PHP 5.5.35 Release

curl_setopt

(PHP 4 >= 4.0.2, PHP 5, PHP 7)

curl_setopt设置一个cURL传输选项

说明

bool curl_setopt ( resource $ch , int $option , mixed $value )

为 cURL 会话句柄设置选项。

参数

ch

curl_init() 返回的 cURL 句柄。

option

需要设置的CURLOPT_XXX选项。

value

将设置在option选项上的值。

以下 option 参数的 value应该被设置成 bool 类型:

选项 value 设置为 备注
CURLOPT_AUTOREFERER TRUE 时将根据 Location: 重定向时,自动设置 header 中的Referer:信息。
CURLOPT_BINARYTRANSFER 设为 TRUE ,将在启用 CURLOPT_RETURNTRANSFER 时,返回原生的(Raw)输出。 From PHP 5.1.3, this option has no effect: the raw output will always be returned when CURLOPT_RETURNTRANSFER is used.
CURLOPT_COOKIESESSION 设为 TRUE 时将开启新的一次 cookie 会话。它将强制 libcurl 忽略之前会话时存的其他 cookie。 libcurl 在默认状况下无论是否为会话,都会储存、加载所有 cookie。会话 cookie 是指没有过期时间,只存活在会话之中。
CURLOPT_CERTINFO TRUE 将在安全传输时输出 SSL 证书信息到 STDERR 在 cURL 7.19.1 中添加。 PHP 5.3.2 后有效。 需要开启 CURLOPT_VERBOSE 才有效。
CURLOPT_CONNECT_ONLY TRUE 将让库执行所有需要的代理、验证、连接过程,但不传输数据。此选项用于 HTTP、SMTP 和 POP3。 在 7.15.2 中添加。 PHP 5.5.0 起有效。
CURLOPT_CRLF 启用时将Unix的换行符转换成回车换行符。
CURLOPT_DNS_USE_GLOBAL_CACHE TRUE 会启用一个全局的DNS缓存。此选项非线程安全的,默认已开启。
CURLOPT_FAILONERROR 当 HTTP 状态码大于等于 400,TRUE 将将显示错误详情。 默认情况下将返回页面,忽略 HTTP 代码。
CURLOPT_FILETIME TRUE 时,会尝试获取远程文档中的修改时间信息。 信息可通过curl_getinfo()函数的CURLINFO_FILETIME 选项获取。
CURLOPT_FOLLOWLOCATION TRUE 时将会根据服务器返回 HTTP 头中的 "Location: " 重定向。(注意:这是递归的,"Location: " 发送几次就重定向几次,除非设置了 CURLOPT_MAXREDIRS,限制最大重定向次数。)。
CURLOPT_FORBID_REUSE TRUE 在完成交互以后强制明确的断开连接,不能在连接池中重用。
CURLOPT_FRESH_CONNECT TRUE 强制获取一个新的连接,而不是缓存中的连接。
CURLOPT_FTP_USE_EPRT TRUE 时,当 FTP 下载时,使用 EPRT (和 LPRT)命令。 设置为 FALSE 时禁用 EPRT 和 LPRT,仅仅使用PORT 命令。
CURLOPT_FTP_USE_EPSV TRUE 时,在FTP传输过程中,回到 PASV 模式前,先尝试 EPSV 命令。设置为 FALSE 时禁用 EPSV。
CURLOPT_FTP_CREATE_MISSING_DIRS TRUE 时,当 ftp 操作不存在的目录时将创建它。
CURLOPT_FTPAPPEND TRUE 为追加写入文件,而不是覆盖。
CURLOPT_TCP_NODELAY TRUE 时禁用 TCP 的 Nagle 算法,就是减少网络上的小包数量。 PHP 5.2.1 有效,编译时需要 libcurl 7.11.2 及以上。
CURLOPT_FTPASCII CURLOPT_TRANSFERTEXT 的别名。
CURLOPT_FTPLISTONLY TRUE 时只列出 FTP 目录的名字。
CURLOPT_HEADER 启用时会将头文件的信息作为数据流输出。
CURLINFO_HEADER_OUT TRUE 时追踪句柄的请求字符串。 从 PHP 5.1.3 开始可用。CURLINFO_ 的前缀是有意的(intentional)。
CURLOPT_HTTPGET TRUE 时会设置 HTTP 的 method 为 GET,由于默认是 GET,所以只有 method 被修改时才需要这个选项。
CURLOPT_HTTPPROXYTUNNEL TRUE 会通过指定的 HTTP 代理来传输。
CURLOPT_MUTE TRUE 时将完全静默,无论是何 cURL 函数。 在 cURL 7.15.5 中移出(可以使用 CURLOPT_RETURNTRANSFER 作为代替)
CURLOPT_NETRC TRUE 时,在连接建立时,访问~/.netrc文件获取用户名和密码来连接远程站点。
CURLOPT_NOBODY TRUE 时将不输出 BODY 部分。同时 Mehtod 变成了 HEAD。修改为 FALSE 时不会变成 GET。
CURLOPT_NOPROGRESS

TRUE 时关闭 cURL 的传输进度。

Note:

PHP 默认自动设置此选项为 TRUE,只有为了调试才需要改变设置。

CURLOPT_NOSIGNAL TRUE 时忽略所有的 cURL 传递给 PHP 进行的信号。在 SAPI 多线程传输时此项被默认启用,所以超时选项仍能使用。 cURL 7.10时被加入。
CURLOPT_POST TRUE 时会发送 POST 请求,类型为:application/x-www-form-urlencoded,是 HTML 表单提交时最常见的一种。
CURLOPT_PUT TRUE 时允许 HTTP 发送文件。要被 PUT 的文件必须在 CURLOPT_INFILECURLOPT_INFILESIZE 中设置。
CURLOPT_RETURNTRANSFER TRUEcurl_exec()获取的信息以字符串返回,而不是直接输出。
CURLOPT_SAFE_UPLOAD TRUE 禁用 @ 前缀在 CURLOPT_POSTFIELDS 中发送文件。 意味着 @ 可以在字段中安全得使用了。 可使用 CURLFile 作为上传的代替。 PHP 5.5.0 中添加,默认值 FALSE。 PHP 5.6.0 改默认值为 TRUE
CURLOPT_SSL_VERIFYPEER FALSE 禁止 cURL 验证对等证书(peer's certificate)。要验证的交换证书可以在 CURLOPT_CAINFO 选项中设置,或在 CURLOPT_CAPATH中设置证书目录。 自cURL 7.10开始默认为 TRUE。从 cURL 7.10开始默认绑定安装。
CURLOPT_TRANSFERTEXT TRUE 对 FTP 传输使用 ASCII 模式。对于LDAP,它检索纯文本信息而非 HTML。在 Windows 系统上,系统不会把 STDOUT 设置成二进制 模式。
CURLOPT_UNRESTRICTED_AUTH TRUE 在使用CURLOPT_FOLLOWLOCATION重定向 header 中的多个 location 时继续发送用户名和密码信息,哪怕主机名已改变。
CURLOPT_UPLOAD TRUE 准备上传。
CURLOPT_VERBOSE TRUE 会输出所有的信息,写入到STDERR,或在CURLOPT_STDERR中指定的文件。

以下 optionvalue应该被设置成 integer

选项 设置value 备注
CURLOPT_BUFFERSIZE 每次读入的缓冲的尺寸。当然不保证每次都会完全填满这个尺寸。 在cURL 7.10中被加入。
CURLOPT_CLOSEPOLICY CURLCLOSEPOLICY_* 中的一个。

Note:

此选项已被废弃,它不会被实现,永远不会有效果啦。

PHP 5.6.0 中移除。
CURLOPT_CONNECTTIMEOUT 在尝试连接时等待的秒数。设置为0,则无限等待。
CURLOPT_CONNECTTIMEOUT_MS 尝试连接等待的时间,以毫秒为单位。设置为0,则无限等待。 如果 libcurl 编译时使用系统标准的名称解析器( standard system name resolver),那部分的连接仍旧使用以秒计的超时解决方案,最小超时时间还是一秒钟。 在 cURL 7.16.2 中被加入。从 PHP 5.2.3 开始可用。
CURLOPT_DNS_CACHE_TIMEOUT 设置在内存中缓存 DNS 的时间,默认为120秒(两分钟)。
CURLOPT_FTPSSLAUTH FTP验证方式(启用的时候):CURLFTPAUTH_SSL (首先尝试SSL),CURLFTPAUTH_TLS (首先尝试TLS)或CURLFTPAUTH_DEFAULT (让cURL 自个儿决定)。 在 cURL 7.12.2 中被加入。
CURLOPT_HTTP_VERSION CURL_HTTP_VERSION_NONE (默认值,让 cURL 自己判断使用哪个版本),CURL_HTTP_VERSION_1_0 (强制使用 HTTP/1.0)或CURL_HTTP_VERSION_1_1 (强制使用 HTTP/1.1)。
CURLOPT_HTTPAUTH

使用的 HTTP 验证方法。选项有:CURLAUTH_BASICCURLAUTH_DIGESTCURLAUTH_GSSNEGOTIATECURLAUTH_NTLMCURLAUTH_ANYCURLAUTH_ANYSAFE

可以使用 | 位域(OR)操作符结合多个值,cURL 会让服务器选择受支持的方法,并选择最好的那个。

CURLAUTH_ANYCURLAUTH_BASIC | CURLAUTH_DIGEST | CURLAUTH_GSSNEGOTIATE | CURLAUTH_NTLM 的别名。

CURLAUTH_ANYSAFECURLAUTH_DIGEST | CURLAUTH_GSSNEGOTIATE | CURLAUTH_NTLM 的别名。

CURLOPT_INFILESIZE 希望传给远程站点的文件尺寸,字节(byte)为单位。 注意无法用这个选项阻止 libcurl 发送更多的数据,确切发送什么取决于 CURLOPT_READFUNCTION
CURLOPT_LOW_SPEED_LIMIT 传输速度,每秒字节(bytes)数,根据CURLOPT_LOW_SPEED_TIME秒数统计是否因太慢而取消传输。
CURLOPT_LOW_SPEED_TIME 当传输速度小于CURLOPT_LOW_SPEED_LIMIT时(bytes/sec),PHP会根据CURLOPT_LOW_SPEED_TIME来判断是否因太慢而取消传输。
CURLOPT_MAXCONNECTS 允许的最大连接数量。达到限制时,会通过CURLOPT_CLOSEPOLICY决定应该关闭哪些连接。
CURLOPT_MAXREDIRS 指定最多的 HTTP 重定向次数,这个选项是和CURLOPT_FOLLOWLOCATION一起使用的。
CURLOPT_PORT 用来指定连接端口。
CURLOPT_POSTREDIR 位掩码, 1 (301 永久重定向), 2 (302 Found) 和 4 (303 See Other) 设置 CURLOPT_FOLLOWLOCATION 时,什么情况下需要再次 HTTP POST 到重定向网址。 cURL 7.19.1 中添加,PHP 5.3.2 开始可用。
CURLOPT_PROTOCOLS

