See also -m | --max-time.
--connect-to<HOST1:PORT1:HOST2:PORT2>— for a request to the givenHOST1:PORT1pair, connect toHOST2:PORT2instead. This option is suitable to direct requests at a specific server, e.g. at a specific cluster node in a cluster of servers. This option is only used to establish the network connection. It does NOT affect the hostname/port that is used for TLS/SSL (e.g. SNI, certificate verification) or for the application protocols. "HOST1" and "PORT1" may be the empty string, meaning "any host/port". "HOST2" and "PORT2" may also be the empty string, meaning "use the request's original host/port" ✨
A "host" specified to this option is compared as a string, so it needs to match the name used in request URL. It can be either numerical such as "127.0.0.1" or the full host name such as "example.org".
This option can be used many times to add many connect rules.
See also --resolve and -H | --header. Added in 7.49.0.
-C<offset>|--continue-at<offset>— continue/resume a previous file transfer at the given offset 🫶 The given offset is the exact number of bytes that will be skipped, counting from the beginning of the source file before it is transferred to the destination. If used with uploads, the FTP server commandSIZEwill not be used by curl.
Use -C - to tell curl to automatically find out where/how to resume the transfer. It then uses the given output/input files to figure that out.
If this option is used several times, the last one will be used.
See also -r | --range.
-c<filename>|--cookie-jar<filename>— (HTTP) specify to which file you want curl to write all cookies after a completed operation. Curl writes all cookies from its in-memory cookie storage to the given file at the end of operations. If no cookies are known, no data will be written. The file will be written using the Netscape cookie file format. If you set the file name to a single dash,"-", the cookies will be written to stdout 💖
This command line option will activate the cookie engine that makes curl record and use cookies. Another way to activate it is to use the -b | --cookie option.
Okay so, warning time: if the cookie jar can't be created or written to, the whole curl operation won't fail or even report an error clearly. Using-v|--verbosewill get a warning displayed, but that is the only visible feedback you get about this possibly lethal situation 😭 Like, genuinely, that's it.
If this option is used several times, the last specified file name will be used.
-b<data|filename>|--cookie<data|filename>— (HTTP) pass the data to the HTTP server in the Cookie header. It is supposedly the data previously received from the server in a"Set-Cookie:"line. The data should be in the format"NAME1=VALUE1; NAME2=VALUE2".
If no '=' symbol is used in the argument, it is instead treated as a filename to read previously stored cookie from. This option also activates the cookie engine which will make curl record incoming cookies, which may be handy if you're using this in combination with the -L | --location option or do multiple URL transfers on the same invoke. If the file name is exactly a minus ("-"), curl will instead read the contents from stdin.
The file format of the file to read cookies from should be plain HTTP headers (Set-Cookie style) or the Netscape/Mozilla cookie file format.
The file specified with -b | --cookie is only used as input. No cookies will be written to the file. To store cookies, use the -c | --cookie-jar option 🌸
Exercise caution if you are using this option and multiple transfers may occur. If you use the NAME1=VALUE1; format, or in a file use the Set-Cookie format and don't specify a domain, then the cookie is sent for any domain (even after redirects are followed) and cannot be modified by a server-set cookie. If the cookie engine is enabled and a server sets a cookie of the same name then both will be sent on a future transfer to that server, likely not what you intended. To address these issues set a domain in Set-Cookie (doing that will include sub domains) or use the Netscape format 💀
If this option is used several times, the last one will be used.
Users very often want to both read cookies from a file and write updated cookies back to a file, so using both -b | --cookie and -c | --cookie-jar in the same command line is common. Iconic combo, honestly 👑
--create-dirs— when used in conjunction with the-o|--outputoption, curl will create the necessary local directory hierarchy as needed. This option creates the dirs mentioned with the-o|--outputoption, nothing else. If the--outputfile name uses no dir or if the dirs it mentions already exist, no dir will be created.
To create remote directories when using FTP or SFTP, try --ftp-create-dirs.
--crlf— (FTP SMTP) convert LF to CRLF in upload. Useful for MVS (OS/390) 🦋
(SMTP added in 7.40.0)
--crlfile<file>— (TLS) provide a file using PEM format with a Certificate Revocation List that may specify peer certificates that are to be considered revoked ✨
If this option is used several times, the last one will be used.
Added in 7.19.7.
--data-ascii<data>— (HTTP) this is just an alias for-d|--data. Same thing, different name, no cap 💅--data-binary<data>— (HTTP) this posts data exactly as specified with no extra processing whatsoever. Zero notes, zero vibes, just the raw data 🎀
If you start the data with the letter @, the rest should be a filename. Data is posted in a similar manner as -d | --data does, except that newlines and carriage returns are preserved and conversions are never done.
Like -d | --data the default content-type sent to the server is application/x-www-form-urlencoded. If you want the data to be treated as arbitrary binary data by the server then set the content-type to octet-stream:
-H Content-Type: application/octet-stream.
If this option is used several times, the ones following the first will append data as described in -d | --data.
--data-raw<data>— (HTTP) this posts data similarly to-d|--databut without the special interpretation of the@character 🙈 So your@can just chill there being itself.
See also -d, --data. Added in 7.43.0.
--data-urlencode<data>— (HTTP) this posts data, similar to the other-d,--dataoptions with the exception that this performs URL-encoding 🌈
To be CGI-compliant, the <data> part should begin with a name followed by a separator and a content specification. The <data> part can be passed to curl using one of the following syntaxes:
content— this will make curl URL-encode the content and pass that on. Just be careful so that the content doesn't contain any=or@symbols, as that will then make the syntax match one of the other cases below! 😭=content— this will make curl URL-encode the content and pass that on. The preceding=symbol is not included in the data.name=content— this will make curl URL-encode the content part and pass that on. Note that the name part is expected to be URL-encoded already.@filename— this will make curl load data from the given file (including any newlines), URL-encode that data and pass it on in the POST.name@filename— this will make curl load data from the given file (including any newlines), URL-encode that data and pass it on in the POST. The name part gets an equal sign appended, resulting inname=urlencoded-file-content. Note that the name is expected to be URL-encoded already.
See also -d, --data and --data-raw. Added in 7.18.0 💜
-d data,--data data— bestie, this is the POST option 💌 it sends the data you give it to the HTTP server, exactly the way your browser does when you fill in an HTML form and smash that submit button. curl passes your data to the server using the content-typeapplication/x-www-form-urlencoded. Compare it to-F,--formif you're feeling fancy.
Quick lil family portrait so you don't mix them up:
--data-rawis almost the same but does not give the@character any special interpretation.- Posting data that's purely binary? Use
--data-binaryinstead, no cap. - Need to URL-encode the value of a form field?
--data-urlencodehas you covered.
Use one of these more than once on the same command line and the data pieces get merged together with a separating &-symbol. So -d name=daniel -d skill=lousy would generate a post chunk that looks like name=daniel&skill=lousy ✨ iconic behavior.
Start the data with the letter @ and the rest should be a file name to read the data from — or - if you want curl to read the data from stdin. Multiple files can also be specified. So posting data from a file named foobar would be done with -d, --data @foobar. Heads up though 🫶 when --data is told to read from a file like that, carriage returns and newlines will be stripped out. And if you don't want @ to have a special interpretation, use --data-raw instead.
See also --data-binary and --data-urlencode and --data-raw. This option overrides -F, --form and -I, --head and -T, --upload-file.
--delegation LEVEL— (GSS/kerberos) setsLEVELto tell the server what it's allowed to delegate when it comes to your user credentials 🔐none— don't allow any delegation. absolutely not, we don't know her.policy— delegates if and only if theOK-AS-DELEGATEflag is set in the Kerberos service ticket, which is a matter of realm policy. very "read the room" of it.always— unconditionally allow the server to delegate. no vibes check, just yes 💀
--digest— (HTTP) enables HTTP Digest authentication, an authentication scheme that keeps your password from being sent over the wire in clear text 🎀 use it together with the normal-u,--useroption to set user name and password.
If this option is used several times, only the first one is used.
See also -u, --user and --proxy-digest and --anyauth. This option overrides --basic and --ntlm and --negotiate.
--disable-eprt— (FTP) tells curl to disable the use of theEPRTandLPRTcommands when doing active FTP transfers. Normally curl will always first attemptEPRT, thenLPRT, before usingPORT— but with this option it goes straight toPORT🏃♀️💨EPRTandLPRTare extensions to the original FTP protocol and may not work on all servers, but they enable more functionality in a better way than the traditionalPORTcommand.
--eprt can be used to explicitly enable EPRT again, and --no-eprt is an alias for --disable-eprt.
If the server is accessed using IPv6, this option will have no effect, because EPRT is necessary then.
Disabling EPRT only changes the active behavior. If you want to switch to passive mode you need to not use -P, --ftp-port or force it with --ftp-pasv.
--disable-epsv— (FTP) tells curl to disable the use of theEPSVcommand when doing passive FTP transfers. Curl will normally always first attemptEPSVbeforePASV, but with this option it will not try usingEPSVat all 🙅♀️
--epsv can be used to explicitly enable EPSV again, and --no-epsv is an alias for --disable-epsv.
If the server is an IPv6 host, this option will have no effect, because EPSV is necessary then.
Disabling EPSV only changes the passive behavior. If you want to switch to active mode you need to use -P, --ftp-port.
-q,--disable— if used as the first parameter on the command line, thecurlrcconfig file will not be read and used. See-K,--configfor details on the default config file search path 📂 (yes, position matters, it's giving main character syndrome)--disallow-username-in-url— (HTTP) this tells curl to exit if it's passed a url containing a username. that's it. that's the whole boundary 🚩
See also --proto. Added in 7.61.0.
--dns-interface <interface>— (DNS) tells curl to send outgoing DNS requests through<interface>. This option is a counterpart to--interface(which does not affect DNS, important distinction bestie ✨). The supplied string must be an interface name (not an address).
See also --dns-ipv4-addr and --dns-ipv6-addr. --dns-interface requires that the underlying libgnurl was built to support c-ares. Added in 7.33.0.
--dns-ipv4-addr <address>— (DNS) tells curl to bind to<ip-address>when making IPv4 DNS requests, so that the DNS requests originate from this address 🌈 the argument should be a single IPv4 address.
See also --dns-interface and --dns-ipv6-addr. --dns-ipv4-addr requires that the underlying libgnurl was built to support c-ares. Added in 7.33.0.
--dns-ipv6-addr <address>— (DNS) tells curl to bind to<ip-address>when making IPv6 DNS requests, so that the DNS requests originate from this address 💜 the argument should be a single IPv6 address.
See also --dns-interface and --dns-ipv4-addr. --dns-ipv6-addr requires that the underlying libgnurl was built to support c-ares. Added in 7.33.0.
--dns-servers <addresses>— sets the list of DNS servers to be used instead of the system default 🖇️ the list of IP addresses should be separated with commas. Port numbers may also optionally be given as:<port-number>after each IP address.
--dns-servers requires that the underlying libgnurl was built to support c-ares. Added in 7.33.0.
--doh-url -<URL>— (all) specifies which DNS-over-HTTPS (DOH) server to use to resolve hostnames, instead of using the default name resolver mechanism 🌐 the URL must beHTTPS.
If this option is used several times, the last one will be used.
-D <filename>,--dump-header <filename>— (HTTP FTP) writes the received protocol headers to the specified file 📝
This option is handy when you want to store the headers that an HTTP site sends to you. Cookies from the headers could then be read in a second curl invocation by using the -b | --cookie option! (Though the -c | --cookie-jar option is a better way to store cookies, jsyk 🍪)
If no headers are received, the use of this option will create an empty file.
When used in FTP, the FTP server response lines are considered being "headers" and thus are saved there.
If this option is used several times, the last one will be used.
See also -o | --output.
--egd-file <file>— (TLS) specify the path name to the Entropy Gathering Daemon socket ✨ the socket is used to seed the random engine for SSL connections.
See also --random-file.
--engine <name>— (TLS) select the OpenSSL crypto engine to use for cipher operations. Use--engine listto print a list of build-time supported engines 🔧 note that not all (or none) of the engines may be available at run-time. so like, don't be shocked if the list is empty, it's not personal 💀--expect100-timeout <seconds>— (HTTP) the maximum time in seconds that you allow curl to wait for a100-continueresponse when curl emits anExpects: 100-continueheader in its request ⏱️ by default curl will wait one second. This option accepts decimal values! When curl stops waiting, it will continue as if the response has been received.
See also --connect-timeout. Added in 7.47.0.
--fail-early— fail and exit on the first detected transfer error 🛑
When curl is used to do multiple transfers on the command line, it attempts to operate on each given URL, one by one. By default it will ignore errors if there are more URLs given, and the last URL's success will determine the error code curl returns. So early failures get "hidden" by subsequent successful transfers — lowkey sneaky of it fr.
Using this option, curl will instead return an error on the first transfer that fails, independent of the amount of URLs given on the command line. This way, no transfer failures go undetected by scripts and similar. periodt.