CURLPROTO_*的位掩码。 启用时,会限制 libcurl 在传输过程中可使用哪些协议。 这将允许你在编译libcurl时支持众多协议,但是限制只用允许的子集。默认 libcurl 将使用所有支持的协议。 参见CURLOPT_REDIR_PROTOCOLS

可用的协议选项为:CURLPROTO_HTTPCURLPROTO_HTTPSCURLPROTO_FTPCURLPROTO_FTPSCURLPROTO_SCPCURLPROTO_SFTPCURLPROTO_TELNETCURLPROTO_LDAPCURLPROTO_LDAPSCURLPROTO_DICTCURLPROTO_FILECURLPROTO_TFTPCURLPROTO_ALL

在 cURL 7.19.4 中被加入。
CURLOPT_PROXYAUTH HTTP 代理连接的验证方式。使用在CURLOPT_HTTPAUTH中的位掩码。 当前仅仅支持 CURLAUTH_BASICCURLAUTH_NTLM 在 cURL 7.10.7 中被加入。
CURLOPT_PROXYPORT 代理服务器的端口。端口也可以在CURLOPT_PROXY中设置。
CURLOPT_PROXYTYPE 可以是 CURLPROXY_HTTP (默认值) CURLPROXY_SOCKS4CURLPROXY_SOCKS5CURLPROXY_SOCKS4ACURLPROXY_SOCKS5_HOSTNAME 在 cURL 7.10 中被加入。
CURLOPT_REDIR_PROTOCOLS CURLPROTO_* 值的位掩码。如果被启用,位掩码会限制 libcurl 在 CURLOPT_FOLLOWLOCATION开启时,使用的协议。 默认允许除 FILE 和 SCP 外所有协议。 这和 7.19.4 前的版本无条件支持所有支持的协议不同。关于协议常量,请参照CURLOPT_PROTOCOLS 在 cURL 7.19.4 中被加入。
CURLOPT_RESUME_FROM 在恢复传输时,传递字节为单位的偏移量(用来断点续传)。
CURLOPT_SSL_VERIFYHOST 设置为 1 是检查服务器SSL证书中是否存在一个公用名(common name)。译者注:公用名(Common Name)一般来讲就是填写你将要申请SSL证书的域名 (domain)或子域名(sub domain)。 设置成 2,会检查公用名是否存在,并且是否与提供的主机名匹配。 在生产环境中,这个值应该是 2(默认值)。 值 1 的支持在 cURL 7.28.1 中被删除了。
CURLOPT_SSLVERSION CURL_SSLVERSION_DEFAULT (0), CURL_SSLVERSION_TLSv1 (1), CURL_SSLVERSION_SSLv2 (2), CURL_SSLVERSION_SSLv3 (3), CURL_SSLVERSION_TLSv1_0 (4), CURL_SSLVERSION_TLSv1_1 (5) , CURL_SSLVERSION_TLSv1_2 (6) 中的其中一个。

Note:

你最好别设置这个值,让它使用默认值。 设置为 2 或 3 比较危险,在 SSLv2 和 SSLv3 中有弱点存在。

CURLOPT_TIMECONDITION 设置如何对待 CURLOPT_TIMEVALUE。 使用 CURL_TIMECOND_IFMODSINCE,仅在页面 CURLOPT_TIMEVALUE 之后修改,才返回页面。没有修改则返回 "304 Not Modified" 头,假设设置了 CURLOPT_HEADERTRUECURL_TIMECOND_IFUNMODSINCE则起相反的效果。 默认为 CURL_TIMECOND_IFMODSINCE
CURLOPT_TIMEOUT 允许 cURL 函数执行的最长秒数。
CURLOPT_TIMEOUT_MS 设置cURL允许执行的最长毫秒数。 如果 libcurl 编译时使用系统标准的名称解析器( standard system name resolver),那部分的连接仍旧使用以秒计的超时解决方案,最小超时时间还是一秒钟。 在 cURL 7.16.2 中被加入。从 PHP 5.2.3 起可使用。
CURLOPT_TIMEVALUE 秒数,从 1970年1月1日开始。这个时间会被 CURLOPT_TIMECONDITION使。默认使用CURL_TIMECOND_IFMODSINCE
CURLOPT_MAX_RECV_SPEED_LARGE If a download exceeds this speed (counted in bytes per second) on cumulative average during the transfer, the transfer will pause to keep the average rate less than or equal to the parameter value. Defaults to unlimited speed. Added in cURL 7.15.5. Available since PHP 5.4.0.
CURLOPT_MAX_SEND_SPEED_LARGE If an upload exceeds this speed (counted in bytes per second) on cumulative average during the transfer, the transfer will pause to keep the average rate less than or equal to the parameter value. Defaults to unlimited speed. Added in cURL 7.15.5. Available since PHP 5.4.0.
CURLOPT_SSH_AUTH_TYPES A bitmask consisting of one or more of CURLSSH_AUTH_PUBLICKEY, CURLSSH_AUTH_PASSWORD, CURLSSH_AUTH_HOST, CURLSSH_AUTH_KEYBOARD. Set to CURLSSH_AUTH_ANY to let libcurl pick one. Added in cURL 7.16.1.
CURLOPT_IPRESOLVE Allows an application to select what kind of IP addresses to use when resolving host names. This is only interesting when using host names that resolve addresses using more than one version of IP, possible values are CURL_IPRESOLVE_WHATEVER, CURL_IPRESOLVE_V4, CURL_IPRESOLVE_V6, by default CURL_IPRESOLVE_WHATEVER. Added in cURL 7.10.8.

对于下面的这些optionvalue应该被设置成 string

选项 设置的value 备注
CURLOPT_CAINFO 一个保存着1个或多个用来让服务端验证的证书的文件名。这个参数仅仅在和CURLOPT_SSL_VERIFYPEER一起使用时才有意义。 . 可能需要绝对路径。
CURLOPT_CAPATH 一个保存着多个CA证书的目录。这个选项是和CURLOPT_SSL_VERIFYPEER一起使用的。
CURLOPT_COOKIE 设定 HTTP 请求中"Cookie: "部分的内容。多个 cookie 用分号分隔,分号后带一个空格(例如, "fruit=apple; colour=red")。
CURLOPT_COOKIEFILE 包含 cookie 数据的文件名,cookie 文件的格式可以是 Netscape 格式,或者只是纯 HTTP 头部风格,存入文件。如果文件名是空的,不会加载 cookie,但 cookie 的处理仍旧启用。
CURLOPT_COOKIEJAR 连接结束后,比如,调用 curl_close 后,保存 cookie 信息的文件。
CURLOPT_CUSTOMREQUEST

HTTP 请求时,使用自定义的 Method 来代替"GET""HEAD"。对 "DELETE" 或者其他更隐蔽的 HTTP 请求有用。 有效值如 "GET""POST""CONNECT"等等;也就是说,不要在这里输入整行 HTTP 请求。例如输入"GET /index.html HTTP/1.0\r\n\r\n"是不正确的。

Note:

不确定服务器支持这个自定义方法则不要使用它。

CURLOPT_EGDSOCKET 类似CURLOPT_RANDOM_FILE,除了一个Entropy Gathering Daemon套接字。
CURLOPT_ENCODING HTTP请求头中"Accept-Encoding: "的值。 这使得能够解码响应的内容。 支持的编码有"identity""deflate""gzip"。如果为空字符串"",会发送所有支持的编码类型。 在 cURL 7.10 中被加入。
CURLOPT_FTPPORT 这个值将被用来获取供FTP"PORT"指令所需要的IP地址。 "PORT" 指令告诉远程服务器连接到我们指定的IP地址。这个字符串可以是纯文本的IP地址、主机名、一个网络接口名(UNIX下)或者只是一个'-'来使用默认的 IP 地址。
CURLOPT_INTERFACE 发送的网络接口(interface),可以是一个接口名、IP 地址或者是一个主机名。
CURLOPT_KEYPASSWD 使用 CURLOPT_SSLKEYCURLOPT_SSH_PRIVATE_KEYFILE 私钥时候的密码。 在 cURL 7.16.1 中添加。
CURLOPT_KRB4LEVEL KRB4 (Kerberos 4) 安全级别。下面的任何值都是有效的(从低到高的顺序):"clear""safe""confidential""private".。如果字符串以上这些,将使用"private"。 这个选项设置为 NULL 时将禁用 KRB4 安全认证。目前 KRB4 安全认证只能用于 FTP 传输。
CURLOPT_POSTFIELDS 全部数据使用HTTP协议中的 "POST" 操作来发送。 要发送文件,在文件名前面加上@前缀并使用完整路径。 文件类型可在文件名后以 ';type=mimetype' 的格式指定。 这个参数可以是 urlencoded 后的字符串,类似'para1=val1&para2=val2&...',也可以使用一个以字段名为键值,字段数据为值的数组。 如果value是一个数组,Content-Type头将会被设置成multipart/form-data 从 PHP 5.2.0 开始,使用 @ 前缀传递文件时,value 必须是个数组。 从 PHP 5.5.0 开始, @ 前缀已被废弃,文件可通过 CURLFile 发送。 设置 CURLOPT_SAFE_UPLOADTRUE 可禁用 @ 前缀发送文件,以增加安全性。
CURLOPT_PROXY HTTP 代理通道。
CURLOPT_PROXYUSERPWD 一个用来连接到代理的"[username]:[password]"格式的字符串。
CURLOPT_RANDOM_FILE 一个被用来生成 SSL 随机数种子的文件名。
CURLOPT_RANGE "X-Y"的形式,其中X和Y都是可选项获取数据的范围,以字节计。HTTP传输线程也支持几个这样的重复项中间用逗号分隔如"X-Y,N-M"
CURLOPT_REFERER 在HTTP请求头中"Referer: "的内容。
CURLOPT_SSH_HOST_PUBLIC_KEY_MD5 包含 32 位长的 16 进制数值。这个字符串应该是远程主机公钥(public key) 的 MD5 校验值。在不匹配的时候 libcurl 会拒绝连接。 此选项仅用于 SCP 和 SFTP 的传输。 cURL 7.17.1 中添加。
CURLOPT_SSH_PUBLIC_KEYFILE The file name for your public key. If not used, libcurl defaults to $HOME/.ssh/id_dsa.pub if the HOME environment variable is set, and just "id_dsa.pub" in the current directory if HOME is not set. Added in cURL 7.16.1.
CURLOPT_SSH_PRIVATE_KEYFILE The file name for your private key. If not used, libcurl defaults to $HOME/.ssh/id_dsa if the HOME environment variable is set, and just "id_dsa" in the current directory if HOME is not set. If the file is password-protected, set the password with CURLOPT_KEYPASSWD. Added in cURL 7.16.1.
CURLOPT_SSL_CIPHER_LIST 一个SSL的加密算法列表。例如RC4-SHATLSv1都是可用的加密列表。
CURLOPT_SSLCERT 一个包含 PEM 格式证书的文件名。
CURLOPT_SSLCERTPASSWD 使用CURLOPT_SSLCERT证书需要的密码。
CURLOPT_SSLCERTTYPE 证书的类型。支持的格式有"PEM" (默认值), "DER""ENG" 在 cURL 7.9.3中 被加入。
CURLOPT_SSLENGINE 用来在CURLOPT_SSLKEY中指定的SSL私钥的加密引擎变量。
CURLOPT_SSLENGINE_DEFAULT 用来做非对称加密操作的变量。
CURLOPT_SSLKEY 包含 SSL 私钥的文件名。
CURLOPT_SSLKEYPASSWD

CURLOPT_SSLKEY中指定了的SSL私钥的密码。

Note:

由于这个选项包含了敏感的密码信息,记得保证这个PHP脚本的安全。

CURLOPT_SSLKEYTYPE CURLOPT_SSLKEY中规定的私钥的加密类型,支持的密钥类型为"PEM"(默认值)、"DER""ENG"
CURLOPT_URL 需要获取的 URL 地址,也可以在curl_init() 初始化会话的时候。
CURLOPT_USERAGENT 在HTTP请求中包含一个"User-Agent: "头的字符串。
CURLOPT_USERPWD 传递一个连接中需要的用户名和密码,格式为:"[username]:[password]"

以下optionvalue应该被设置成数组:

选项 可选value 备注
CURLOPT_HTTP200ALIASES HTTP 200 响应码数组,数组中的响应码被认为是正确的响应,而非错误。 在 cURL 7.10.3 中被加入。
CURLOPT_HTTPHEADER 设置 HTTP 头字段的数组。格式: array('Content-type: text/plain', 'Content-length: 100')
CURLOPT_POSTQUOTE 在 FTP 请求执行完成后,在服务器上执行的一组array格式的 FTP 命令。
CURLOPT_QUOTE 一组先于 FTP 请求的在服务器上执行的FTP命令。

以下 optionvalue应该被设置成流资源 (例如使用fopen()):

选项 可选value
CURLOPT_FILE 设置输出文件,默认为STDOUT (浏览器)。
CURLOPT_INFILE 上传文件时需要读取的文件。
CURLOPT_STDERR 错误输出的地址,取代默认的STDERR
CURLOPT_WRITEHEADER 设置 header 部分内容的写入的文件地址。

以下optionvalue应该是有效的函数或者闭包:

选项 value
CURLOPT_HEADERFUNCTION 设置一个回调函数,这个函数有两个参数,第一个是cURL的资源句柄,第二个是输出的 header 数据。header数据的输出必须依赖这个函数,返回已写入的数据大小。
CURLOPT_PASSWDFUNCTION 设置一个回调函数,有三个参数,第一个是cURL的资源句柄,第二个是一个密码提示符,第三个参数是密码长度允许的最大值。返回密码的值。
CURLOPT_PROGRESSFUNCTION

设置一个回调函数,有五个参数,第一个是cURL的资源句柄,第二个是预计要下载的总字节(bytes)数。第三个是目前下载的字节数,第四个是预计传输中总上传字节数,第五个是目前上传的字节数。

Note:

只有设置 CURLOPT_NOPROGRESS 选项为 FALSE 时才会调用这个回调函数。

返回非零值将中断传输。 传输将设置 CURLE_ABORTED_BY_CALLBACK 错误。

CURLOPT_READFUNCTION 回调函数名。该函数应接受三个参数。第一个是 cURL resource;第二个是通过选项 CURLOPT_INFILE 传给 cURL 的 stream resource;第三个参数是最大可以读取的数据的数量。回 调函数必须返回一个字符串,长度小于或等于请求的数据量(第三个参数)。一般从传入的 stream resource 读取。返回空字符串作为 EOF(文件结束) 信号。
CURLOPT_WRITEFUNCTION 回调函数名。该函数应接受两个参数。第一个是 cURL resource;第二个是要写入的数据字符串。数 据必须在函数中被保存。 函数必须准确返回写入数据的字节数,否则传输会被一个错误所中 断。

其他值:

Option 设置 value
CURLOPT_SHARE curl_share_init() 返回的结果。 使 cURL 可以处理共享句柄里的数据。

返回值

成功时返回 TRUE, 或者在失败时返回 FALSE

更新日志

版本 说明
5.6.0 默认 CURLOPT_SAFE_UPLOADTRUE
5.6.0 移出 CURLOPT_CLOSEPOLICY和相关的值。
5.5.0 添加 cURL 资源作为 CURLOPT_PROGRESSFUNCTION 回调函数的第一个参数。
5.5.0 引入 CURLOPT_SHARE
5.3.0 引入 CURLOPT_PROGRESSFUNCTION
5.2.10 引入 CURLOPT_PROTOCOLS, and CURLOPT_REDIR_PROTOCOLS.
5.1.0 引入 CURLOPT_AUTOREFERER, CURLOPT_BINARYTRANSFER, CURLOPT_FTPSSLAUTH, CURLOPT_PROXYAUTH, and CURLOPT_TIMECONDITION.
5.0.0 引入 CURLOPT_FTP_USE_EPRT, CURLOPT_NOSIGNAL, CURLOPT_UNRESTRICTED_AUTH, CURLOPT_BUFFERSIZE, CURLOPT_HTTPAUTH, CURLOPT_PROXYPORT, CURLOPT_PROXYTYPE, CURLOPT_SSLCERTTYPE, and CURLOPT_HTTP200ALIASES.

范例

Example #1 初始化一个新的cURL会话并获取一个网页

<?php
// 创建一个新cURL资源
$ch curl_init();

// 设置URL和相应的选项
curl_setopt($chCURLOPT_URL"http://www.example.com/");
curl_setopt($chCURLOPT_HEADERfalse);

// 抓取URL并把它传递给浏览器
curl_exec($ch);

//关闭cURL资源,并且释放系统资源
curl_close($ch);
?>

Example #2 上传文件 (PHP 5.5.0 后被废弃)

<?php

/* http://localhost/upload.php:
print_r($_POST);
print_r($_FILES);
*/

$ch curl_init();

$data = array('name' => 'Foo''file' => '@/home/user/test.png');

curl_setopt($chCURLOPT_URL'http://localhost/upload.php');
curl_setopt($chCURLOPT_POST1);
curl_setopt($chCURLOPT_SAFE_UPLOADfalse); //  PHP 5.6.0 后必须开启
curl_setopt($chCURLOPT_POSTFIELDS$data);

curl_exec($ch);
?>

以上例程会输出:

Array
(
    [name] => Foo
)
Array
(
    [file] => Array
        (
            [name] => test.png
            [type] => image/png
            [tmp_name] => /tmp/phpcpjNeQ
            [error] => 0
            [size] => 279
        )

)

注释

Note:

传递一个数组到CURLOPT_POSTFIELDS,cURL会把数据编码成 multipart/form-data,而然传递一个URL-encoded字符串时,数据会被编码成 application/x-www-form-urlencoded

参见

add a note add a note

User Contributed Notes 139 notes

up
106
rmckay at webaware dot com dot au
3 years ago
Please everyone, stop setting CURLOPT_SSL_VERIFYPEER to false or 0. If your PHP installation doesn't have an up-to-date CA root certificate bundle, download the one at the curl website and save it on your server:

http://curl.haxx.se/docs/caextract.html

Then set a path to it in your php.ini file, e.g. on Windows:

curl.cainfo=c:\php\cacert.pem

Turning off CURLOPT_SSL_VERIFYPEER allows man in the middle (MITM) attacks, which you don't want!
up
6
luca dot manzo at bbsitalia dot com
10 years ago
If you're getting trouble with cookie handling in curl:

- curl manages tranparently cookies in a single curl session
- the option
<?php curl_setopt($ch, CURLOPT_COOKIEJAR, "/tmp/cookieFileName"); ?>

makes curl to store the cookies in a file at the and of the curl session

- the option
<?php curl_setopt($ch, CURLOPT_COOKIEFILE, "/tmp/cookieFileName"); ?>

makes curl to use the given file as source for the cookies to send to the server.

so to handle correctly cookies between different curl session, the you have to do something like this:

<?php
       $ch
= curl_init();
      
curl_setopt ($ch, CURLOPT_URL, $url);
      
curl_setopt ($ch, CURLOPT_COOKIEJAR, COOKIE_FILE_PATH);
      
curl_setopt ($ch, CURLOPT_COOKIEFILE, COOKIE_FILE_PATH);

      
curl_setopt ($ch, CURLOPT_RETURNTRANSFER, 1);
      
$result = curl_exec ($ch);
      
curl_close($ch);
       return
$result;
?>

in particular this is NECESSARY if you are using PEAR_SOAP libraries to build a webservice client over https and the remote server need to establish a session cookie. in fact each soap message is sent using a different curl session!!