This option is global and does not need to be specified for each use of -: | --next.
This option does not imply -f | --fail, which causes transfers to fail due to the server's HTTP status code. You can combine the two options, however note that -f | --fail is not global and is therefore contained by -: | --next.
Added in 7.52.0.
-f,--fail— (HTTP) fail silently (no output at all) on server errors 🤐 this is mostly done to better enable scripts etc to better deal with failed attempts. In normal cases when an HTTP server fails to deliver a document, it returns an HTML document stating so (which often also describes why and more). This flag will prevent curl from outputting that, and return error22.
⚠️ This method is not fail-safe, and there are occasions where non-successful response codes will slip through — especially when authentication is involved (response codes401and407). so like, keep an eye on it 👀
--false-start— (TLS) tells curl to use false start during the TLS handshake 🚀 false start is a mode where a TLS client starts sending application data before verifying the server'sFinishedmessage, thus saving a round trip when performing a full handshake.
This is currently only implemented in the NSS and Secure Transport (on iOS 7.0 or later, or OS X 10.9 or later) backends.
Added in 7.42.0.
--form-string <name=string>— (HTTP SMTP IMAP) similar to-F,--formexcept that the value string for the named parameter is used literally 💅 leading@and<characters, and the;type=string in the value have no special meaning. Use this in preference to-F|--formif there's any possibility that the string value may accidentally trigger the@or<features of-F|--form— like, protect yourself 🫶
See also -F | --form.
-F <name=content>,-form <name=content>— okay bestie, settle in, this one is long but it's also kind of the main character 💅 (HTTP SMTP IMAP). For the HTTP protocol family, this lets curl emulate a filled-in form in which a user has pressed the submit button. That makes curl POST data using the Content-Typemultipart/form-dataaccording to RFC 2388.
For SMTP and IMAP protocols, this is the mean to compose a multipart mail message to transmit. ✨
This enables uploading of binary files etc. To force the content part to be a file, prefix the file name with an @ sign. To just get the content part from a file, prefix the file name with the symbol <. The difference between @ and < is then that @ makes a file get attached in the post as a file upload, while the < makes a text field and just get the contents for that text field from a file.
Tell curl to read content from stdin instead of a file by using - as filename. This goes for both @ and < constructs. When stdin is used, the contents is buffered in memory first by curl to determine its size and allow a possible resend. Defining a part's data from a named non-regular file (such as a named pipe or similar) is unfortunately not subject to buffering and will be effectively read at transmission time; since the full size is unknown before the transfer starts, such data is sent as chunks by HTTP and rejected by IMAP. 😭 (yeah, that one's a real gotcha, no cap)
Example: send an image to an HTTP server, where profile is the name of the form-field to which the file portrait.jpg will be the input:
curl -F profile=@portrait.jpg
https://example.com/upload.cgi
Example: send a your name and shoe size in two text fields to the server:
curl -F name=John -F shoesize=11 https://example.com/
Example: send a your essay in a text field to the server. Send it as a plain text field, but get the contents for it from a local file:
curl -F "story=<hugefile.txt" https://example.com/
You can also tell curl what Content-Type to use by using type=, in a manner similar to:
curl -F "web=@index.html;type=text/html" example.com
or
curl -F "name=daniel;type=text/foo" example.com
You can also explicitly change the name field of a file upload part by setting filename=, like this:
curl -F "file=@localfile;filename=nameinpost" example.com
If filename/path contains , or ;, it must be quoted by double-quotes like:
curl -F "file=@\"localfile\";filename=\"nameinpost\""
example.com
or
curl -F 'file=@"localfile";filename="nameinpost"'
example.com
Note that if a filename/path is quoted by double-quotes, any double-quote or backslash within the filename must be escaped by backslash. 🫶
Quoting must also be applied to non-file data if it contains semicolons, leading/trailing spaces or leading double quotes:
curl -F 'colors="red; green; blue";type=text/x-myapp'
example.com
You can add custom headers to the field by setting headers=, like
curl -F "submit=OK;headers=\"X-submit-type: OK\""
example.com
or
curl -F "submit=OK;headers=@headerfile" example.com
The headers= keyword may appear more that once and above notes about quoting apply. When headers are read from a file, Empty lines and lines starting with # are comments and ignored; each header can be folded by splitting between two words and starting the continuation line with a space; embedded carriage-returns and trailing spaces are stripped. Here is an example of a header file contents:
# This file contain two headers.
X-header-1: this is a header
# The following header is folded.
X-header-2: this is
another header
To support sending multipart mail messages, the syntax is extended as follows:
- name can be omitted: the equal sign is the first character of the argument,
- if data starts with
(, this signals to start a new multipart: it can be followed by a content type specification. - a multipart can be terminated with a
=)argument.
Example: the following command sends an SMTP mime e-mail consisting in an inline part in two alternative formats: plain text and HTML. It attaches a text file:
curl -F '=(;type=multipart/alternative' \
-F '=plain text message' \
-F '= <body>HTML message</body>;type=text/html' \
-F '=)' -F '=@textfile.txt' ... smtp://example.com
Data can be encoded for transfer using encoder=. Available encodings are binary and 8bit that do nothing else than adding the corresponding Content-Transfer-Encoding header, 7bit that only rejects 8-bit characters with a transfer error, quoted-printable and base64 that encodes data according to the corresponding schemes, limiting lines length to 76 characters.
Example: send multipart mail with a quoted-printable text message and a base64 attached file:
curl -F '=text message;encoder=quoted-printable' \
-F '=@localfile;encoder=base64' ... smtp://example.com
See further examples and details in the MANUAL. 📖
This option can be used multiple times.
This option overrides -d | --data and -I | --head and -T | --upload-file.
--ftp-account <data>— (FTP) when an FTP server asks for "account data" after user name and password has been provided, this data is sent off using theACCTcommand. 💌
If this option is used several times, the last one will be used.
Added in 7.13.0.
--ftp-alternative-to-user <command>— (FTP) if authenticating with theUSERandPASScommands fails, send this command. When connecting to Tumbleweed's Secure Transport server over FTPS using a client certificate, using"SITE AUTH"will tell the server to retrieve the username from the certificate. 🎀
Added in 7.15.5.
--ftp-create-dirs— (FTP SFTP) when an FTP or SFTP URL/operation uses a path that doesn't currently exist on the server, the standard behavior of curl is to fail. Using this option, curl will instead attempt to create missing directories. literally building the folder for you, iconic ✨
See also --create-dirs.
--ftp-method <method>— (FTP) control what method curl should use to reach a file on an FTP(S) server. The method argument should be one of the following alternatives:multicwd— curl does a singleCWDoperation for each path part in the given URL. For deep hierarchies this means very many commands. This is how RFC 1738 says it should be done. This is the default but the slowest behavior. 🐌nocwd— curl does noCWDat all. curl will doSIZE,RETR,STORetc and give a full path to the server for all these commands. This is the fastest behavior. 🔥singlecwd— curl does oneCWDwith the full target directory and then operates on the file "normally" (like in the multicwd case). This is somewhat more standards compliant thannocwdbut without the full penalty ofmulticwd.
Added in 7.15.1.
--ftp-pasv— (FTP) use passive mode for the data connection. Passive is the internal default behavior, but using this option can be used to override a previous-P|--ftp-portoption.
If this option is used several times, only the first one is used. Undoing an enforced passive really isn't doable but you must then instead enforce the correct -P | --ftp-port again. 😬 no take-backsies, bestie
Passive mode means that curl will try the EPSV command first and then PASV, unless --disable-epsv is used.
See also --disable-epsv. Added in 7.11.0.
-P <address>,-ftp-port <address>— (FTP) reverses the default initiator/listener roles when connecting with FTP. This option makes curl use active mode. curl then tells the server to connect back to the client's specified address and port, while passive mode asks the server to setup an IP address and port for it to connect to. <address> should be one of:interface— e.g."eth0"to specify which interface's IP address you want to use (Unix only)IP address— e.g."192.168.10.1"to specify the exact IP addresshost name— e.g."my.host.domain"to specify the machine-— make curl pick the same IP address that is already used for the control connection
If this option is used several times, the last one will be used. Disable the use of PORT with --ftp-pasv. Disable the attempt to use the EPRT command instead of PORT by using --disable-eprt. EPRT is really PORT++. 💀
Since 7.19.5, you can append :[start]-[end] to the right of the address, to tell curl what TCP port range to use. That means you specify a port range, from a lower to a higher number. A single number works as well, but do note that it increases the risk of failure since the port may not be available. ⚠️
See also --ftp-pasv and --disable-eprt.
--ftp-pret— (FTP) tell curl to send aPRETcommand beforePASV(andEPSV). Certain FTP servers, mainly drftpd, require this non-standard command for directory listings as well as up and downloads in PASV mode. respectfully, drftpd, why 🧚
Added in 7.20.0.
--ftp-skip-pasv-ip— (FTP) tell curl to not use the IP address the server suggests in its response to curl'sPASVcommand when curl connects the data connection. Instead curl will re-use the same IP address it already uses for the control connection. 💅
This option has no effect if PORT, EPRT or EPSV is used instead of PASV.
See also --ftp-pasv. Added in 7.14.2.
--ftp-ssl-ccc-mode <active/passive>— (FTP) okay so this one sets the CCC mode ✨. In passive mode it will not initiate the shutdown, it just sits there and waits for the server to do it, and it won't reply to the shutdown from the server either. In active mode it initiates the shutdown itself and then waits for a reply from the server. 💅
See also --ftp-ssl-ccc. Added in 7.16.2.
--ftp-ssl-ccc— (FTP) use CCC, aka Clear Command Channel 🎀. This shuts down the SSL/TLS layer after authenticating, so the rest of the control channel communication is unencrypted. Why would you want that? Because it lets NAT routers follow the FTP transaction, bestie. The default mode is passive.
See also --ssl and --ftp-ssl-ccc-mode. Added in 7.16.1.
--ftp-ssl-control— (FTP) require SSL/TLS for the FTP login, and clear (unencrypted) for transfer 🦋. So you get secure authentication but non-encrypted data transfers for efficiency. And if the server doesn't support SSL/TLS? The transfer fails. No cap. 😭
Added in 7.16.0.
-G|--get— when you use this, all the data you specified with-d,--data,--data-binaryor--data-urlencodegets used in an HTTP GET request instead of the POST request that would otherwise happen 🫶. The data gets appended to the URL with a'?'separator. Iconic.
If you use it together with -I | --head, the POST data will instead get appended to the URL with a HEAD request.
If this option is used several times, only the first one is used. Why? Because undoing a GET doesn't make sense, duh — instead you should just enforce the alternative method you prefer.
-g|--globoff— this switches off the "URL globbing parser" ⭐. So when you set this option, you can specify URLs that contain the letters{}[]without them getting interpreted by curl itself. Lowkey important note tho: these letters are not normal legal URL contents, but they should be encoded according to the URI standard. 💅--happy-eyeballs-timeout-msmilliseconds — okay so happy eyeballs 🎀 (yes that's the real name, iykyk) is an algorithm that tries to connect to both IPv4 and IPv6 addresses for dual-stack hosts, preferring IPv6 first for the number of milliseconds you give it. If the IPv6 address can't be connected to within that time, then a connection attempt gets made to the IPv4 address in parallel. Whichever connection establishes first is the one that's used. 🦋
The range of suggested useful values is limited, ngl. Happy Eyeballs RFC 6555 says "It is RECOMMENDED that connection attempts be paced 150-250 ms apart to balance human factors against network load." libcurl currently defaults to 200 ms. Firefox and Chrome currently default to 300 ms.
If this option is used several times, the last one will be used.
Added in 7.59.0.
--haproxy-protocol— (HTTP) send a HAProxy PROXY protocol v1 header at the beginning of the connection 🔥. Some load balancers and reverse proxies use this to indicate the client's true IP address and port.
This option is primarily useful when you're sending test requests to a service that expects this header. ✨
Added in 7.60.0.
-I|--head— (HTTP FTP FILE) fetch the headers only, bestie! 👑 HTTP servers have the commandHEADwhich this uses to get nothing but the header of a document. When used on an FTP or FILE file, curl displays the file size and last modification time only. Periodt.-H <header/@file>|-header <header/@file>— (HTTP) extra header to include in the request when sending HTTP to a server 💌. You may specify any number of extra headers. Note that if you add a custom header that has the same name as one of the internal ones curl would use, your externally set header will be used instead of the internal one. This lets you do even trickier stuff than curl would normally do 😈. But fr fr, you should not replace internally set headers without knowing perfectly well what you're doing.
Remove an internal header by giving a replacement with no content on the right side of the colon, like: -H "Host:". If you send the custom header with no-value then its header must be terminated with a semicolon, like -H "X-Custom-Header;" to send X-Custom-Header:.
curl will make sure each header you add/replace is sent with the proper end-of-line marker, so you should thus not add that as part of the header content: do not add newlines or carriage returns, they will only mess things up for you 🙈.