I hope this can help someone
Luca
up
6
ron at ttvavanti dot nl
11 years ago
If you specify a CAINFO, note that the file must be in PEM format! (If not, it won't work).
Using Openssl you can use:
openssl x509 -in <cert> -inform d -outform PEM -out cert.pem
To create a pem formatted certificate from a binary certificate (the one you get if you download the ca somewhere).
up
11
Philippe dot Jausions at 11abacus dot com
9 years ago
Clarification on the callback methods:

- CURLOPT_HEADERFUNCTION is for handling header lines received *in the response*,
- CURLOPT_WRITEFUNCTION is for handling data received *from the response*,
- CURLOPT_READFUNCTION is for handling data passed along *in the request*.

The callback "string" can be any callable function, that includes the array(&$obj, 'someMethodName') format.

-Philippe
up
23
jade dot skaggs at gmail dot com
8 years ago
After much struggling, I managed to get a SOAP request requiring HTTP authentication to work.  Here's some source that will hopefully be useful to others.

         <?php

         $credentials
= "username:password";
        
        
// Read the XML to send to the Web Service
        
$request_file = "./SampleRequest.xml";
       
$fh = fopen($request_file, 'r');
       
$xml_data = fread($fh, filesize($request_file));
       
fclose($fh);
               
       
$url = "http://www.example.com/services/calculation";
       
$page = "/services/calculation";
       
$headers = array(
           
"POST ".$page." HTTP/1.0",
           
"Content-type: text/xml;charset=\"utf-8\"",
           
"Accept: text/xml",
           
"Cache-Control: no-cache",
           
"Pragma: no-cache",
           
"SOAPAction: \"run\"",
           
"Content-length: ".strlen($xml_data),
           
"Authorization: Basic " . base64_encode($credentials)
        );
      
       
$ch = curl_init();
       
curl_setopt($ch, CURLOPT_URL,$url);
       
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
       
curl_setopt($ch, CURLOPT_TIMEOUT, 60);
       
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
       
curl_setopt($ch, CURLOPT_USERAGENT, $defined_vars['HTTP_USER_AGENT']);
       
       
// Apply the XML to our curl call
       
curl_setopt($ch, CURLOPT_POST, 1);
       
curl_setopt($ch, CURLOPT_POSTFIELDS, $xml_data);

       
$data = curl_exec($ch);

        if (
curl_errno($ch)) {
            print
"Error: " . curl_error($ch);
        } else {
           
// Show me the result
           
var_dump($data);
           
curl_close($ch);
        }

?>
up
10
dweingart at pobox dot com
13 years ago
If you want to Curl to follow redirects and you would also like Curl to echo back any cookies that are set in the process, use this:

<?php curl_setopt($ch, CURLOPT_COOKIEJAR, '-'); ?>

'-' means stdout

-dw
up
17
joeterranova at gmail dot com
5 years ago
It appears that setting CURLOPT_FILE before setting CURLOPT_RETURNTRANSFER doesn't work, presumably because CURLOPT_FILE depends on CURLOPT_RETURNTRANSFER being set.

So do this:

<?php
curl_setopt
($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_FILE, $fp);
?>

not this:

<?php
curl_setopt
($ch, CURLOPT_FILE, $fp);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
?>
up
11
badman
2 years ago
Many hosters use PHP safe_mode or/and open_basedir, so you can't use CURLOPT_FOLLOWLOCATION. If you try, you see message like this:
CURLOPT_FOLLOWLOCATION cannot be activated when safe_mode is enabled or an open_basedir is set in [you script name & path] on line XXX

First, I try to use zsalab function (http://us2.php.net/manual/en/function.curl-setopt.php#102121) from this page, but for some reason it did not work properly. So, I wrote my own.

It can be use instead of curl_exec. If server HTTP response codes is 30x, function will forward the request as long as the response is not different from 30x (for example, 200 Ok). Also you can use POST.

function curlExec(/* Array */$curlOptions='', /* Array */$curlHeaders='', /* Array */$postFields='')
{
  $newUrl = '';
  $maxRedirection = 10;
  do
  {
    if ($maxRedirection<1) die('Error: reached the limit of redirections');

    $ch = curl_init();
    if (!empty($curlOptions)) curl_setopt_array($ch, $curlOptions);
    if (!empty($curlHeaders)) curl_setopt($ch, CURLOPT_HTTPHEADER, $curlHeaders);
    if (!empty($postFields))
    {
      curl_setopt($ch, CURLOPT_POST, 1);
      curl_setopt($ch, CURLOPT_POSTFIELDS, $postFields);
    }
   
    if (!empty($newUrl)) curl_setopt($ch, CURLOPT_URL, $newUrl); // redirect needed
   
    $curlResult = curl_exec($ch);
    $code = curl_getinfo($ch, CURLINFO_HTTP_CODE);

    if ($code == 301 || $code == 302 || $code == 303 || $code == 307)
    {
      preg_match('/Location:(.*?)\n/', $curlResult, $matches);
      $newUrl = trim(array_pop($matches));
      curl_close($ch);

      $maxRedirection--;
      continue;
    }
    else // no more redirection
    {
      $code = 0;
      curl_close($ch);
    }
  }
  while($code);
  return $curlResult;
}
up
7
anderseta at gmail dot com
6 years ago
If you wish to find the size of the file you are streaming and use it as your header this is how:

<?php

function write_function($curl_resource, $string)
{
    if(
curl_getinfo($curl_resource, CURLINFO_SIZE_DOWNLOAD) <= 2000)
    {
       
header('Expires: 0');
       
header('Cache-Control: must-revalidate, post-check=0, pre-check=0');
       
header('Pragma: public');
       
header('Content-Description: File Transfer');
       
header("Content-Transfer-Encoding: binary");
       
header("Content-Type: ".curl_getinfo($curl_resource, CURLINFO_CONTENT_TYPE)."");
       
header("Content-Length: ".curl_getinfo($curl_resource, CURLINFO_CONTENT_LENGTH_DOWNLOAD)."");
    }
   
    print
$string;

    return
mb_strlen($string, '8bit');
}

?>

1440 is the the default number of bytes curl will call the write function (BUFFERSIZE does not affect this, i actually think you can not change this value), so it means the headers are going to be set only one time.

write_function must return the exact number of bytes of the string, so you can return a value with mb_strlen.
up
16
Ed Cradock
6 years ago
PUT requests are very simple, just make sure to specify a content-length header and set post fields as a string.

Example:

<?php
function doPut($url, $fields)
{
  
$fields = (is_array($fields)) ? http_build_query($fields) : $fields;

   if(
$ch = curl_init($url))
   {
     
curl_setopt($ch, CURLOPT_CUSTOMREQUEST, 'PUT');
     
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
     
curl_setopt($ch, CURLOPT_HTTPHEADER, array('Content-Length: ' . strlen($fields)));
     
curl_setopt($ch, CURLOPT_POSTFIELDS, $fields);
     
curl_exec($ch);

     
$status = curl_getinfo($ch, CURLINFO_HTTP_CODE);

     
curl_close($ch);

      return (int)
$status;
   }
   else
   {
      return
false;
   }
}

if(
doPut('http://example.com/api/a/b/c', array('foo' => 'bar')) == 200)
  
// do something
else
  
// do something else.
?>

You can grab the request data on the other side with:

<?php
if($_SERVER['REQUEST_METHOD'] == 'PUT')
{
  
parse_str(file_get_contents('php://input'), $requestData);

  
// Array ( [foo] => bar )
  
print_r($requestData);

  
// Do something with data...
}
?>

DELETE  can be done in exactly the same way.
up
6
Dustin Hawkins
10 years ago
To further expand upon use of CURLOPT_CAPATH and CURLOPT_CAINFO...

In my case I wanted to prevent curl from talking to any HTTPS server except my own using a self signed certificate. To do this, you'll need openssl installed and access to the HTTPS Server Certificate (server.crt by default on apache)

You can then use a command simiar to this to translate your apache certificate into one that curl likes.

$ openssl x509 -in server.crt -out outcert.pem -text

Then set CURLOPT_CAINFO equal to the the full path to outcert.pem and turn on CURLOPT_SSL_VERIFYPEER.

If you want to use the CURLOPT_CAPATH option, you should create a directory for all the valid certificates you have created, then use the c_rehash script that is included with openssl to "prepare" the directory.

If you dont use the c_rehash utility, curl will ignore any file in the directory you set.
up
5
eion at bigfoot dot com
9 years ago
If you are trying to use CURLOPT_FOLLOWLOCATION and you get this warning:
Warning: curl_setopt() [function.curl-setopt]: CURLOPT_FOLLOWLOCATION cannot be activated when in safe_mode or an open_basedir is set...

then you will want to read http://www.php.net/ChangeLog-4.php which says "Disabled CURLOPT_FOLLOWLOCATION in curl when open_basedir or safe_mode are enabled." as of PHP 4.4.4/5.1.5.  This is due to the fact that curl is not part of PHP and doesn't know the values of open_basedir or safe_mode, so you could comprimise your webserver operating in safe_mode by redirecting (using header('Location: ...')) to "file://" urls, which curl would have gladly retrieved.

Until the curl extension is changed in PHP or curl (if it ever will) to deal with "Location:" headers, here is a far from perfect remake of the curl_exec function that I am using.

Since there's no curl_getopt function equivalent, you'll have to tweak the function to make it work for your specific use.  As it is here, it returns the body of the response and not the header.  It also doesn't deal with redirection urls with username and passwords in them.