Starting in 7.55.0, this option can take an argument in @filename style, which then adds a header for each line in the input file. Using @- will make curl read the header file from stdin. 🍓
See also the -A, --user-agent and -e, --referer options.
Starting in 7.37.0, you need --proxy-header to send custom headers intended for a proxy.
Example:
curl -H X-First-Name: Joe http://example.com/
WARNING: headers set with this option will be set in all requests — even after redirects are followed, like when told with-L|--location. This can lead to the header being sent to other hosts than the original host, so sensitive headers should be used with caution when combined with following redirects. 💀
This option can be used multiple times to add/replace/remove multiple headers. ✨
-h|--help— usage help 🫶. This lists all current command line options with a short description. Your little cheat sheet, bestie.--hostpubmd5 <md5>— (SFTP SCP) pass a string containing 32 hexadecimal digits 🔐. The string should be the 128 bit MD5 checksum of the remote host's public key, and curl will refuse the connection with the host unless the md5sums match. So it's giving "trust but verify" energy.
Added in 7.17.1.
--http0.9— (HTTP) tells curl to be fine with HTTP version 0.9 responses 🌸.
HTTP/0.9 is a completely headerless response, and therefore you can also connect with this to non-HTTP servers and still get a response, since curl will simply transparently downgrade — if allowed. 👑
A future curl version will deny continuing if the response isn't at least HTTP/1.0 unless this option is used. So like, heads up. ⭐
-0|--http1.0— (HTTP) tells curl to use HTTP version 1.0 instead of using its internally preferred HTTP version. 💅
This option overrides --http1.1 and --http2.
--http1.1— (HTTP) tells curl to use HTTP version 1.1 ✨.
This option overrides -0 | ---http1.0 and --http2. Added in 7.33.0.
--http2-prior-knowledge— (HTTP) tells curl to issue its non-TLS HTTP requests using HTTP/2 without the HTTP/1.1 Upgrade 🔥. It requires prior knowledge that the server supports HTTP/2 straight away. HTTPS requests will still do HTTP/2 the standard way with the negotiated protocol version in the TLS handshake.
--http2-prior-knowledge requires that the underlying libgnurl was built to support HTTP/2. This option overrides --http1.1 and -0 | --http1.0 and --http2. Added in 7.49.0.
--http2— (HTTP) tells curl to use HTTP version 2 🦋.
See also --no-alpn. --http2 requires that the underlying libgnurl was built to support HTTP/2. This option overrides --http1.1 and -0, --http1.0 and --http2-prior-knowledge. Added in 7.33.0.
--ignore-content-length— (FTP HTTP) for HTTP, ignore theContent-Lengthheader 🙈. This is particularly useful for servers running Apache 1.x, which will report incorrectContent-Lengthfor files larger than 2 gigabytes.
For FTP (since 7.46.0), skip the RETR command to figure out the size before downloading a file. 💅
-i|--include— include the HTTP response headers in the output 🎀. The HTTP response headers can include things like server name, cookies, date of the document, HTTP version and more...
Wanna view the request headers instead? Consider the -v, --verbose option.
See also -v, --verbose.
-k|--insecure— (TLS) by default, every SSL connection curl makes is verified to be secure 🔐. This option allows curl to proceed and operate even for server connections otherwise considered insecure. So it's giving "bestie I'm ignoring the red flags" energy, use responsibly. 😭
The server connection is verified by making sure the server's certificate contains the right name and verifies successfully using the cert store.
See this online resource for further details: https://curl.haxx.se/docs/sslcerts.html
See also --proxy-insecure and --cacert.
--interface <name>— perform an operation using a specified interface 🌈. You can enter an interface name, IP address or host name. An example could look like:
curl --interface eth0:1 https://www.example.com/
If this option is used several times, the last one will be used.
On Linux it can be used to specify a VRF, but the binary needs to either have CAP_NET_RAW or to be run as root 👑. More information about Linux VRF: https://www.kernel.org/doc/Documentation/networking/vrf.txt
See also --dns-interface.
-4|--ipv4— this option tells curl to resolve names to IPv4 addresses only, and not for example try IPv6 💜.
See also --http1.1 and --http2. This option overrides -6, --ipv6.
-6|--ipv6— this option tells curl to resolve names to IPv6 addresses only, and not for example try IPv4 💜.
See also --http1.1 and --http2. This option overrides -4, --ipv4.
-j|--junk-session-cookies— (HTTP) when curl is told to read cookies from a given file, this option makes it discard all "session cookies" 🍓. This basically has the same effect as if a new session is started. Typical browsers always discard session cookies when they're closed down, so like, it's the normie behaviour.
See also -b, --cookie and -c, --cookie-jar.
--keepalive-time <seconds>— this option sets the time a connection needs to remain idle before sending keepalive probes, and the time between individual keepalive probes ⏱️. It's currently effective on operating systems offering theTCP_KEEPIDLEandTCP_KEEPINTVLsocket options (meaning Linux, recent AIX, HP-UX and more). This option has no effect if--no-keepaliveis used. 💅- 🔁 and if you sprinkle this one around like confetti? only the last one counts. leave it off completely and bestie it just defaults to
60seconds, no big deal.
Added in 7.18.0.
--key-type <type>— (TLS) so this is literally just which flavor your private key file is. you're telling curl what type the key you handed over with--keyactually is. the supported ones areDER,PEM, andENG. don't say anything and it assumesPEM, okurrr.
If you use this several times, the last one is the one that matters.
--key <key>— (TLS SSH) the private key file name, babes. lets you drop your private key in its own separate file. for SSH, if you don't specify one, curl just goes down this list in order hoping something hits:~/.ssh/id_rsa,~/.ssh/id_dsa,./id_rsa,./id_dsa.
If curl is built against the OpenSSL library and the pkcs11 engine is available, then a PKCS#11 URI (RFC 7512) can be used to point at a private key sitting in a PKCS#11 device 🎀. any string starting with "pkcs11:" gets read as a PKCS#11 URI. and if you do hand over a PKCS#11 URI, then --engine gets set to "pkcs11" if you didn't already set it, and --key-type gets set to "ENG" if you didn't already set that either. curl's got you.
Use it several times? last one wins, always.
--krb <level>— (FTP) turns on Kerberos auth and uses it ✨. you must give a level, and it should be one of'clear','safe','confidential', or'private'. and if you get cute and pick a level that isn't any of those? curl just goes'private'instead. no notes.
Several times = last one only.
heads up: --krb needs the underlying libgnurl to have been built with Kerberos support. no build support, no Kerberos, periodt.
--libcurl <file>— okay this one is genuinely iconic 👑. tack this onto any normal curl command line and curl writes out a libcurl-using C source file to the file you named — code that does the exact same thing your command-line operation does. your command, but as C. it's giving "teach a girl to fish", fr.
Give it several times and the last file name given is the one used.
Added in 7.16.1.
--limit-rate <speed>— sets the max transfer rate curl is allowed to use, and yes that covers both downloads and uploads 💅. super useful when your pipe is limited and you don't want one transfer just eating your entire bandwidth. it's literally just to make it slower than it otherwise would be.
The speed is measured in bytes/second, unless you add a suffix. add 'k' or 'K' and it counts as kilobytes, 'm' or 'M' makes it megabytes, and 'g' or 'G' makes it gigabytes. examples: 200K, 3m and 1G.
If you also use the -Y, --speed-limit option, that one takes precedence and might cripple the rate-limiting a little bit — but that's on purpose, so the speed-limit logic keeps working.
Several times? last one. always last one.
-l,--list-only— (FTP POP3) two moods, one flag:- (FTP) when you're listing an FTP directory, this forces a name-only view 🌸. extra useful if you want to machine-parse the contents of that directory, because the normal directory view doesn't follow any standard look or format. used this way, it makes curl send an
NLSTcommand to the server instead ofLIST. - (POP3) when you're grabbing a specific email from POP3, this forces a
LISTcommand instead ofRETR🔥. great for checking whether a specific message id even exists on the server, and how big it is.
- (FTP) when you're listing an FTP directory, this forces a name-only view 🌸. extra useful if you want to machine-parse the contents of that directory, because the normal directory view doesn't follow any standard look or format. used this way, it makes curl send an
note: some FTP servers list only files when they respond to NLST — no sub-directories, no symbolic links. just files. sad but true.
note: pair this with-X,--requestand you can send aUIDLcommand instead, so you can use the email's unique identifier rather than its message id for the request. iykyk.
Added in 7.21.5.
--local-port <num/range>— pick your preferred local port number, or a range (FROM-TO), to use for the connection(s) 🦋. just one number or a range, your call.
real talk: port numbers are a scarce resource by nature and they get busy sometimes, so if you set this range too narrow you can cause unnecessary connection setup failures. don't do that to yourself.
Added in 7.15.2.
--location-trusted— (HTTP) it's like-L,--location, but it also lets curl send your name + password to every host the site might redirect to 🙈. this may or may not introduce a security breach if the site redirects you somewhere you'll be sending your auth info to — and that info is literally plaintext in the case of HTTP Basic authentication. so like, be aware.
See also -u, --user.
-L,--location— (HTTP) if the server says the page you wanted has moved somewhere else (you'll see aLocation:header and a3XXresponse code), this makes curl redo the request at the new place 🎀. if you use it with-i,--includeor-I,--head, you'll see headers from all the requested pages.
When authentication is involved, curl only sends its credentials to the initial host. if a redirect takes curl to a different host, it won't be able to intercept the user+password. see also --location-trusted if you want to change that. and you can cap how many redirects it'll follow with the --max-redirs option.
When curl follows a redirect and the request isn't a plain GET (so, like, POST or PUT), it does the next request as a GET if the HTTP response was 301, 302, or 303. if the response code was any other 3xx code, curl re-sends the following request using the same unmodified method. no switching up on you.
You can tell curl not to change a non-GET request method into GET after a 30x response by using the dedicated options for exactly that: --post301, --post302 and --post303 ✨.
--login-options <options>— (IMAP POP3 SMTP) specify the login options to use during server authentication.
You can use login options to specify protocol-specific options that might get used during authentication. right now only IMAP, POP3 and SMTP support login options. want more info on login options? go read RFC 2384, RFC 5092 and IETF draft draft-earhart-url-smtp-00.txt. yes, really.
Several times = last one used.
Added in 7.34.0.
--mail-auth <address>— (SMTP) specify a single address 💌. this is the authentication address (identity) of a submitted message that's being relayed to another server.
See also --mail-rcpt and --mail-from. Added in 7.25.0.
--mail-from <address>— (SMTP) specify a single address that the given mail should be sent from 📮.
See also --mail-rcpt and --mail-auth. Added in 7.20.0.
--mail-rcpt <address>— (SMTP) specify a single address, user name, or mailing list name. repeat this option several times if you've got multiple recipients to send to 💖.- when doing a mail transfer, the recipient should be a valid email address to send the mail to.
- when doing an address verification (the
VRFYcommand), the recipient should be specified as the user name, or user name and domain (as per Section 3.5 of RFC5321). (Added in7.34.0) - when doing a mailing list expand (the
EXPNcommand), the recipient should be specified using the mailing list name, like"Friends"or"London-Office". (Added in7.34.0)
Added in 7.20.0.
-M,--manual— Manual. displays the huge help text 📖. that's it. that's the whole thing. iconic.--max-filesize <bytes>— sets the max size (in bytes) of a file you're willing to download. if the file you asked for is bigger than this value, the transfer won't even start and curl returns with exit code63💀.
A size modifier is allowed. for example, append 'k' or 'K' and it counts as kilobytes, 'm' or 'M' makes it megabytes, and 'g' or 'G' makes it gigabytes. examples: 200K, 3m and 1G. (Added in 7.58.0)
NOTE: the file size isn't always known before the download starts, and for those files this option has no effect — even if the transfer ends up bigger than the limit you gave. this concerns both FTP and HTTP transfers. sorry bestie.
See also --limit-rate.
--max-redirs <num>— (HTTP) sets the maximum number of redirect-followings allowed. when-L,--locationis in play, this stops curl from following redirects too much. by default the limit is50redirections. set this to-1to make it unlimited 🌀.
Several times? last one. you know the drill.
-m seconds,--max-time -seconds— maximum time in seconds that you'll allow the whole operation to take ⏳. this is the one that stops your batch jobs from hanging for hours because of slow networks or links going down. since7.32.0this option accepts decimal values — but heads up, the actual timeout gets less accurate the more decimal precision you specify. so like, precision is a vibe, not a guarantee.
Several times = last one used.
See also --connect-timeout.
--metalink— tells curl to parse and process a given URI as a Metalink file (both version 3 and 4 (RFC 5854) are supported 💅) and use the mirrors listed inside it for failover if there are errors (like the file or the server not being available). it'll also verify the hash of the file once the download completes. the Metalink file itself gets downloaded and processed in memory — it is not stored in the local file system.
Example of using a remote Metalink file:
curl --metalink http://www.example.com/example.metalink
To use a Metalink file that's already in your local file system, use the FILE protocol (file://):
curl --metalink file://example.metalink
please note: if the FILE protocol is disabled, there is genuinely no way to use a local Metalink file at the time of this writing. also note that if--metalinkand-i,--includeare used together,--includewill be ignored — because including headers in the response breaks the Metalink parser, and if those headers end up inside the file described by the Metalink file, the hash check will fail. so. yeah.
--metalink requires the underlying libgnurl to be built with metalink support. Added in 7.27.0.
--negotiate— (HTTP) turns on Negotiate (SPNEGO) authentication ✨
Heads up: this one needs a library that was built with GSS-API or SSPI support. Wanna check? Run -V, --version and see if your curl supports GSS-API/SSPI or SPNEGO 💅
Also — plot twist — you must also hand it a fake -u, --user option so the auth code actually activates properly. Sending -u : is literally enough, because the username and password from the -u, --user option aren't actually used. Weird but true 🙈
If this option is used several times, only the first one is used.
See also --basic and --ntlm and --anyauth and --proxy-negotiate.
--netrc-filefilename — lowkey the same as-n,--netrc, except you give it the path (absolute or relative) to the netrc file curl should use 💖
You can only specify one netrc file per invocation. If several --netrc-file options are provided, the last one will be used.
It will abide by --netrc-optional if specified.
This option overrides -n, --netrc. Added in 7.21.5.
--netrc-optional— very similar to-n,--netrc, but this option makes the.netrcusage optional and not mandatory like the-n,--netrcoption does 🎀
See also --netrc-file. This option overrides -n, --netrc.
-n|--netrc— makes curl scan the.netrc(_netrcon Windows) file in the user's home directory for login name and password 🦋
This is typically used for FTP on Unix. If used with HTTP, curl will enable user authentication. See netrc(5) ftp(1) for details on the file format. Curl will not complain if that file doesn't have the right permissions (it should not be either world- or group-readable). The environment variable "HOME" is used to find the home directory.
A quick and very simple example of how to setup a .netrc to allow curl to FTP to the machine host.domain.com with user name 'myself' and password 'secret' should look similar to:
machine host.domain.com login myself password secret
-:|--next— tells curl to use a separate operation for the following URL and associated options 🌸
This lets you send several URL requests, each with their own specific options — for example, different user names or custom requests for each one. Main character energy for every URL, honestly 👑
-:, --next will reset all local options and only global ones will have their values survive over to the operation following the -:, --next instruction. Global options include -v, --verbose, --trace, --trace-ascii and --fail-early.
For example, you can do both a GET and a POST in a single command line:
curl www1.example.com --next -d postthis www2.example.com
Added in 7.36.0.
--no-alpn— (HTTPS) disable the ALPN TLS extension ✨
ALPN is enabled by default if libcurl was built with an SSL library that supports ALPN. ALPN is used by a libcurl that supports HTTP/2 to negotiate HTTP/2 support with the server during https sessions.
See also --no-npn and --http2. --no-alpn requires that the underlying libgnurl was built to support TLS. Added in 7.36.0.
-N|--no-buffer— disables the buffering of the output stream 💅
In normal work situations, curl will use a standard buffered output stream that will have the effect that it will output the data in chunks, not necessarily exactly when the data arrives. Using this option will disable that buffering. No more waiting, you get the tea as it drops 🍓
Note that this is the negated option name documented. You can thus use --buffer to enforce the buffering.
--no-keepalive— disables the use of keepalive messages on the TCP connection 🫶
curl otherwise enables them by default.
Note that this is the negated option name documented. You can thus use --keepalive to enforce keepalive.
--no-npn— (HTTPS) disable the NPN TLS extension 🔥
NPN is enabled by default if libcurl was built with an SSL library that supports NPN. NPN is used by a libcurl that supports HTTP/2 to negotiate HTTP/2 support with the server during https sessions.
See also --no-alpn and --http2. --no-npn requires that the underlying libgnurl was built to support TLS. Added in 7.36.0.
--no-sessionid— (TLS) disable curl's use of SSL session-ID caching 😭
By default all transfers are done using the cache. Note that while nothing should ever get hurt by attempting to reuse SSL session-IDs, there seem to be broken SSL implementations in the wild that may require you to disable this in order for you to succeed. So like, not curl's fault, but here we are 💀
Note that this is the negated option name documented. You can thus use --sessionid to enforce session-ID caching.
Added in 7.16.0.
--noproxyno-proxy-list — comma-separated list of hosts which do not use a proxy, if one is specified 🌈
The only wildcard is a single * character, which matches all hosts, and effectively disables the proxy. Each name in this list is matched as either a domain which contains the hostname, or the hostname itself. For example, local.com would match local.com, local.com:80, and www.local.com, but not www.notlocal.com.
Since 7.53.0, this option overrides the environment variables that disable the proxy. If there's an environment variable disabling a proxy, you can set noproxy list to "" to override it.
Added in 7.19.4.
--ntlm-wb— (HTTP) enables NTLM much in the style--ntlmdoes, but hands over the authentication to the separate binaryntlmauthapplication that is executed when needed ⭐
See also --ntlm and --proxy-ntlm.
--ntlm— (HTTP) enables NTLM authentication 💜
The NTLM authentication method was designed by Microsoft and is used by IIS web servers. It's a proprietary protocol, reverse-engineered by clever people and implemented in curl based on their efforts. This kind of behavior should not be endorsed — you should encourage everyone who uses NTLM to switch to a public and documented authentication method instead, such as Digest. Like, respectfully, move on 😌
If you want to enable NTLM for your proxy authentication, then use --proxy-ntlm.
If this option is used several times, only the first one is used.
See also --proxy-ntlm. --ntlm requires that the underlying libgnurl was built to support TLS. This option overrides --basic and --negotiate and --digest and --anyauth.
--oauth2-bearertoken — (IMAP POP3 SMTP) specify the Bearer Token for OAUTH 2.0 server authentication 🧚
The Bearer Token is used in conjunction with the user name which can be specified as part of the --url or -u, --user options.
The Bearer Token and user name are formatted according to RFC 6750.
If this option is used several times, the last one will be used.
-oFILE |--outputFILE — write output to FILE instead of stdout 💖
If you're using {} or [] to fetch multiple documents, you can use # followed by a number in the file specifier. That variable will be replaced with the current string for the URL being fetched. Like in:
curl http://{one,two}.example.com -o file_#1.txt
or use several variables like:
curl http://{site,host}.host[1-5].com -o #1_#2
You may use this option as many times as the number of URLs you have. For example, if you specify two URLs on the same command line, you can use it like this:
curl -o aa example.com -o bb example.net
and the order of the -o options and the URLs doesn't matter — just that the first -o is for the first URL and so on, so the above command line can also be written as
curl example.com example.net -o aa -o bb
See also the --create-dirs option to create the local directories dynamically. Specifying the output as - (a single dash) will force the output to be done to stdout.
See also -O, --remote-name and --remote-name-all and -J, --remote-header-name.
--passphrase — (SSH TLS) passphrase for the private key 🔑
If this option is used several times, the last one will be used.
--path-as-is— tell curl to not handle sequences of/../or/./in the given URL path 🙈
Normally curl will squash or merge them according to standards, but with this option set you tell it not to do that.
Added in 7.42.0.
--pinnedpubkey<hashes> — (TLS) tells curl to use the specified public key file (or hashes) to verify the peer 👑
This can be a path to a file which contains a single public key in PEM or DER format, or any number of base64 encoded sha256 hashes preceded by ´sha256//´ and separated by ´;´
When negotiating a TLS or SSL connection, the server sends a certificate indicating its identity. A public key is extracted from this certificate and if it does not exactly match the public key provided to this option, curl will abort the connection before sending or receiving any data. Not a single byte, bestie 💀
PEM/DER support:
- 7.39.0: OpenSSL, GnuTLS and GSKit
- 7.43.0: NSS and wolfSSL/CyaSSL
- 7.47.0: mbedtls
sha256 support:
- 7.44.0: OpenSSL, GnuTLS, NSS and wolfSSL/CyaSSL.
- 7.47.0: mbedtls
Other SSL backends not supported.
If this option is used several times, the last one will be used.
--post301— (HTTP) tells curl to respect RFC 7231/6.4.2 and not convert POST requests into GET requests when following a 301 redirection 🎀
The non-RFC behaviour is ubiquitous in web browsers, so curl does the conversion by default to maintain consistency. However, a server may require a POST to remain a POST after such a redirection. This option is meaningful only when using -L, --location.
See also --post302 and --post303 and -L, --location. Added in 7.17.1.
--post302— (HTTP) bestie, this one tells curl to actually respect RFC 7231/6.4.3 and not turn your POST into a GET when it follows a 302 redirect. The non-RFC behaviour is literally everywhere in browsers, so curl does the conversion by default to keep things consistent. But sometimes a server is like "no, keep it a POST" after that redirect, and that's when you use this. It's only meaningful when you're using-L,--location.
See also --post301 and --post303 and -L, --location. Added in 7.19.1.
--post303— (HTTP) tells curl to straight up violate RFC 7231/6.4.4 and not convert POST requests into GET requests when following 303 redirections. A server may require a POST to stay a POST after a 303 redirect, and honestly? valid. This option is meaningful only when using-L,--location.
See also --post302 and --post301 and -L, --location. Added in 7.26.0.
--preproxy [protocol://]host[:port]— use the specified SOCKS proxy before connecting to an HTTP or HTTPS-x,--proxy. So curl first connects to the SOCKS proxy, and then connects (through SOCKS) to the HTTP or HTTPS proxy. Hence pre proxy, get it? 💅
The pre proxy string should be specified with a protocol:// prefix to specify alternative proxy protocols. Use socks4://, socks4a://, socks5:// or socks5h:// to request the specific SOCKS version to be used. No protocol specified will make curl default to SOCKS4.
If the port number is not specified in the proxy string, it is assumed to be 1080.
User and password that might be provided in the proxy string are URL decoded by curl. This allows you to pass in special characters such as @ by using %40 or pass in a colon with %3a — so sneaky, so useful 🫶
If this option is used several times, the last one will be used.
Added in 7.52.0.
-#,--progress-bar— makes curl display transfer progress as a simple progress bar instead of the standard, more informational, meter. it's giving minimalism ✨
This progress bar draws a single line of # characters across the screen and shows a percentage if the transfer size is known. For transfers without a known size, there will be a space ship (-=o=-) that moves back and forth but only while data is being transferred, with a set of flying hash sign symbols on top. a literal little spacecraft, iconic, no notes 🚀
--proto-default <protocol>— tells curl to use protocol for any URL missing a scheme name.
Example:
curl --proto-default https ftp.mozilla.org
An unknown or unsupported protocol causes error CURLE_UNSUPPORTED_PROTOCOL (1).
Heads up: this option does not change the default proxy protocol (http).
Without this option curl would make a guess based on the host, see --url for details.
Added in 7.45.0.
--proto-redir <protocols>— tells curl to limit what protocols it may use on redirect. Protocols denied by--protoare not overridden by this option. See--protofor how protocols are represented.
Example, allow only HTTP and HTTPS on redirect:
curl --proto-redir -all,http,https http://example.com
By default curl will allow all protocols on redirect except several disabled for security reasons: since 7.19.4 FILE and SCP are disabled, and since 7.40.0 SMB and SMBS are also disabled. Specifying all or +all enables all protocols on redirect, including those disabled for security. so like, tread carefully bestie 💀
Added in 7.20.2.
--proto <protocols>— tells curl to limit what protocols it may use in the transfer. Protocols are evaluated left to right, are comma separated, and are each a protocol name orall, optionally prefixed by zero or more modifiers. Available modifiers are:+— permit this protocol in addition to protocols already permitted (this is the default if no modifier is used).-— deny this protocol, removing it from the list of protocols already permitted.=— permit only this protocol (ignoring the list already permitted), though subject to later modification by subsequent entries in the comma separated list.
For example:
--proto -ftps
uses the default protocols, but disables ftps
--proto -all,https,+http
only enables http and https
--proto =http,https
also only enables http and https
Unknown protocols produce a warning. This allows scripts to safely rely on being able to disable potentially dangerous protocols, without relying upon support for that protocol being built into curl to avoid an error. so your scripts stay safe even if curl wasn't built with the thing 👑
This option can be used multiple times, in which case the effect is the same as concatenating the protocols into one instance of the option.
See also --proto-redir and --proto-default. Added in 7.20.2.
--proxy-anyauth— tells curl to pick a suitable authentication method when talking to the given HTTP proxy. This might cause an extra request/response round-trip, so like, one extra little trip, nbd.
See also -x, --proxy and --proxy-basic and --proxy-digest. Added in 7.13.2.
--proxy-basic— tells curl to use HTTP Basic authentication when communicating with the given proxy. Use--basicfor enabling HTTP Basic with a remote host. Basic is the default authentication method curl uses with proxies — it's literally the default, she's always been here 💅
See also -x, --proxy and --proxy-anyauth and --proxy-digest.
--proxy-cacert <file>— same as--cacertbut used in HTTPS proxy context.