<?php
   
function curl_redir_exec($ch)
    {
        static
$curl_loops = 0;
        static
$curl_max_loops = 20;
        if (
$curl_loops++ >= $curl_max_loops)
        {
           
$curl_loops = 0;
            return
FALSE;
        }
       
curl_setopt($ch, CURLOPT_HEADER, true);
       
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
       
$data = curl_exec($ch);
        list(
$header, $data) = explode("\n\n", $data, 2);
       
$http_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
        if (
$http_code == 301 || $http_code == 302)
        {
           
$matches = array();
           
preg_match('/Location:(.*?)\n/', $header, $matches);
           
$url = @parse_url(trim(array_pop($matches)));
            if (!
$url)
            {
               
//couldn't process the url to redirect to
               
$curl_loops = 0;
                return
$data;
            }
           
$last_url = parse_url(curl_getinfo($ch, CURLINFO_EFFECTIVE_URL));
            if (!
$url['scheme'])
               
$url['scheme'] = $last_url['scheme'];
            if (!
$url['host'])
               
$url['host'] = $last_url['host'];
            if (!
$url['path'])
               
$url['path'] = $last_url['path'];
           
$new_url = $url['scheme'] . '://' . $url['host'] . $url['path'] . ($url['query']?'?'.$url['query']:'');
           
curl_setopt($ch, CURLOPT_URL, $new_url);
           
debug('Redirecting to', $new_url);
            return
curl_redir_exec($ch);
        } else {
           
$curl_loops=0;
            return
$data;
        }
    }
?>
up
4
JScott jscott401 at gmail dot com
5 years ago
Some additional notes for curlopt_writefunction. I struggled with this at first because it really isn't documented very well.

When you write a callback function and use it with curlopt_writefunction it will be called MULTIPLE times. Your function MUST return the ammount of data written to it each time. It is very picky about this. Here is a snippet from my code that may help you

<?php
curl_setopt
($this->curl_handle, CURLOPT_WRITEFUNCTION, array($this, "receiveResponse"));

// later on in the class I wrote my receive Response method

private function receiveResponse($curlHandle,$xmldata)
                {
                       
$this->responseString = $xmldata;
                       
$this->responseXML .=  $this->responseString;
                       
$this->length = strlen($xmldata);
                       
$this->size += $this->length;
                        return
$this->length;

                }
?>

Now I did this for a class. If you aren't doing OOP then you will obviously need to modify this for your own use.

CURL calls your script MULTIPLE times because the data will not always be sent all at once. Were talking internet here so its broken up into packets. You need to take your data and concatenate it all together until it is all written. I was about to pull my damn hair out because I would get broken chunks of XML back from the server and at random lengths. I finally figured out what was going on. Hope this helps
up
7
fnjordy at gmail dot com
7 years ago
Note that CURLOPT_RETURNTRANSFER when used with CURLOPT_WRITEFUNCTION has effectively three settings: default, true, and false.

default - callbacks will be called as expected.
true - content will be returned but callback function will not be called.
false - content will be output and callback function will not be called.

Note that CURLOPT_HEADERFUNCTION callbacks are always called.
up
3
jancister at gmail dot com
1 year ago
Please note that if you want to handle progress using CURLOPT_PROGRESSFUNCTION option, you need to take into consideration what version of PHP are you using. Since version 5.5.0, compatibility-breaking change was introduced in number/order of the arguments passed to the callback function, and cURL resource is now passed as first argument.

Prior to version 5.5.0:
<?php
// ...
curl_setopt($resource, CURLOPT_PROGRESSFUNCTION, 'progressCallback');
curl_setopt($resource, CURLOPT_NOPROGRESS, false);
// ...
function progressCallback($download_size = 0, $downloaded = 0, $upload_size = 0, $uploaded = 0)
{
   
// Handle progress
}
?>

From version 5.5.0:
<?php
// ...
curl_setopt($resource, CURLOPT_PROGRESSFUNCTION, 'progressCallback');
curl_setopt($resource, CURLOPT_NOPROGRESS, false);
// ...
function progressCallback($resource, $download_size = 0, $downloaded = 0, $upload_size = 0, $uploaded = 0)
{
   
// Handle progress
}
?>

However, if your code needs to be compatible with PHP version both before and after 5.5.0, consider adding a version check:
<?php
// ...
curl_setopt($resource, CURLOPT_PROGRESSFUNCTION, 'progressCallback');
curl_setopt($resource, CURLOPT_NOPROGRESS, false);
// ...
function progressCallback($resource, $download_size = 0, $downloaded = 0, $upload_size = 0, $uploaded = 0)
{
  
/**
    * $resource parameter was added in version 5.5.0 breaking backwards compatibility;
    * if we are using PHP version lower than 5.5.0, we need to shift the arguments
    * @see http://php.net/manual/en/function.curl-setopt.php#refsect1-function.curl-setopt-changelog
    */
   
if (version_compare(PHP_VERSION, '5.5.0') < 0) {
       
$uploaded = $upload_size;
       
$upload_size = $downloaded;
       
$downloaded = $download_size;
       
$download_size = $resource;
    }

   
// Handle progress
}
?>
up
18
Steve Kamerman
4 years ago
If you want cURL to timeout in less than one second, you can use CURLOPT_TIMEOUT_MS, although there is a bug/"feature"  on "Unix-like systems" that causes libcurl to timeout immediately if the value is < 1000 ms with the error "cURL Error (28): Timeout was reached".  The explanation for this behavior is:

"If libcurl is built to use the standard system name resolver, that portion of the transfer will still use full-second resolution for timeouts with a minimum timeout allowed of one second."

What this means to PHP developers is "You can use this function without testing it first, because you can't tell if libcurl is using the standard system name resolver (but you can be pretty sure it is)"

The problem is that on (Li|U)nix, when libcurl uses the standard name resolver, a SIGALRM is raised during name resolution which libcurl thinks is the timeout alarm.

The solution is to disable signals using CURLOPT_NOSIGNAL.  Here's an example script that requests itself causing a 10-second delay so you can test timeouts:

<?php
if (!isset($_GET['foo'])) {
       
// Client
       
$ch = curl_init('http://localhost/test/test_timeout.php?foo=bar');
       
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
       
curl_setopt($ch, CURLOPT_NOSIGNAL, 1);
       
curl_setopt($ch, CURLOPT_TIMEOUT_MS, 200);
       
$data = curl_exec($ch);
       
$curl_errno = curl_errno($ch);
       
$curl_error = curl_error($ch);
       
curl_close($ch);

        if (
$curl_errno > 0) {
                echo
"cURL Error ($curl_errno): $curl_error\n";
        } else {
                echo
"Data received: $data\n";
        }
} else {
       
// Server
       
sleep(10);
        echo
"Done.";
}
?>
up
6
c00lways at gmail dot com
8 years ago
if you would like to send xml request to a server (lets say, making a soap proxy),
you have to set

<?php
curl_setopt
($ch, CURLOPT_HTTPHEADER, Array("Content-Type: text/xml"));
?>

makesure you watch for cache issue:
the below code will prevent cache...

<?php
curl_setopt
($ch, CURLOPT_FORBID_REUSE, 1);
curl_setopt($ch, CURLOPT_FRESH_CONNECT, 1);
?>

hope it helps ;)
up
3
rob at infoglobe dot net
9 years ago
Options not included in the above, but that work (Taken from the libcurl.a C documentation)

CURLOPT_FTP_SSL

Pass a long using one of the values from below, to make libcurl use your desired level of SSL for the ftp transfer. (Added in 7.11.0)

CURLFTPSSL_NONE

Don't attempt to use SSL.

CURLFTPSSL_TRY

Try using SSL, proceed as normal otherwise.

CURLFTPSSL_CONTROL

Require SSL for the control connection or fail with CURLE_FTP_SSL_FAILED.

CURLFTPSSL_ALL

Require SSL for all communication or fail with CURLE_FTP_SSL_FAILED.
up
5
m at mar dot lt
3 years ago
Be careful when changing CURLOPT_SSL_VERIFYHOST or other options to true (boolean). It may cause insecure behavior [1]

This is because boolean true casts into integer 1, and CURLOPT_SSL_VERIFYHOST = 1 is not secure behavior.

The *correct* value here is CURLOPT_SSL_VERIFYHOST = 2. By setting this value equal to 1 the peer certificate must contain a Common Name field, but it doesn't matter what name it says.

[1] Martin Georgiev and Subodh Iyengar and Suman Jana and Rishita Anubhai and Dan Boneh and Vitaly Shmatikov, The most dangerous code in the world: validating SSL certificates in non-browser software, ACM CCS '12, pp. 38-49, 2012
up
7
regan dot corey at gmail dot com
3 years ago
I spent a couple of days trying to POST a multi-dimensional array of form fields, including a file upload, to a remote server to update a product. Here are the breakthroughs that FINALLY allowed the script to run as desired.

Firstly, the HTML form used input names like these:
<input type="text" name="product[name]" />
<input type="text" name="product[cost]" />
<input type="file" name="product[thumbnail]" />
in conjunction with two other form inputs not part of the product array
<input type="text" name="method" value="put" />
<input type="text" name="mode" />

I used several cURL options, but the only two (other than URL) that mattered were:
curl_setopt($handle, CURLOPT_POST, true);
curl_setopt($handle, CURLOPT_POSTFIELDS, $postfields);
Pretty standard so far.
Note: headers didn't need to be set, cURL automatically sets headers (like content-type: multipart/form-data; content-length...) when you pass an array into CURLOPT_POSTFIELDS.
Note: even though this is supposed to be a PUT command through an HTTP POST form, no special PUT options needed to be passed natively through cURL. Options such as
curl_setopt($handle, CURLOPT_HTTPHEADER, array('X-HTTP-Method-Override: PUT', 'Content-Length: ' . strlen($fields)));
or
curl_setopt($handle, CURLOPT_PUT, true);
or
curl_setopt($handle, CURLOPT_CUSTOMREQUEST, "PUT);
were not needed to make the code work.

The fields I wanted to pass through cURL were arranged into an array something like this:
$postfields = array("method" => $_POST["method"],
                    "mode" => $_POST["mode"],
                    "product" => array("name" => $_POST["product"],
                                        "cost" => $_POST["product"]["cost"],
                                        "thumbnail" => "@{$_FILES["thumbnail"]["tmp_name"]};type={$_FILES["thumbnail"]["type"]}")
                    );

-Notice how the @ precedes the temporary filename, this creates a link so PHP will upload/transfer an actual file instead of just the file name, which would happen if the @ isn't included.
-Notice how I forcefully set the mime-type of the file to upload. I was having issues where images filetypes were defaulting to octet-stream instead of image/png or image/jpeg or whatever the type of the selected image.