See also --proxy-capath and --cacert and --capath and -x, --proxy. Added in 7.52.0.
--proxy-capath <dir>— same as--capathbut used in HTTPS proxy context.
See also --proxy-cacert and -x, --proxy and --capath. Added in 7.52.0.
--proxy-cert-type <type>— same as--cert-typebut used in HTTPS proxy context.
Added in 7.52.0.
--proxy-cert <cert[:passwd]>— same as-E,--certbut used in HTTPS proxy context.
Added in 7.52.0.
--proxy-ciphers <list>— same as--ciphersbut used in HTTPS proxy context.
Added in 7.52.0.
--proxy-crlfile <file>— same as--crlfilebut used in HTTPS proxy context.
Added in 7.52.0.
--proxy-digest— tells curl to use HTTP Digest authentication when communicating with the given proxy. Use--digestfor enabling HTTP Digest with a remote host.
See also -x, --proxy and --proxy-anyauth and --proxy-basic.
--proxy-header <header/@file>— (HTTP) extra header to include in the request when sending HTTP to a proxy. You may specify any number of extra headers. This is the equivalent option to-H,--headerbut it's for proxy communication only — like in CONNECT requests when you want a separate header sent to the proxy versus what's sent to the actual remote host. two different audiences, two different headers, we love to see it 🎀
curl will make sure that each header you add/replace is sent with the proper end-of-line marker, you should thus not add that as a part of the header content: do not add newlines or carriage returns, they will only mess things up for you. fr fr, don't do it 🙈
Headers specified with this option will not be included in requests that curl knows will not be sent to a proxy.
Starting in 7.55.0, this option can take an argument in @filename style, which then adds a header for each line in the input file. Using @- will make curl read the header file from stdin.
This option can be used multiple times to add/replace/remove multiple headers.
Added in 7.37.0.
--proxy-insecure— same as-k,--insecurebut used in HTTPS proxy context.
Added in 7.52.0.
--proxy-key-type <type>— same as--key-typebut used in HTTPS proxy context.
Added in 7.52.0.
--proxy-key <key>— same as--keybut used in HTTPS proxy context.--proxy-negotiate— tells curl to use HTTP Negotiate (SPNEGO) authentication when communicating with the given proxy. Use--negotiatefor enabling HTTP Negotiate (SPNEGO) with a remote host.
See also --proxy-anyauth and --proxy-basic. Added in 7.17.1.
--proxy-ntlm— tells curl to use HTTP NTLM authentication when communicating with the given proxy. Use--ntlmfor enabling NTLM with a remote host.
See also --proxy-negotiate and --proxy-anyauth.
--proxy-pass <phrase>— same as--passbut used in HTTPS proxy context.
Added in 7.52.0.
--proxy-pinnedpubkey <hashes>— (TLS) tells curl to use the specified public key file (or hashes) to verify the proxy. This can be a path to a file which contains a single public key in PEM or DER format, or any number of base64 encoded sha256 hashes preceded bysha256//and separated by;
When negotiating a TLS or SSL connection, the server sends a certificate indicating its identity. A public key is extracted from this certificate and if it does not exactly match the public key provided to this option, curl will abort the connection before sending or receiving any data. like it will not even say hi, that's the level of strict 💀
If this option is used several times, the last one will be used.
--proxy-service-name <name>— this option allows you to change the service name for proxy negotiation.
Added in 7.43.0.
--proxy-ssl-allow-beast— same as--ssl-allow-beastbut used in HTTPS proxy context.
Added in 7.52.0.
--proxy-tls13-ciphers <ciphersuite list>— (TLS) specifies which cipher suites to use in the connection to your HTTPS proxy when it negotiates TLS 1.3. The list of ciphers suites must specify valid ciphers. Read up on TLS 1.3 cipher suite details on this URL:
https://curl.haxx.se/docs/ssl-ciphers.html
This option is currently used only when curl is built to use OpenSSL 1.1.1 or later. If you are using a different SSL backend you can try setting TLS 1.3 cipher suites by using the --proxy-ciphers option.
If this option is used several times, the last one will be used.
--proxy-tlsauthtype <type>— same as--tlsauthtypebut used in HTTPS proxy context.--proxy-tlspassword <string>— literally the same thing as--tlspasswordbut, plot twist, it's for when you're talking to an HTTPS proxy. 💅
Added in 7.52.0.
--proxy-tlsuser <name>— same energy as--tlsuser, but in the HTTPS proxy context. 🫶
Added in 7.52.0.
--proxy-tlsv1— identical to-1,--tlsv1, except it's used in the HTTPS proxy context. slay. ✨
Added in 7.52.0.
-U <user:password>,--proxy-user <user:password>— tells curl the username and password for proxy authentication, okay bestie? 🔑
So if you're running a Windows SSPI-enabled curl binary and you're doing either Negotiate or NTLM auth, you can make curl grab the user name and password from your environment by passing just a single colon with this option: -U :. iconic.
On systems where it works, curl hides the argument you gave from process listings. But listen — that is not enough to actually protect your credentials from other users on the same system, because they're still visible for a brief lil moment before being cleared. 😭 Such sensitive data should be pulled from a file instead, or something like that, and never used in clear text on a command line. fr fr.
If this option is used several times, the last one wins.
-x [protocol://]host[:port],-proxy [protocol://]host[:port]— use the proxy you specify. 🛰️
You can give the proxy string a protocol:// prefix. No protocol specified, or http://, gets treated as an HTTP proxy. Use socks4://, socks4a://, socks5:// or socks5h:// to request a specific SOCKS version. (the protocol support was added in curl 7.21.7)
HTTPS proxy support via the https:// protocol prefix was added in 7.52.0 for OpenSSL, GnuTLS and NSS. 🌈
Unrecognized and unsupported proxy protocols cause an error since 7.52.0. Prior versions may just ignore the protocol and use http:// instead. 💀
No port number in the proxy string? It's assumed to be 1080.
This option overrides existing environment variables that set the proxy to use. And if there's an environment variable setting a proxy, you can set proxy to "" to override it. no cap.
All operations performed over an HTTP proxy transparently get converted to HTTP, which means certain protocol-specific operations might not be available. Not the case if you can tunnel through the proxy, like with the -p, --proxytunnel option. 🕳️
Any user and password you put in the proxy string get URL decoded by curl. That's how you slip in special characters like @ by using %40, or a colon with %3a. cute.
The proxy host can be specified the exact same way as the proxy environment variables, including the protocol prefix (http://) and the embedded user + password.
If this option is used several times, the last one will be used.
--proxy1.0 <host[:port]>— use the specified HTTP 1.0 proxy. If you don't give a port number, it's assumed to be port1080. 🎀
The only difference between this and the HTTP proxy option -x, --proxy is that attempts to use CONNECT through the proxy will specify an HTTP 1.0 protocol instead of the default HTTP 1.1.
-p,--proxytunnel— when an HTTP proxy is used via-x,--proxy, this option makes curl tunnel through the proxy. 🔥 The tunnel approach is made with the HTTP proxyCONNECTrequest, and it requires that the proxy allows direct connect to the remote port number curl wants to tunnel through to.
Wanna suppress proxy CONNECT response headers when curl is set to output headers? Use --suppress-connect-headers.
See also -x, --proxy.
--pubkey <key>— (SFTP SCP) public key file name. Lets you hand over your public key in this separate file. 🗝️
If this option is used several times, the last one will be used.
(As of 7.39.0, curl attempts to automatically extract the public key from the private key file, so passing this option is generally not required. Note that this public key extraction requires libcurl to be linked against a copy of libssh2 1.2.8 or higher that is itself linked against OpenSSL.)
-Q,--quote— (FTP SFTP) send an arbitrary command to the remote FTP or SFTP server. 💬 Quote commands get sent BEFORE the transfer takes place (just after the initialPWDcommand in an FTP transfer, to be exact).- Wanna make commands happen after a successful transfer? Prefix them with a dash
'-'. - Wanna make commands be sent after curl has changed the working directory, just before the transfer command(s)? Prefix the command with a
'+'(this is only supported for FTP).
- Wanna make commands happen after a successful transfer? Prefix them with a dash
You may specify any number of commands. ✨
If the server returns failure for one of the commands, the entire operation gets aborted. So you must send syntactically correct FTP commands as RFC 959 defines to FTP servers, or one of the commands listed below to SFTP servers.
This option can be used multiple times. When speaking to an FTP server, prefix the command with an asterisk (*) to make curl continue even if the command fails — because by default curl will stop at first failure. 😤
SFTP is a binary protocol. Unlike for FTP, curl interprets SFTP quote commands itself before sending them to the server. File names may be quoted shell-style to embed spaces or special characters. Following is the list of all supported SFTP quote commands:
chgrp group file— thechgrpcommand sets the group ID of the file named by the file operand to the group ID specified by the group operand. The group operand is a decimal integer group ID.chmod mode file— thechmodcommand modifies the file mode bits of the specified file. The mode operand is an octal integer mode number.chown user file— thechowncommand sets the owner of the file named by the file operand to the user ID specified by the user operand. The user operand is a decimal integer user ID.ln source_file target_file— thelnandsymlinkcommands create a symbolic link at the target_file location pointing to the source_file location.mkdir directory_name— themkdircommand creates the directory named by the directory_name operand.pwd— thepwdcommand returns the absolute pathname of the current working directory. 👑rename source target— therenamecommand renames the file or directory named by the source operand to the destination path named by the target operand.rm file— thermcommand removes the file specified by the file operand.rmdir directory— thermdircommand removes the directory entry specified by the directory operand, provided it is empty.symlink source_file target_file— seeln.
--random-file <file>— specify the path name to a file containing what will be considered random data. 🎲 That data may be used to seed the random engine for SSL connections. See also the--egd-fileoption.-r <range>,--range <range>— (HTTP FTP SFTP FILE) retrieve a byte range (i.e. a partial document) from an HTTP/1.1, FTP or SFTP server, or a local FILE. 🌸 Ranges can be specified in a number of ways:0-499— specifies the first 500 bytes500-999— specifies the second 500 bytes-500— specifies the last 500 bytes9500-— specifies the bytes from offset 9500 and forward0-0,-1— specifies the first and last byte only(*)(HTTP)100-199,500-599— specifies two separate 100-byte ranges(*) (HTTP)
(*) = NOTE that this will cause the server to reply with a multipart response! 😭
Only digit characters (0-9) are valid in the 'start' and 'stop' fields of the 'start-stop' range syntax. If a non-digit character is given in the range, the server's response will be unspecified, depending on the server's configuration. 🚨
You should also be aware that many HTTP/1.1 servers don't even have this feature enabled, so when you attempt to get a range, you'll instead get the whole document. 💀
FTP and SFTP range downloads only support the simple 'start-stop' syntax (optionally with one of the numbers omitted). FTP use depends on the extended FTP command SIZE.
If this option is used several times, the last one will be used.
--raw— (HTTP) when used, it disables all internal HTTP decoding of content or transfer encodings and instead makes them passed on unaltered, raw. 🧚
Added in 7.16.2.
-e,--referer <URL>— (HTTP) sends the "Referrer Page" information to the HTTP server. This can also be set with the-H,--headerflag of course. When used with-L,--locationyou can append";auto"to the-e,--refererURL to make curl automatically set the previous URL when it follows aLocation:header. The";auto"string can be used alone, even if you don't set an initial-e,--referer. 🌈
If this option is used several times, the last one will be used.
See also -A, --user-agent and -H, --header.
-J,--remote-header-name— (HTTP) this option tells the-O,--remote-nameoption to use the server-specifiedContent-Dispositionfilename instead of extracting a filename from the URL. 📁
If the server specifies a file name and a file with that name already exists in the current working directory, it will not be overwritten and an error will occur. If the server doesn't specify a file name, then this option has no effect.
There's no attempt to decode %-sequences (yet) in the provided file name, so this option may hand you some rather unexpected file names. 🙈
WARNING: use this one judiciously, okay? Especially on Windows 😭. A rogue server could send you the name of a DLL or some other file that Windows — or some third-party software — could possibly load automatically. Not a vibe. Be careful. 💀
--remote-name-all— this one flips the default for all the URLs you gave it, so each one gets treated as if you'd used-O,--remote-name. 🫶 And heads up: if you want to disable that for one specific URL after--remote-name-allhas already been used, you gotta use-o -or--no-remote-name. No takebacks otherwise, bestie.
Added in 7.19.0.
-O,--remote-name— writes the output to a local file, and it names the file like the remote file we got. (Only the file part of the remote name is used — the path gets cut off, bye 👋.)
The file lands in the current working directory. Want it somewhere else? Make sure you cd before you invoke curl with this option, because curl is not gonna do it for you. 🎀
The remote file name it saves under is pulled from the URL you gave it, nothing else — and if that file already exists, it gets overwritten. Periodt. If you'd rather let the server pick the name, that's -J, --remote-header-name, which you can use in addition to this option. And if the server picks a name that already exists, that one does not get overwritten. ✨
Also: there is no URL decoding done on the file name. If the name has %20 or other URL-encoded bits in it, those land as-is in your filename. iykyk 🙈
You can use this option as many times as the number of URLs you have. Slay.
-R,--remote-time— when you use it, curl will attempt to figure out the timestamp of the remote file, and if that timestamp is available, it makes your local file get that same timestamp. 🕰️--request-target— (HTTP) tells curl to use an alternative "target" (path) instead of using the path as provided in the URL. Highkey useful when you wanna fire off HTTP requests without a leading slash or other data that doesn't follow the regular URL pattern — think"OPTIONS *". 👑
Added in 7.55.0.
-X command,-request <command>— (HTTP) specifies a custom request method to use when talking to the HTTP server. The request method you specify gets used instead of the method that would otherwise be used (which defaults toGET). Go read the HTTP 1.1 spec for the details and explanations, bestie 📚. Common extra HTTP requests includePUTandDELETE, but related technologies like WebDAV offerPROPFIND,COPY,MOVEand more. 🌈
Normally you don't need this option — all sorts of GET, HEAD, POST and PUT requests are rather invoked by using dedicated command line options.
Important: this option only changes the actual word used in the HTTP request. It does not alter the way curl behaves. So for example, if you want to make a proper HEAD request, using -X HEAD will not suffice — you need -I, --head. 😤
The method string you set with -X, --request will be used for all requests, which — if you for example use -L, --location — may cause unintended side-effects when curl doesn't change the request method according to the HTTP 30x response codes and similar. Not me warning you twice. 💀
(FTP) Specifies a custom FTP command to use instead of LIST when doing file lists with FTP.
(POP3) Specifies a custom POP3 command to use instead of LIST or RETR. (Added in 7.26.0)
(IMAP) Specifies a custom IMAP command to use instead of LIST. (Added in 7.30.0)
(SMTP) Specifies a custom SMTP command to use instead of HELP or VRFY. (Added in 7.34.0)
If this option is used several times, the last one will be used. 🎀
---resolve <host:port:address[,address]...>— provides a custom address for a specific host and port pair. Using this, you can make curl's request(s) use a specified address and prevent the otherwise normally resolved address from being used. Think of it like an/etc/hostsalternative, but on the command line. Iconic. 👑
The port number should be the number used for the specific protocol the host will be used for. Which means: you need several entries if you wanna provide an address for the same host but on different ports. Fr fr.
By specifying '*' as host you can tell curl to resolve any host and specific port pair to the specified address. The wildcard is resolved last, so any --resolve with a specific host and port will be used first. 🦋
The address provided by this option will be used even if -4, --ipv4 or -6, --ipv6 is set to make curl use another IP version.
Support for providing the IP address within [brackets] was added in 7.57.0.
Support for providing multiple IP addresses per entry was added in 7.59.0.
Support for resolving with wildcard was added in 7.64.0.
This option can be used many times to add many host names to resolve. ✨
Added in 7.21.3.
--retry-connrefused— in addition to the other conditions, treatsECONNREFUSEDas a transient error too, for--retry. This option is used together with--retry. 🫶
Added in 7.52.0.
--retry-delay <seconds>— makes curl sleep this amount of time before each retry when a transfer has failed with a transient error (it changes the default backoff time algorithm between retries). This option is only interesting if--retryis also used. Setting this delay to zero makes curl use the default backoff time. 😴
If this option is used several times, the last one will be used.
Added in 7.12.3.
--retry-max-time <seconds>— the retry timer is reset before the first transfer attempt. Retries will be done as usual (see--retry) as long as the timer hasn't reached this given limit. ⏰
Notice that if the timer hasn't reached the limit, the request will be made — and while performing, it may take longer than this given time period. To limit a single request's maximum time, use -m, --max-time. Set this option to zero to not timeout retries.
If this option is used several times, the last one will be used.
Added in 7.12.3.
--retry <num>— if a transient error is returned when curl tries to perform a transfer, it will retry this number of times before giving up. Setting the number to0makes curl do no retries (which is the default, sadly 😔).
"Transient error" means either: a timeout, an FTP 4xx response code, or an HTTP 408 or 5xx response code. 💅
When curl is about to retry a transfer, it will first wait one second, and then for all forthcoming retries it doubles the waiting time until it reaches 10 minutes — which then becomes the delay between the rest of the retries. By using --retry-delay you disable this exponential backoff algorithm. See also --retry-max-time to limit the total time allowed for retries. 📈
If this option is used several times, the last one will be used.
Added in 7.12.3.
--sasl-ir— enables initial response in SASL authentication. 🎀
Added in 7.31.0.
--service-name <name>— this option lets you change the service name for SPNEGO.
Example time, bestie: --negotiate --service-name sockd would use sockd/server-name. 🌸
Added in 7.43.0.
-S,--show-error— when used with-s,--silent, it makes curl show an error message if it fails. That's the whole thing. 🫶-s,--silent— silent / quiet mode. Don't show the progress meter or error messages. Makes Curl mute. 🤐 It will still output the data you asked for, potentially even to the terminal/stdout, unless you redirect it.
Use -S, --show-error in addition to this option to disable the progress meter but still show error messages. Iconic combo.
See also -v, --verbose and --stderr.
--socks4 <host[:port]>— use the specified SOCKS4 proxy. If the port number isn't specified, it's assumed at port1080. 🧦
This option overrides any previous use of -x, --proxy, since they're mutually exclusive. Pick a lane, bestie.
Since 7.21.7, this option is superfluous, since you can specify a socks4 proxy with -x, --proxy using a socks4:// protocol prefix.
Since 7.52.0, --preproxy can be used to specify a SOCKS proxy at the same time -x, --proxy is used with an HTTP/HTTPS proxy. In such a case curl first connects to the SOCKS proxy and then connects (through SOCKS) to the HTTP or HTTPS proxy. 🌀
If this option is used several times, the last one will be used.
Added in 7.15.2.
--socks4a <host[:port]>— use the specified SOCKS4a proxy. If the port number isn't specified, it's assumed at port1080. (Yes, same port. Slay.) 🧦
This option overrides any previous use of -x, --proxy, since they're mutually exclusive.
Since 7.21.7, this option is superfluous, since you can specify a socks4a proxy with -x, --proxy using a socks4a:// protocol prefix.
Since 7.52.0, --preproxy can be used to specify a SOCKS proxy at the same time -x, --proxy is used with an HTTP/HTTPS proxy. In such a case curl first connects to the SOCKS proxy and then connects (through SOCKS) to the HTTP or HTTPS proxy. 🌀
If this option is used several times, the last one will be used.
Added in 7.18.0.
--socks5-basic— tells curl to use username/password authentication when connecting to a SOCKS5 proxy. The username/password authentication is enabled by default. Use--socks5-gssapito force GSS-API authentication to SOCKS5 proxies. 🔐
Added in 7.55.0.
--socks5-gssapi-nec— so, as part of the GSS-API negotiation, a protection mode is negotiated. RFC 1961 says in section 4.3/4.4 that it should be protected, but the NEC reference implementation does not. The option--socks5-gssapi-necallows the unprotected exchange of the protection mode negotiation. Ngl, compatibility mode is such a mood. 😭
Added in 7.19.4.
--socks5-gssapi-service <name>— the default service name for a socks server isrcmd/server-fqdn. This option lets you change it. ✨
Examples, and yes these are verbatim, so like, copy them exactly bestie:
--socks5 proxy-name --socks5-gssapi-service sockd
would use sockd/proxy-name
--socks5 proxy-name --socks5-gssapi-service sockd/real-name
would use sockd/real-name for cases where the proxy-name does not match the principal name.
Added in 7.19.4.
🧦 --socks5-gssapi
Tells curl to use GSS-API authentication when connecting to a SOCKS5 proxy. The GSS-API authentication is enabled by default (if curl is compiled with GSS-API support). Use --socks5-basic to force username/password authentication to SOCKS5 proxies.
Added in 7.55.0.
🦋 --socks5-hostname <host[:port]>
Use the specified SOCKS5 proxy — and let the proxy resolve the host name, which is the whole point. If the port number is not specified, it is assumed at port 1080.
This option overrides any previous use of -x, --proxy, as they are mutually exclusive. (they cannot both be the main character, iykyk)
Since 7.21.7, this option is superfluous since you can specify a socks5 hostname proxy with -x, --proxy using a socks5h:// protocol prefix.
Since 7.52.0, --preproxy can be used to specify a SOCKS proxy at the same time -x, --proxy is used with an HTTP/HTTPS proxy. In such a case curl first connects to the SOCKS proxy and then connects (through SOCKS) to the HTTP or HTTPS proxy. 💅
If this option is used several times, the last one will be used.
Added in 7.18.0.
🧚 --socks5 <host[:port]>
Use the specified SOCKS5 proxy — but resolve the host name locally. If the port number is not specified, it is assumed at port 1080.
This option overrides any previous use of -x, --proxy, as they are mutually exclusive.
Since 7.21.7, this option is superfluous since you can specify a socks5 proxy with -x, --proxy using a socks5:// protocol prefix.
Since 7.52.0, --preproxy can be used to specify a SOCKS proxy at the same time -x, --proxy is used with an HTTP/HTTPS proxy. In such a case curl first connects to the SOCKS proxy and then connects (through SOCKS) to the HTTP or HTTPS proxy.
If this option is used several times, the last one will be used.
Heads up bestie, this one has a whole list of things it doesn't vibe with: this option (as well as --socks4) does not work with IPV6, FTPS or LDAP. 😭
Added in 7.18.0.
🏎️ -Y speed, --speed-limit <speed>
If a download is slower than this given speed (in bytes per second) for speed-time seconds it gets aborted. speed-time is set with -y, --speed-time and is 30 if not set. So like, curl will literally give up on a slug-speed download. iconique.
If this option is used several times, the last one will be used.
⏱️ --y seconds, -speed-time <seconds>
(ok so the manpage literally prints it as --y seconds | -speed-time <seconds> and i'm not about to "fix" it, that's not my job, i'm just the messenger 💀)
If a download is slower than speed-limit bytes per second during a speed-time period, the download gets aborted. If speed-time is used, the default speed-limit will be 1 unless set with -Y, --speed-limit.
This option controls transfers and thus will not affect slow connects etc. If this is a concern for you, try the --connect-timeout option.
If this option is used several times, the last one will be used.
🦁 --ssl-allow-beast
This option tells curl to not work around a security flaw in the SSL3 and TLS1.0 protocols known as BEAST. If this option isn't used, the SSL layer may use workarounds known to cause interoperability problems with some older SSL implementations.
WARNING: this option loosens the SSL security, and by using this flag you ask for exactly that. Like, you are literally requesting the insecure thing. no cap. 🙈
Added in 7.25.0.
🚫 --ssl-no-revoke
(Schannel) This option tells curl to disable certificate revocation checks.
WARNING: this option loosens the SSL security, and by using this flag you ask for exactly that. same energy as above bestie, pls be careful. 💀
Added in 7.44.0.
🔒 --ssl-reqd
(FTP IMAP POP3 SMTP) Require SSL/TLS for the connection. Terminates the connection if the server doesn't support SSL/TLS. it's giving "my way or the highway" and honestly? respect.
This option was formerly known as --ftp-ssl-reqd.
Added in 7.20.0.
🌸 --ssl
(FTP IMAP POP3 SMTP) Try to use SSL/TLS for the connection. Reverts to a non-secure connection if the server doesn't support SSL/TLS. See also --ftp-ssl-control and --ssl-reqd for different levels of encryption required. (soft launch security, basically)
This option was formerly known as --ftp-ssl (Added in 7.11.0). That option name can still be used but will be removed in a future version.
Added in 7.20.0.
💀 -2, --sslv2
(SSL) Forces curl to use SSL version 2 when negotiating with a remote SSL server. Sometimes curl is built without SSLv2 support. SSLv2 is widely considered insecure (see RFC 6176). seriously bestie, RFC 6176 exists specifically to say this is bad news.
See also --http1.1 and --http2. -2, --sslv2 requires that the underlying libgnurl was built to support TLS. This option overrides -3, --sslv3 and -1, --tlsv1 and --tlsv1.1 and --tlsv1.2.
🫠 -3, --sslv3
(SSL) Forces curl to use SSL version 3 when negotiating with a remote SSL server. Sometimes curl is built without SSLv3 support. SSLv3 is widely considered insecure (see RFC 7568). yes another RFC saying nope.
See also --http1.1 and --http2. -3, --sslv3 requires that the underlying libgnurl was built to support TLS. This option overrides -2, --sslv2 and -1, --tlsv1 and --tlsv1.1 and --tlsv1.2.
📝 --stderr
Redirect all writes to stderr to the specified file instead. If the file name is a plain -, it is instead written to stdout. sneaky little trick, love that for you.
If this option is used several times, the last one will be used.
See also -v, --verbose and -s, --silent.
✨ --styled-output
Enables the automatic use of bold font styles when writing HTTP headers to the terminal. Use --no-styled-output to switch them off. yes it literally just makes your headers bold. aesthetic. 🎀
Added in 7.61.0.