I then tried passing $postfields straight into curl_setopt($this->handle, CURLOPT_POSTFIELDS, $postfields); but it didn't work.
I tried using http_build_query($postfields); but that didn't work properly either.
In both cases either the file wouldn't be treated as an actual file and the form data wasn't being sent properly. The problem was HTTP's methods of transmitting arrays. While PHP and other languages can figure out how to handle arrays passed via forms, HTTP isn't quite as sofisticated. I had to rewrite the $postfields array like so:
$postfields = array("method" => $_POST["method"],
                    "mode" => $_POST["mode"],
                    "product[name]" => $_POST["product"],
                    "product[cost]" => $_POST["product"]["cost"],
                    "product[thumbnail]" => "@{$_FILES["thumbnail"]["tmp_name"]}");
curl_setopt($handle, CURLOPT_POSTFIELDS, $postfields);

This, without the use of http_build_query, solved all of my problems. Now the receiving host outputs both $_POST and $_FILES vars correctly.
up
7
PHP at RHaworth dot net
5 years ago
When CURLOPT_FOLLOWLOCATION and CURLOPT_HEADER are both true and redirect/s have happened then the header returned by curl_exec() will contain all the headers in the redirect chain in the order they were encountered.
up
14
Richard
1 year ago
CURL_SSLVERSION_TLSv1_0 (4), CURL_SSLVERSION_TLSv1_1 (5) or CURL_SSLVERSION_TLSv1_2 (6) only work for PHP versions using curl 7.34 or newer.
up
9
sgamon at yahoo dot com
8 years ago
If you are doing a POST, and the content length is 1,025 or greater, then curl exploits a feature of http 1.1: 100 (Continue) Status.

See http://www.w3.org/Protocols/rfc2616/rfc2616-sec8.html#sec8.2.3

* it adds a header, "Expect: 100-continue". 
* it then sends the request head, waits for a 100 response code, then sends the content

Not all web servers support this though.  Various errors are returned depending on the server.  If this happens to you, suppress the "Expect" header with this command:

<?php
curl_setopt
($ch, CURLOPT_HTTPHEADER, array('Expect:'));
?>

See http://www.gnegg.ch/2007/02/the-return-of-except-100-continue/
up
7
yann dot corno at free dot fr
13 years ago
About the CURLOPT_HTTPHEADER option, it took me some time to figure out how to format the so-called 'Array'. It fact, it is a list of strings. If Curl was already defining a header item, yours will replace it. Here is an example to change the Content Type in a POST:

<?php curl_setopt ($ch, CURLOPT_HTTPHEADER, Array("Content-Type: text/xml")); ?>

Yann
up
6
mr at coder dot tv
10 years ago
Sometimes you can't use CURLOPT_COOKIEJAR and CURLOPT_COOKIEFILE becoz of the server php-settings(They say u may grab any files from server using these options). Here is the solution
1)Don't use CURLOPT_FOLLOWLOCATION
2)Use curl_setopt($ch, CURLOPT_HEADER, 1)
3)Grab from the header cookies like this:
preg_match_all('|Set-Cookie: (.*);|U', $content, $results);   
$cookies = implode(';', $results[1]);
4)Set them using curl_setopt($ch, CURLOPT_COOKIE,  $cookies);

Good Luck, Yevgen
up
7
Adam Monsen
4 years ago
CURLOPT_POST must be left unset if you want the Content-Type header set to "multipart/form-data" (e.g., when CURLOPT_POSTFIELDS is an array). If you set CURLOPT_POSTFIELDS to an array and have CURLOPT_POST set to TRUE, Content-Length will be -1 and most sane servers will reject the request. If you set CURLOPT_POSTFIELDS to an array and have CURLOPT_POST set to FALSE, cURL will send a GET request.
up
10
saidk at phirebranding dot com
7 years ago
Passing in PHP's $_SESSION into your cURL call:

<?php
session_start
();
$strCookie = 'PHPSESSID=' . $_COOKIE['PHPSESSID'] . '; path=/';
session_write_close();

$curl_handle = curl_init('enter_external_url_here');
curl_setopt( $curl_handle, CURLOPT_COOKIE, $strCookie );
curl_exec($curl_handle);
curl_close($curl_handle);
?>

This worked great for me.  I was calling pages from the same server and needed to keep the $_SESSION variables.  This passes them over.  If you want to test, just print_r($_SESSION);

Enjoy!
up
2
support at lostair dot com
5 years ago
Force Curl Request To Go To A Particular IP Address

Yes, there is a method of passing an IP address to curl.  Excellent for services with multiple IP addresses and also to take DNS out of the equation for testing/debugging.

<?php
   
function fetch_page($url, $host_ip = NULL)
    {

     
$ch = curl_init();

      if (!
is_null($host_ip))
      {
       
$urldata = parse_url($url);

       
//  Ensure we have the query too, if there is any...
       
if (!empty($urldata['query']))
         
$urldata['path'] .= "?".$urldata['query'];

       
//  Specify the host (name) we want to fetch...
       
$headers = array("Host: ".$urldata['host']);
       
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);

       
//  create the connecting url (with the hostname replaced by IP)
       
$url = $urldata['scheme']."://".$host_ip.$urldata['path'];
      }

     
curl_setopt($chCURLOPT_URL, $url);
     
curl_setopt ($ch, CURLOPT_HEADER, 0);
     
curl_setopt($chCURLOPT_RETURNTRANSFER, true);

     
$result = curl_exec ($ch);
     
curl_close ($ch);

      return
$result;
    }
?>
up
5
qeremy [atta] gmail [dotta] com
3 years ago
If you are trying to update something on your server and you need to handle this update operation by PUT;

<?php
curl_setopt
($ch, CURLOPT_CUSTOMREQUEST, "PUT");
curl_setopt($ch, CURLOPT_PUT, 1);
?>

are "useless" without;

<?php
curl_setopt
($ch, CURLOPT_HTTPHEADER, array('X-HTTP-Method-Override: PUT'));
?>

Example;

Updating a book data in database identified by "id 1";

--cURL Part--
<?php
$data
= http_build_query($_POST);
// or
$data = http_build_query(array(
  
'name'  => 'PHP in Action',
  
'price' => 10.9
));

$ch = curl_init();
curl_setopt($ch, CURLOPT_URL, "http://api.localhost/rest/books/1");
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
// curl_setopt($ch, CURLOPT_CUSTOMREQUEST, "PUT"); // no need anymore
// or
// curl_setopt($ch, CURLOPT_PUT, 1); // no need anymore
curl_setopt($ch, CURLOPT_HTTPHEADER, array('X-HTTP-Method-Override: PUT'));
curl_setopt($ch, CURLOPT_POSTFIELDS, $data);
$ce = curl_exec($ch);
curl_close($ch);
print_r($ce);
?>

--API class--
<?php
public function putAction() {
    echo
"putAction() -> id: ". $this->_getParam('id') ."\n";
   
print_r($_POST);
   
// do stuff with post data
   
...
?>

--Output--
putAction() -> id: 15
Array
(
    [name] => PHP in Action
    [price] => 10.9
)

---Keywords--
rest, restfull api, restfull put, curl put, curl customrequest put
up
2
ashw1 - at - no spam - post - dot - cz
8 years ago
In case you wonder how come, that cookies don't work under Windows, I've googled for some answers, and here is the result: Under WIN you need to input absolute path of the cookie file.

This piece of code solves it:

<?php

if ($cookies != '')
    {
    if (
substr(PHP_OS, 0, 3) == 'WIN')
        {
$cookies = str_replace('\\','/', getcwd().'/'.$cookies);}
   
curl_setopt($ch, CURLOPT_COOKIEJAR, $cookies);
   
curl_setopt($ch, CURLOPT_COOKIEFILE, $cookies);
    }

?>
up
5
adrian at foeder dot de
3 years ago
if you want to do a GET request with additional body data it will become tricky not to implicitly change the request to a POST, like many notes below correctly state.
So to do the analogy of command line's

curl -XGET 'http://example.org?foo=bar' -d '<baz>some additional data</baz>'

in PHP you'll do, besides your other necessary stuff,

<?php
    curl_setopt
($curlHandle, CURLOPT_CUSTOMREQUEST, 'GET');
   
curl_setopt($curlHandle, CURLOPT_POSTFIELDS, '<baz>some additional data</baz>');
?>

during my experiments, every other "similar" way, like e.g. CURLOPT_HTTPGET, didn't send the additional data or fell into POST.
up
2
phpnet at wafflehouse dot de
10 years ago
Resetting CURLOPT_FILE to STDOUT won't work by calling curl_setopt() with the STDOUT constant or a php://output stream handle (at least I get error messages when trying the code from phpnet at andywaite dot com). Instead, one can simply reset it as a side effect of CURLOPT_RETURNTRANSFER. Just say

<?php curl_setopt($this->curl,CURLOPT_RETURNTRANSFER,0); ?>

and following calls to curl_exec() will output to STDOUT again.
up
2
ROXORT at TGNOOB dot FR
10 years ago
<?php
/*
  Here is a script that is usefull to :
  - login to a POST form,
  - store a session cookie,
  - download a file once logged in.
*/

// INIT CURL
$ch = curl_init();

// SET URL FOR THE POST FORM LOGIN
curl_setopt($ch, CURLOPT_URL, 'http://www.example.com/Members/Login.php');

// ENABLE HTTP POST
curl_setopt ($ch, CURLOPT_POST, 1);

// SET POST PARAMETERS : FORM VALUES FOR EACH FIELD
curl_setopt ($ch, CURLOPT_POSTFIELDS, 'fieldname1=fieldvalue1&fieldname2=fieldvalue2');

// IMITATE CLASSIC BROWSER'S BEHAVIOUR : HANDLE COOKIES
curl_setopt ($ch, CURLOPT_COOKIEJAR, 'cookie.txt');

# Setting CURLOPT_RETURNTRANSFER variable to 1 will force cURL
# not to print out the results of its query.
# Instead, it will return the results as a string return value
# from curl_exec() instead of the usual true/false.
curl_setopt ($ch, CURLOPT_RETURNTRANSFER, 1);

// EXECUTE 1st REQUEST (FORM LOGIN)
$store = curl_exec ($ch);

// SET FILE TO DOWNLOAD
curl_setopt($ch, CURLOPT_URL, 'http://www.example.com/Members/Downloads/AnnualReport.pdf');

// EXECUTE 2nd REQUEST (FILE DOWNLOAD)
$content = curl_exec ($ch);

// CLOSE CURL
curl_close ($ch);

/*
  At this point you can do do whatever you want
  with the downloaded file stored in $content :
  display it, save it as file, and so on.
*/
?>
up
2
ikendra at yken dot nospam dot org
10 years ago
Using cURL, I needed to call a third-party script which was returning binary data as attachment to pass on retrieved data again as attachment.

Problem was that the third-party script occassionally returned HTTP errors and I wanted to avoid passing on zero-length attachment in such case.

Combination of using CURLOPT_FAILONERROR and CURLOPT_HEADERFUNCTION callback helped to process the third-party script HTTP errors neatly:

<?php
function curlHeaderCallback($resURL, $strHeader) {
    if (
preg_match('/^HTTP/i', $strHeader)) {
       
header($strHeader);
       
header('Content-Disposition: attachment; filename="file-name.zip"');
    }
    return
strlen($strHeader);
}

$strURL = 'http://www.example.com/script-whichs-dumps-binary-attachment.php';

$resURL = curl_init();
curl_setopt($resURL, CURLOPT_URL, $strURL);
curl_setopt($resURL, CURLOPT_BINARYTRANSFER, 1);
curl_setopt($resURL, CURLOPT_HEADERFUNCTION, 'curlHeaderCallback');
curl_setopt($resURL, CURLOPT_FAILONERROR, 1);

curl_exec ($resURL);

$intReturnCode = curl_getinfo($resURL, CURLINFO_HTTP_CODE);
curl_close ($resURL);

if (
$intReturnCode != 200) {
    print
'was error: ' . $intReturnCode;
}
?>
up
7
Madcat
2 years ago
If you have a mixture of strings starting with @ (at character) and files in CURLOPT_POSTFIELDS you have a problem (such as posting a tweet with attached media) because curl tries to interpret anything starting with @ as a file.

<?php

$postfields
= array(
   
'upload_file' => '@file_to_upload.png',
   
'upload_text' => '@text_to_upload'
);

$curl = curl_init();
curl_setopt($curl, CURLOPT_URL, 'http://example.com/upload-test');
curl_setopt($curl, CURLOPT_POSTFIELDS, $postfields);
curl_exec($curl);
curl_close($curl);

?>

To get around this, prepend the text string with the NULL character like so:

<?php
    $postfields
= array(
       
'upload_file' => '@file_to_upload.png',
       
'upload_text' => sprintf("\0%s", '@text_to_upload')
    );
?>

Original source: http://bit.ly/AntMle
up
4
S\
5 years ago
When using CURLOPT_POSTFIELDS with an array as parameter, you have to pay high attention to user input. Unvalidated user input will lead to serious security issues.

<?php

/**
* test.php:
*/
$ch = curl_init('http://example.com');

curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, array(
   
'foo' => $_GET['bar']
));

curl_exec($ch);

?>

Requesting "test.php?bar=@/home/user/test.png" will send "test.png" to example.com.
Make sure you remove the leading "@" from user input.
up
5
dotpointer at gmail dot com
4 years ago
I noted something when using CURLOPT_POSTFIELDS in combination with arrays from PHP.

You may supply an array, but there may not be any sub-arrays in this array, as this will give Array-to-string-conversion notice.

Example:

<?php
$ch
= curl_init();

# this works
$data = array('name' => 'value');

# this gives "Notice: Array to string conversion..."
$data = array('name' => array('subname' => 'subvalue'));

curl_setopt($ch, CURLOPT_URL, 'http://localhost/test.php');
curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, $data);

curl_exec($ch);
?>
up
4
Pawel Antczak
6 years ago
Hello.
During problems with "CURLOPT_FOLLOWLOCATION cannot be activated when in safe_mode or an open_basedir is set"
I was looking for solution.
I've found few methods on this page, but none of them was good enough, so I made one.
<?php
function curl_redirect_exec($ch, &$redirects, $curlopt_header = false) {
   
curl_setopt($ch, CURLOPT_HEADER, true);
   
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
   
$data = curl_exec($ch);
   
$http_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
    if (
$http_code == 301 || $http_code == 302) {
        list(
$header) = explode("\r\n\r\n", $data, 2);
       
$matches = array();
       
preg_match('/(Location:|URI:)(.*?)\n/', $header, $matches);
       
$url = trim(array_pop($matches));
       
$url_parsed = parse_url($url);
        if (isset(
$url_parsed)) {
           
curl_setopt($ch, CURLOPT_URL, $url);
           
$redirects++;
            return
curl_redirect_exec($ch, $redirects);
        }
    }
    if (
$curlopt_header)
        return
$data;
    else {
        list(,
$body) = explode("\r\n\r\n", $data, 2);
        return
$body;
    }
}
?>

Main issue in existing functions was lack of information, how many redirects was done.
This one will count it.
First parameter as usual.
Second should be already initialized integer, it will be incremented by number of done redirects.
You can set CURLOPT_HEADER if You need it.
up
8
zsalab
5 years ago
Handling redirections with curl if safe_mode or open_basedir is enabled. The function working transparent, no problem with header and returntransfer options. You can handle the max redirection with the optional second argument (the function is set the variable to zero if max redirection exceeded).
Second parameter values:
- maxredirect is null or not set: redirect maximum five time, after raise PHP warning
- maxredirect is greather then zero: no raiser error, but parameter variable set to zero
- maxredirect is less or equal zero: no follow redirections

<?php
function curl_exec_follow(/*resource*/ $ch, /*int*/ &$maxredirect = null) {
   
$mr = $maxredirect === null ? 5 : intval($maxredirect);
    if (
ini_get('open_basedir') == '' && ini_get('safe_mode' == 'Off')) {
       
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, $mr > 0);
       
curl_setopt($ch, CURLOPT_MAXREDIRS, $mr);
    } else {
       
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, false);
        if (
$mr > 0) {
           
$newurl = curl_getinfo($ch, CURLINFO_EFFECTIVE_URL);

           
$rch = curl_copy_handle($ch);
           
curl_setopt($rch, CURLOPT_HEADER, true);
           
curl_setopt($rch, CURLOPT_NOBODY, true);
           
curl_setopt($rch, CURLOPT_FORBID_REUSE, false);
           
curl_setopt($rch, CURLOPT_RETURNTRANSFER, true);
            do {
               
curl_setopt($rch, CURLOPT_URL, $newurl);
               
$header = curl_exec($rch);
                if (
curl_errno($rch)) {
                   
$code = 0;
                } else {
                   
$code = curl_getinfo($rch, CURLINFO_HTTP_CODE);
                    if (
$code == 301 || $code == 302) {
                       
preg_match('/Location:(.*?)\n/', $header, $matches);
                       
$newurl = trim(array_pop($matches));
                    } else {
                       
$code = 0;
                    }
                }
            } while (
$code && --$mr);
           
curl_close($rch);
            if (!
$mr) {
                if (
$maxredirect === null) {
                   
trigger_error('Too many redirects. When following redirects, libcurl hit the maximum amount.', E_USER_WARNING);
                } else {
                   
$maxredirect = 0;
                }
                return
false;
            }
           
curl_setopt($ch, CURLOPT_URL, $newurl);
        }
    }
    return
curl_exec($ch);
}
?>
up
4
xektrum at gmail dot com
6 years ago
As of php 5.3 CURLOPT_PROGRESSFUNCTION its supported here's how:

<?php

function callback($download_size, $downloaded, $upload_size, $uploaded)
{
   
// do your progress stuff here
}

$ch = curl_init('http://www.example.com');

// This is required to curl give us some progress
// if this is not set to false the progress function never
// gets called
curl_setopt($ch, CURLOPT_NOPROGRESS, false);

// Set up the callback
curl_setopt($ch, CURLOPT_PROGRESSFUNCTION, 'callback');

// Big buffer less progress info/callbacks
// Small buffer more progress info/callbacks
curl_setopt($ch, CURLOPT_BUFFERSIZE, 128);

$data = curl_exec($ch);

?>

Hope this help.
up
3
tim dot php at ebw dot ca
12 years ago
The page http://curl.haxx.se/libcurl/c/curl_easy_setopt.html at the cURL site has a list of all the CURLOPTS, including many not mentioned here. Also see http://curl.haxx.se/libcurl/php/examples/ for cURL examples in PHP.
up
1
joelhy
3 months ago
Please notice that CURLINFO_HEADER_OUT and CURLOPT_VERBOSE option does not work together:
"When CURLINFO_HEADER_OUT is set to TRUE than CURLOPT_VERBOSE does not work."(from https://bugs.php.net/bug.php?id=65348).
This took me an hour or two to figure it out.
up
1
Simeon Anastasov
1 year ago
One note of importance when you open several cURL handles simultaneously: If you want to share cookies via cookie-jar file among all your handles - be sure to curl_close() one before using the cookie-jar file from the other.

It appears that during cURL handler execution the cookies are kept in some sort of handler specific internal session storage and only upon explicit curl_close() call or interpreter exit garbage collection these cookies are actually flushed to the file on the hard disk ( I guess for performance reasons ).

I hope this note will save you couple of hours debugging :)
up
4
Ojas Ojasvi
8 years ago
<?php
/*
* Author: Ojas Ojasvi
* Released: September 25, 2007
* Description: An example of the disguise_curl() function in order to grab contents from a website while remaining fully camouflaged by using a fake user agent and fake headers.
*/

$url = 'http://www.php.net';

// disguises the curl using fake headers and a fake user agent.
function disguise_curl($url)
{
 
$curl = curl_init();

 
// Setup headers - I used the same headers from Firefox version 2.0.0.6
  // below was split up because php.net said the line was too long. :/
 
$header[0] = "Accept: text/xml,application/xml,application/xhtml+xml,";
 
$header[0] .= "text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5";
 
$header[] = "Cache-Control: max-age=0";
 
$header[] = "Connection: keep-alive";
 
$header[] = "Keep-Alive: 300";
 
$header[] = "Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7";
 
$header[] = "Accept-Language: en-us,en;q=0.5";
 
$header[] = "Pragma: "; // browsers keep this blank.

 
curl_setopt($curl, CURLOPT_URL, $url);
 
curl_setopt($curl, CURLOPT_USERAGENT, 'Googlebot/2.1 (+http://www.google.com/bot.html)');
 
curl_setopt($curl, CURLOPT_HTTPHEADER, $header);
 
curl_setopt($curl, CURLOPT_REFERER, 'http://www.google.com');
 
curl_setopt($curl, CURLOPT_ENCODING, 'gzip,deflate');
 
curl_setopt($curl, CURLOPT_AUTOREFERER, true);
 
curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
 
curl_setopt($curl, CURLOPT_TIMEOUT, 10);

 
$html = curl_exec($curl); // execute the curl command
 
curl_close($curl); // close the connection

 
return $html; // and finally, return $html
}

// uses the function and displays the text off the website
$text = disguise_curl($url);
echo
$text;
?>

Ojas Ojasvi
up
1
W at RHaworth dot net
5 years ago
I had problems with the Wikimedia software and sending a POST request where the data was more than 1024 bytes long. I traced this to cURL adding: Expect: 100-continue to the headers.