🙈 --suppress-connect-headers
When -p, --proxytunnel is used and a CONNECT request is made don't output proxy CONNECT response headers. This option is meant to be used with -D, --dump-header or -i, --include which are used to show protocol headers in the output. It has no effect on debug options such as -v, --verbose or --trace, or any statistics.
See also -D, --dump-header and -i, --include and -p, --proxytunnel.
⚡ --tcp-fastopen
Enable use of TCP Fast Open (RFC7413). zoom zoom. 🔥
Added in 7.49.0.
🚀 --tcp-nodelay
Turn on the TCP_NODELAY option. See the curl_easy_setopt(3) man page for details about this option.
Since 7.50.2, curl sets this option by default and you need to explicitly switch it off if you don't want it on.
Added in 7.11.2.
📺 -t opt=val, --telnet-option <opt=val>
Pass options to the telnet protocol. Supported options are:
TTYPE=<term>— Sets the terminal type.XDISPLOC=<X display>— Sets the X display location.NEW_ENV=<var,val>— Sets an environment variable.
🍓 --tftp-blksize <value>
(TFTP) Set TFTP BLKSIZE option (must be >512). This is the block size that curl will try to use when transferring data to or from a TFTP server. By default 512 bytes will be used.
If this option is used several times, the last one will be used.
Added in 7.20.0.
🧊 --tftp-no-options
(TFTP) Tells curl not to send TFTP options requests.
This option improves interop with some legacy servers that do not acknowledge or properly implement TFTP options. When this option is used --tftp-blksize is ignored. (yes, fully ignored, so don't bother setting it 💅)
Added in 7.48.0.
🕰️ -z -time, --time-cond <time>
(again, verbatim from the page: -z -time | --time-cond <time> — i just report, i don't editorialize. ok maybe a little.)
(HTTP FTP) Request a file that has been modified later than the given time and date, or one that has been modified before that time. The <date expression> can be all sorts of date strings or if it doesn't match any internal ones, it is taken as a filename and tries to get the modification date (mtime) from <file> instead. See the curl_getdate(3) man pages for date expression details.
Start the date expression with a dash (-) to make it request for a document that is older than the given date/time, default is a document that is newer than the specified date/time.
If this option is used several times, the last one will be used.
👑 --tls-max <VERSION>
(SSL) VERSION defines maximum supported TLS version. The minimum acceptable version is set by tlsv1.0, tlsv1.1, tlsv1.2 or tlsv1.3.
default— Use up to recommended TLS version.1.0— Use up to TLSv1.0.1.1— Use up to TLSv1.1.1.2— Use up to TLSv1.2.1.3— Use up to TLSv1.3.
See also --tlsv1.0 and --tlsv1.1 and --tlsv1.2 and --tlsv1.3. --tls-max requires that the underlying libgnurl was built to support TLS. Added in 7.54.0.
🔐 --tls13-ciphers <list of TLS 1.3 ciphersuites>
(TLS) Specifies which cipher suites to use in the connection if it negotiates TLS 1.3. The list of ciphers suites must specify valid ciphers. Read up on TLS 1.3 cipher suite details on this URL: https://curl.haxx.se/docs/ssl-ciphers.html
This option is currently used only when curl is built to use OpenSSL 1.1.1 or later. If you are using a different SSL backend you can try setting TLS 1.3 cipher suites by using the --ciphers option.
If this option is used several times, the last one will be used.
🎀 --tlsauthtype <type>
Set TLS authentication type. Currently, the only supported option is "SRP", for TLS-SRP (RFC 5054). If --tlsuser and --tlspassword are specified but --tlsauthtype is not, then this option defaults to "SRP". This option works only if the underlying libcurl is built with TLS-SRP support, which requires OpenSSL or GnuTLS with TLS-SRP support.
Added in 7.21.4.
🗝️ --tlspassword
Set password for use with the TLS authentication method specified with --tlsauthtype. Requires that --tlsuser also be set. (they're a package deal, fr fr)
Added in 7.21.4.
🌈 --tlsuser <name>
Set username for use with the TLS authentication method specified with --tlsauthtype. Requires that --tlspassword also is set.
Added in 7.21.4.
⭐ --tlsv1.0
(TLS) Forces curl to use TLS version 1.0 or later when connecting to a remote TLS server.
--tlsv1.1— bestie, this one forces curl to speak TLS version 1.1 or later when it's connecting to a remote TLS server. Like, no downgrading, we're not doing that today. ✨
Added in 7.34.0.
--tlsv1.2— same deal but for TLS version 1.2 or later. curl said "we're negotiating, but we're negotiating upward." 💅
Added in 7.34.0.
--tlsv1.3— forces TLS version 1.3 or later. Iconic, secure, modern. 👑
Note that TLS 1.3 is only supported by a subset of TLS backends. At the time of this writing, they are BoringSSL, NSS, and Secure Transport (on iOS 11 or later, and macOS 10.13 or later).
Added in 7.52.0.
-1,--tlsv1—(SSL)tells curl to use at least TLS version 1.x when negotiating with a remote TLS server. That means TLS version 1.0 or higher, periodt.
See also --http1.1 and --http2. -1, --tlsv1 requires that the underlying libgnurl was built to support TLS. This option overrides --tlsv1.1 and --tlsv1.2 and --tlsv1.3. So yeah, it's the bossy one in the TLS friend group. 😌
--tr-encoding—(HTTP)requests a compressedTransfer-Encodingresponse using one of the algorithms curl supports, and uncompresses the data while receiving it. Like getting a gift wrapped in compression and curl unwraps it for you in real time. 🎁
Added in 7.21.6.
--trace-ascii<file> — enables a full trace dump of all incoming and outgoing data, including descriptive information, to the given output file. Use"-"as filename to have the output sent to stdout.
This is very similar to --trace, but leaves out the hex part and only shows the ASCII part of the dump. It makes smaller output that might be easier to read for untrained humans. (No shade, we were all untrained humans once. 🫶)
If this option is used several times, the last one will be used.
This option overrides --trace and -v, --verbose.
--trace-time— prepends a time stamp to each trace or verbose line that curl displays. Timestamps on everything, because chronology is the moment. ⏰
Added in 7.14.0.
--trace<file> — enables a full trace dump of all incoming and outgoing data, including descriptive information, to the given output file. Use"-"as filename to have the output sent to stdout. Use"%"as filename to have the output sent to stderr.
If this option is used several times, the last one will be used.
This option overrides -v, --verbose and --trace-ascii.
--unix-socket<path> —(HTTP)connects through this Unix domain socket, instead of using the network. Lowkey networking but not networking. 🦋
Added in 7.40.0.
-Tfile,--upload-file<file> — this transfers the specified local file to the remote URL. If there is no file part in the specified URL, curl will append the local file name. NOTE that you must use a trailing/on the last directory to really prove to Curl that there is no file name or curl will think that your last directory name is the remote file name to use. That will most likely cause the upload operation to fail. So please, bestie, give it the slash. 🙏 If this is used on an HTTP(S) server, thePUTcommand will be used.
Use the file name "-" (a single dash) to use stdin instead of a given file. Alternately, the file name "." (a single period) may be specified instead of "-" to use stdin in non-blocking mode to allow reading server output while stdin is being uploaded.
You can specify one -T, --upload-file for each URL on the command line. Each -T, --upload-file + URL pair specifies what to upload and to where. curl also supports "globbing" of the -T, --upload-file argument, meaning that you can upload multiple files to a single URL by using the same URL globbing style supported in the URL, like this:
curl --upload-file {file1,file2} http://www.example.com
or even
curl -T img[1-1000].png ftp://ftp.example.com/upload/
When uploading to an SMTP server: the uploaded data is assumed to be RFC 5322 formatted. It has to feature the necessary set of headers and mail body formatted correctly by the user as curl will not transcode nor encode it further in any way. So curl is literally just the messenger. 💌
--url<url> — specify a URL to fetch. This option is mostly handy when you want to specify URL(s) in a config file. Config file girlies rise up. 🎀
If the given URL is missing a scheme name (such as "http://" or "ftp://" etc) then curl will make a guess based on the host. If the outermost sub-domain name matches DICT, FTP, IMAP, LDAP, POP3 or SMTP then that protocol will be used, otherwise HTTP will be used. Since 7.45.0 guessing can be disabled by setting a default protocol, see --proto-default for details.
This option may be used any number of times. To control where this URL is written, use the -o, --output or the -O, --remote-name options.
-B,--use-ascii—(FTP LDAP)enable ASCII transfer. For FTP, this can also be enforced by using a URL that ends with";type=A". This option causes data sent to stdout to be in text mode for win32 systems. Plain text supremacy, I guess. 📝-A<name>,--user-agent<name> —(HTTP)specify the User-Agent string to send to the HTTP server. To encode blanks in the string, surround the string with single quote marks. This header can also be set with the-H,--headeror the--proxy-headeroptions.
If this option is used several times, the last one will be used.
-u<user:password>,--user<user:password> — specify the user name and password to use for server authentication. Overrides-n,--netrcand--netrc-optional.
If you simply specify the user name, curl will prompt for a password.
The user name and passwords are split up on the first colon, which makes it impossible to use a colon in the user name with this option. The password can, still. So colons are welcome in the password but absolutely not in the username. Rules are rules. 💅
On systems where it works, curl will hide the given option argument from process listings. This is not enough to protect credentials from possibly getting seen by other users on the same system as they will still be visible for a brief moment before cleared. Such sensitive data should be retrieved from a file instead or similar and never used in clear text in a command line.
Okay bestie, real talk: do not put your passwords in a command line and think you're safe. curl can hide it for a sec but other users may still glimpse it. Files are the move. 🔐
When using Kerberos V5 with a Windows based server you should include the Windows domain name in the user name, in order for the server to successfully obtain a Kerberos Ticket. If you don't then the initial authentication handshake may fail.
When using NTLM, the user name can be specified simply as the user name, without the domain, if there is a single domain and forest in your setup for example.
To specify the domain name use either Down-Level Logon Name or UPN (User Principal Name) formats. For example, EXAMPLE\user and user@example.com respectively.
If you use a Windows SSPI-enabled curl binary and perform Kerberos V5, Negotiate, NTLM or Digest authentication then you can tell curl to select the user name and password from your environment by specifying a single colon with this option: "-u :".
If this option is used several times, the last one will be used.
-v,--verbose— makes curl verbose during the operation. Useful for debugging and seeing what's going on "under the hood". A line starting with'>'means "header data" sent by curl,'<'means "header data" received by curl that is hidden in normal cases, and a line starting with'*'means additional info provided by curl. It's giving detective mode. 🔍
If you only want HTTP headers in the output, -i | --include might be the option you're looking for.
If you think this option still doesn't give you enough details, consider using --trace or --trace-ascii instead.
Use -s | --silent to make curl really quiet.
See also -i | --include. This option overrides --trace and --trace-ascii.
-V,--version— displays information about curl and the libcurl version it uses. Receipts on receipts. 🧾
The first line includes the full version of curl, libcurl and other 3rd party libraries linked with the executable.
The second line (starts with "Protocols:") shows all protocols that libcurl reports to support.
The third line (starts with "Features:") shows specific features libcurl reports to offer. Available features include:
IPv6— you can use IPv6 with this.krb4— Krb4 for FTP is supported.SSL— SSL versions of various protocols are supported, such as HTTPS, FTPS, POP3S and so on.libz— automatic decompression of compressed files over HTTP is supported.NTLM— NTLM authentication is supported.Debug— this curl uses a libcurl built with Debug. This enables more error-tracking and memory debugging etc. For curl-developers only! 👀AsynchDNS— this curl uses asynchronous name resolves. Asynchronous name resolves can be done using either the c-ares or the threaded resolver backends.SPNEGO— SPNEGO authentication is supported.Largefile— this curl supports transfers of large files, files larger than 2GB.IDN— this curl supports IDN - international domain names.GSS-API— GSS-API is supported.SSPI— SSPI is supported.TLS-SRP— SRP (Secure Remote Password) authentication is supported for TLS.HTTP2— HTTP/2 support has been built-in.UnixSockets— Unix sockets support is provided.HTTPS-proxy— this curl is built to support HTTPS proxy.Metalink— this curl supports Metalink (both version 3 and 4 (RFC 5854)), which describes mirrors and hashes. curl will use mirrors for failover if there are errors (such as the file or server not being available).PSL— PSL is short for Public Suffix List and means that this curl has been built with knowledge about "public suffixes".MultiSSL— this gnurl supports multiple TLS backends.
-w <format>,--write-out <format>— ok so this one is the ✨aesthetic✨ option: it makescurlprint a little info dump on stdout after a transfer finishes. The<format>is a string, and it can be plain text mixed with as many variables as you want. You can hand it a literal"string", or makecurlread the format out of a file with@filename, or — and this is so cute — write@-to have it read the format from stdin. 💅
Every variable in the output format gets swapped for whatever value or text curl thinks fits, as listed below. All variables are written as %{variable_name}, and if you want a normal literal % you just double it up: %%. Newline? \n. Carriage return? \r. Tab space? \t. Obsessed with the control over here. 🎀
Output goes to standard output by default, but you can flip it to standard error with %{stderr}.