I added curl_setopt($ch,CURLOPT_HTTPHEADER,array("Expect:")); and that suppresses the Expect line.
up
1
starosielec at googlemail dot com
6 years ago
You can use also use object methods as callback functions. This is usefull if your curl ressource is part of an object handling transfers.
Instead of curl_setopt($curl, CURLOPT_WRITEFUNCTION, "curl_handler_recv") you can use array($object, "method") as value for callback options.

For example:

<?php
class downloader {
    private
$curl;

    function
__construct() {
       
$this->curl = curl_init();
       
curl_setopt($this->curl, CURLOPT_WRITEFUNCTION, array($this, "curl_handler_recv"));
    }

     function
curl_handler_recv($res, $data) {
     
//...
    
}
  
//...
}
?>
up
1
urkle at outoforder dot cc
7 years ago
To send a post as a different content-type (ie.. application/json or text/xml) add this setopt call

<?php
curl_setopt
($ch, CURLOPT_HTTPHEADER,array('Content-Type: application/json'));
?>
up
1
w dot danford at electronics-software dot com
7 years ago
Just a small detail I too easily overlooked.
<?php
/*  If you set:  */
curl_setopt ($ch, CURLOPT_POST, 1);
/* then you must have the data: */
curl_setopt ($ch, CURLOPT_POSTFIELDS, $PostData);
?>
I found with only the CURLOPT_POST set (from copy, paste editing of course) cookies were not getting sent with CURLOPT_COOKIE.  Just something subtle to watch out for.
up
2
heyrocker at yahoo dot com
10 years ago
The examples below for HTTP file upload work great, but I wanted to be able to post multiple files through HTTP upload using HTML arrays as specified in example 38.3 at

http://php.net/features.file-upload

In this case, you need to set the arrays AND keys in the $post_data, it will not work with just the array names. The following example shows how this works:

<?php

    $post_data
= array();
   
   
$post_data['pictures[0]'] = "@cat.jpg";
   
$post_data['pictures[1]'] = "@dog.jpg";
   

   
$ch = curl_init();
   
curl_setopt($ch, CURLOPT_URL, "http://example.com/my_url.php" );
   
curl_setopt($ch, CURLOPT_POST, 1 );
   
curl_setopt($ch, CURLOPT_POSTFIELDS, $post_data);
   
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
   
$postResult = curl_exec($ch);

    if (
curl_errno($ch)) {
       print
curl_error($ch);
    }
   
curl_close($ch);
    print
"$postResult";
?>
up
6
Tim Severien
4 years ago
I've been stuck when using the CURLOPT_CONNECTTIMEOUT_MS constant. In fact, on my PHP version (5.3.1) it's not a number but rather a string. Same thing for CURLOPT_TIMEOUT_MS.

I got this error: Warning: curl_setopt() expects parameter 2 to be long, string given

If you are experiencing simular problems, you can replace the constant with the actual number or (re)define the constant.

CURLOPT_TIMEOUT_MS should be 155
CURLOPT_CONNECTTIMEOUT_MS should be 156

curl_setopt($ch, CURLOPT_CONNECTTIMEOUT_MS, 2500); // error
curl_setopt($ch, 156, 2500); // problem solved
up
1
petelu $ post dot sk
10 years ago
load https:// or  http://example.com/exam.php 
with  POST  data (name=alex&year=18) and apply COOKIEs

<?php
$sessions
= curl_init();
curl_setopt($sessions,CURLOPT_URL,'http://example.com/exam.php');
curl_setopt($sessions, CURLOPT_POST, 1);
curl_setopt($sessions,CURLOPT_POSTFIELDS,'name=alex&year=18');
curl_setopt($sessions,CURLOPT_COOKIEJAR,
dirname(__FILE__).'/cookie.txt');
curl_setopt($sessions,CURLOPT_FOLLOWLOCATION,0);
curl_setopt($sessions, CURLOPT_HEADER , 1);
curl_setopt($sessions, CURLOPT_RETURNTRANSFER,1);
$my_load_page = curl_exec($this->sessions);
?>
up
5
julien veneziano
6 years ago
If you need to send deta in a DELETE request, use:

<?php
$request_body
= 'some data';
$ch = curl_init('http://www.example.com');
       
curl_setopt($ch, CURLOPT_POSTFIELDS, $request_body);
       
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
       
curl_setopt($ch, CURLOPT_HEADER, 0);
       
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
       
curl_setopt($ch, CURLOPT_CUSTOMREQUEST, "DELETE");
       
$response = curl_exec($ch);
var_dump($response);
?>
up
3
Chris at PureFormSolutions dot com
6 years ago
I've found that setting CURLOPT_HTTPHEADER more than once will clear out any headers you've set previously with CURLOPT_HTTPHEADER.

Consider the following:
<?php
   
# ...

   
curl_setopt($cURL,CURLOPT_HTTPHEADER,array (
       
"Content-Type: text/xml; charset=utf-8",
       
"Expect: 100-continue"
   
));

   
# ... do some other stuff ...

   
curl_setopt($cURL,CURLOPT_HTTPHEADER,array (
       
"Accept: application/json"
   
));

   
# ...
?>

Both the Content-Type and Expect I set will not be in the outgoing headers, but Accept will.
up
3
ac at an dot y-co dot de
7 years ago
If you want to connect to a server which requires that you identify yourself with a certificate, use following code. Your certificate and servers certificate are signed by an authority whose certificate is in ca.ctr.

<?php
curl_setopt
($ch, CURLOPT_VERBOSE, '1');
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, '2');
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, '1');
curl_setopt($ch, CURLOPT_CAINFOgetcwd().'/cert/ca.crt');
curl_setopt($ch, CURLOPT_SSLCERT, getcwd().'/cert/mycert.pem');
curl_setopt($ch, CURLOPT_SSLCERTPASSWD, 'password');
?>

If your original certificate is in .pfx format, you have to convert it to .pem using following commands
# openssl pkcs12 -in mycert.pfx -out mycert.key
# openssl rsa -in mycert.key -out mycert.pem
# openssl x509 -in mycert.key >> mycert.pem
up
4
rob
6 years ago
Whats not mentioned in the documentation is that you have to set CURLOPT_COOKIEJAR to a file for the CURL handle to actually use cookies, if it is not set then cookies will not be parsed.
up
2
anonymous
4 years ago
This may be not obvious, but if you specify the CURLOPT_POSTFIELDS and don't specify the CURLOPT_POST - it will still send POST, not GET (as you might think - since GET is default).
So the line:

curl_setopt($ch, CURLOPT_POSTFIELDS, $data);

is synonym to:

curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, $data);

Even if you set the options like this (in this order):

curl_setopt($ch, CURLOPT_POST, 0);
curl_setopt($ch, CURLOPT_POSTFIELDS, $data);

it will send POST, since CURLOPT_POSTFIELDS is latter.
So if you want GET - make sure you don't have CURLOPT_POSTFIELDS specified somewhere.
up
3
miloshio at gmail dot com
3 years ago
Remember:

- 'Server-side' cookies exists as information even before they were set on browser agent(HTTP COOKIE HEADER),
- javascript cookies does NOT exists as information before they were set on browser agent,

so, if you're trying to save cookies using CURLOPT_COOKIEJAR to a local file, that cookie must be server - side cookie, otherwise you are wasting time, javascript-produced cookies only exists when client browser's JS interpreter set them.
up
2
fred at themancan dot com
7 years ago
To find what encoding a given HTTP POST request uses is easy -- passing an array to CURLOPT_POSTFIELDS results in  multipart/form-data:

<?php
curl_setopt
(CURLOPT_POSTFIELDS, array('field1' => 'value'));
?>

Passing a URL-encoded string will result in application/x-www-form-urlencoded:

<?php
curl_setopt
(CURLOPT_POSTFIELDS, array('field1=value&field2=value2'));
?>

I ran across this when integrating with both a warehouse system and an email system; neither would accept multipart/form-data, but both happily accepted application/x-www-form-urlencoded.
up
2
eric at imap dot ch
12 years ago
I managed to use curl to retrieve information from severs on ports other than 80 or 443 (for https) on some installations but not on all.
If you get an "CURLE_COULDNT_CONNECT /* 7 */" error, try adding the port : (for example)

<?php curl_setopt($ch, CURLOPT_PORT, $_SERVER['SERVER_PORT']); ?>
up
1
anon@mouse dot com
5 years ago
If your POST data seems to be disappearing (POST data empty, request is being handled by the server as a GET), try rearranging the order of CURLOPT_POSTFIELDS setting with CURLOPT_NOBODY. CURLOPT_POSTFIELDS has to come AFTER CURLOPT_NOBODY setting because if it comes after it wipes out the Content-Type header that tells your URL target that the request is a POST not a GET.

Not sure if this is expected behavior but it certainly isn't documented (except on Stackoverflow.com, which is supremely unhelpful - BTW, guys over on stack overflow... once you've figured out a PHP problem, posting the solution here would save everyone extra search time).
up
1
Sylvain R
6 years ago
When you are using CURLOPT_FILE to download directly into a file you must close the file handler after the curl_close() otherwise the file will be incomplete and you will not be able to use it until the end of the execution of the php process.

<?php

$fh
= fopen('/tmp/foo', 'w');
$ch = curl_init('http://example.com/foo');
curl_setopt($ch, CURLOPT_FILE, $fh);
curl_exec