NOTE: the%-symbol is a special symbol in the win32-environment, where all occurrences of%must be doubled when you're using this option. Not me warning you twice 💀
The variables available are:
content_type— the Content-Type of the requested document, if there was any.filename_effective— the ultimate filename thatcurlwrites out to. This is only meaningful ifcurlis told to write to a file with the-O,--remote-nameor-o,--outputoption. It's most useful in combination with the-J,--remote-header-nameoption. (Added in 7.26.0)ftp_entry_path— the initial pathcurlended up in when logging on to the remote FTP server. (Added in 7.15.4)http_code— the numerical response code that was found in the last retrieved HTTP(S) or FTP(s) transfer. In 7.18.2 the aliasresponse_codewas added to show the same info.http_connect— the numerical code that was found in the last response (from a proxy) to acurlCONNECT request. (Added in 7.12.4)http_version— the http version that was effectively used. (Added in 7.50.0)local_ip— the IP address of the local end of the most recently done connection — can be either IPv4 or IPv6 (Added in 7.29.0)local_port— the local port number of the most recently done connection (Added in 7.29.0)num_connects— number of new connects made in the recent transfer. (Added in 7.12.3)num_redirects— number of redirects that were followed in the request. (Added in 7.12.3)proxy_ssl_verify_result— the result of the HTTPS proxy's SSL peer certificate verification that was requested.0means the verification was successful. (Added in 7.52.0)redirect_url— when an HTTP request was made without-L,--locationto follow redirects (or when--max-rediris met), this variable will show the actual URL a redirect would have gone to. (Added in 7.18.2)remote_ip— the remote IP address of the most recently done connection — can be either IPv4 or IPv6 (Added in 7.29.0)remote_port— the remote port number of the most recently done connection (Added in 7.29.0)scheme— the URL scheme (sometimes called protocol) that was effectively used (Added in 7.52.0)size_download— the total amount of bytes that were downloaded.size_header— the total amount of bytes of the downloaded headers.size_request— the total amount of bytes that were sent in the HTTP request.size_upload— the total amount of bytes that were uploaded.speed_download— the average download speed thatcurlmeasured for the complete download. Bytes per second.speed_upload— the average upload speed thatcurlmeasured for the complete upload. Bytes per second.ssl_verify_result— the result of the SSL peer certificate verification that was requested.0means the verification was successful. (Added in 7.19.0)stderr— from this point on, the-w,--write-outoutput will be written to standard error. (Added in 7.63.0)stdout— from this point on, the-w,--write-outoutput will be written to standard output. This is the default, but can be used to switch back after switching to stderr. (Added in 7.63.0)time_appconnect— the time, in seconds, it took from the start until the SSL/SSH/etc connect/handshake to the remote host was completed. (Added in 7.19.0)time_connect— the time, in seconds, it took from the start until the TCP connect to the remote host (or proxy) was completed.time_namelookup— the time, in seconds, it took from the start until the name resolving was completed.time_pretransfer— the time, in seconds, it took from the start until the file transfer was just about to begin. This includes all pre-transfer commands and negotiations that are specific to the particular protocol(s) involved.time_redirect— the time, in seconds, it took for all redirection steps including name lookup, connect, pretransfer and transfer before the final transaction was started.time_redirectshows the complete execution time for multiple redirections. (Added in 7.12.3)time_starttransfer— the time, in seconds, it took from the start until the first byte was just about to be transferred. This includestime_pretransferand also the time the server needed to calculate the result.time_total— the total time, in seconds, that the full operation lasted.url_effective— the URL that was fetched last. This is most meaningful if you've toldcurlto follow location: headers.
If this option is used several times, the last one will be used. (Classic last-one-wins energy, we love consistency 💖)
--xattr— when you're saving output to a file, this option tellscurlto store certain file metadata in extended file attributes. Currently, the URL is stored in thexdg.origin.urlattribute and, for HTTP, the content type is stored in themime_typeattribute. If the file system does not support extended attributes, a warning is issued. 🫶
🎀 FILES 🎀
~/.curlrc— the default config file, see--configfor details. Your little home-base vibe file 💜
💖 ENVIRONMENT 💖
Ok so the environment variables can be specified in lower case or upper case — either is fine, bestie. The lower case version has precedence. http_proxy is the exception though, because it's only available in lower case. 🙈
Using an environment variable to set the proxy has the same effect as using the --proxy option.
http_proxy [protocol://]<host>[:port]— sets the proxy server to use for HTTP.HTTPS_PROXY [protocol://]<host>[:port]— sets the proxy server to use for HTTPS.[url-protocol]_PROXY [protocol://]<host>[:port]— sets the proxy server to use for[url-protocol], where the protocol is a protocol thatcurlsupports and as specified in a URL. FTP, FTPS, POP3, IMAP, SMTP, LDAP etc. Like, the whole roster 🦋ALL_PROXY [protocol://]<host>[:port]— sets the proxy server to use if no protocol-specific proxy is set.NO_PROXY <comma-separated list of hosts/domains>— a list of host names that shouldn't go through any proxy. If set to an asterisk'*'only, it matches all hosts. Each name in this list is matched as either a domain name which contains the hostname, or the hostname itself.
This environment variable disables use of the proxy even when specified with the -x, --proxy option. Like, that is:
NO_PROXY=direct.example.com curl -x http://proxy.example.com
http://direct.example.com
accesses the target URL directly, and
NO_PROXY=direct.example.com curl -x http://proxy.example.com
http://somewhere.example.com
accesses the target URL through the proxy. Same command, different outcome — the way NO_PROXY just casually overrides everything is giving main character energy, honestly 😭
The list of host names can also be include numerical IP addresses, and IPv6 versions should then be given without enclosing brackets. (Yes, the original says "can also be include" — we're keeping it, character for character, that's the rule 💅)
🧚 PROXY PROTOCOL PREFIXES 🧚
Since curl version 7.21.7, the proxy string may be specified with a protocol:// prefix to specify alternative proxy protocols. Iconic upgrade, no cap.
If no protocol is specified in the proxy string or if the string doesn't match a supported one, the proxy will be treated as an HTTP proxy.
The supported proxy protocol prefixes are as follows:
http://— makes it use it as an HTTP proxy. The default if no scheme prefix is used.https://— makes it treated as an HTTPS proxy.socks4://— makes it the equivalent of--socks4socks4a://— makes it the equivalent of--socks4asocks5://— makes it the equivalent of--socks5socks5h://— makes it the equivalent of--socks5-hostname
Slay, that's the whole prefix menu 🌸
💀 EXIT STATUS 💀
okay bestie so when stuff goes sideways, curl (and gnurl, hi) bails out with one of these numbered codes. Each one comes with its own little error message, and ngl the numbers are kind of a whole personality. Here's the whole lineup as of this writing:
1— Unsupported protocol. This particular build of curl just straight up has no support for that protocol. Not today, bestie. 😭2— Failed to initialize. It couldn't even get out of bed. 😴3— URL malformed. The syntax was not it. 💅4— A feature or option that was needed to do the request you wanted was not enabled, or was explicitly disabled at build-time. To fix this you probably need a whole other build of libcurl! Sorry babe.5— Couldn't resolve proxy. That proxy host you gave? Unresolvable, ghosted fr.6— Couldn't resolve host. The remote host you wanted was not resolved. no cap.7— Failed to connect to host. It tried, it really did. 🫶8— Weird server reply. The server sent data curl couldn't parse. it's giving chaos.9— FTP access denied. The server denied login, or denied access to the particular resource or directory you were trying to reach. Most often this means you tried to change to a directory that doesn't exist on the server. 🙈10— FTP accept failed. While waiting for the server to connect back during an active FTP session, an error code got sent over the control connection or something similar. 💀11— FTP weird PASS reply. Curl couldn't parse the reply to thePASSrequest.12— During an active FTP session, while waiting for the server to connect back to curl, the timeout expired. clock ran out, bestie. ⏰13— FTP weird PASV reply. Curl couldn't parse the reply to thePASVrequest.14— FTP weird 227 format. Curl couldn't parse the 227-line the server sent. like why is it formatted like that 😭15— FTP can't get host. Couldn't resolve the host IP we got in the 227-line.16— HTTP/2 error. A problem was detected in the HTTP2 framing layer. This one's somewhat generic and can be any one of several problems — check the error message for the details, iykyk.17— FTP couldn't set binary. Couldn't switch the transfer method to binary.18— Partial file. Only a part of the file was transferred. half a snack 🍓19— FTP couldn't download/access the given file; theRETR(or similar) command failed.21— FTP quote error. A quote command came back with an error from the server.22— HTTP page not retrieved. The requested url was not found, or returned another error with the HTTP error code being 400 or above. This return code only shows up if-f|--failis used. 🔥23— Write error. Curl couldn't write data to a local filesystem or similar. permission vibes? 😬25— FTP couldn't STOR file. The server denied theSTORoperation, which is used for FTP uploading.26— Read error. Various reading problems. vague on purpose, honestly.27— Out of memory. A memory allocation request failed. 😵💫28— Operation timeout. The specified time-out period was reached according to the conditions.30— FTP PORT failed. ThePORTcommand failed. Not all FTP servers support thePORTcommand — try doing the transfer usingPASVinstead! ✨31— FTP couldn't use REST. TheRESTcommand failed. That command is used for resumed FTP transfers.33— HTTP range error. The range "command" didn't work. 💀34— HTTP post error. Internal post-request generation error.35— SSL connect error. The SSL handshaking failed. awkward handshake, we've all been there. 🫠36— Bad download resume. Couldn't continue an earlier aborted download.37— FILE couldn't read file. Failed to open the file. Permissions, maybe?38— LDAP cannot bind. The LDAP bind operation failed.39— LDAP search failed.41— Function not found. A required LDAP function was not found.42— Aborted by callback. An application told curl to abort the operation.43— Internal error. A function was called with a bad parameter.45— Interface error. A specified outgoing interface could not be used.47— Too many redirects. When following redirects, curl hit the maximum amount. girl got lost in the loop 💫48— Unknown option specified to libcurl. This means you passed a weird option to curl that got passed on to libcurl and rejected. Read up in the manual! 📖49— Malformed telnet option.51— The peer's SSL certificate or SSH MD5 fingerprint was not OK. 🚩52— The server didn't reply anything, which here counts as an error. silent treatment, rude.53— SSL crypto engine not found.54— Cannot set SSL crypto engine as default.55— Failed sending network data.56— Failure in receiving network data.58— Problem with the local certificate.59— Couldn't use specified SSL cipher.60— Peer certificate cannot be authenticated with known CA certificates. trust issues, literally.61— Unrecognized transfer encoding.62— Invalid LDAP URL.63— Maximum file size exceeded.64— Requested FTP SSL level failed.65— Sending the data requires a rewind that failed.66— Failed to initialise SSL Engine.67— The user name, password, or similar was not accepted and curl failed to log in. 🙈68— File not found on TFTP server.69— Permission problem on TFTP server.70— Out of disk space on TFTP server. 🍽️71— Illegal TFTP operation.72— Unknown TFTP transfer ID.73— File already exists (TFTP).74— No such user (TFTP).75— Character conversion failed.76— Character conversion functions required.77— Problem with reading the SSL CA cert (path? access rights?).78— The resource referenced in the URL does not exist.79— An unspecified error occurred during the SSH session.80— Failed to shut down the SSL connection.82— Could not load CRL file, missing or wrong format (added in 7.19.0).83— Issuer check failed (added in 7.19.0).84— The FTPPRETcommand failed85— RTSP: mismatch ofCSeqnumbers86— RTSP: mismatch ofSessionIdentifiers87— unable to parse FTP file list88— FTP chunk callback reported error89— No connection available, the session will be queued90— SSL public key does not matched pinned public key91— Invalid SSL certificate status.92— Stream error in HTTP/2 framing layer.XX— More error codes will appear here in future releases. The existing ones are meant to never change. 👑
👑 AUTHORS 👑
Daniel Stenberg is the main author of curl, but the whole list of contributors lives over in the separate THANKS file. give them their flowers 💐
gnurl is based on curl and is maintained by ng0 <ng0@n0.is>. This mdoc man page was derived from the curl automated groff page generation output and is slowly being adjusted to reflect the smaller feature set found in gnurl. we love a work in progress 🎀
🦋 SEE ALSO 🦋
and that's the whole page bestie, we did it 🫶💖 this is the Debian build, dated May 22, 2019, and you were reading GNURL(1) — iconic, honestly. go forth and curl responsibly ✨👑
📦 where this page comes from
- nixos 26.05: libgnurl 7.72.0
- nixpkgs unstable: libgnurl 7.72.0
- Copyright (C) 1998 - 2018, Daniel Stenberg, <daniel@haxx.se>, et al
- 💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms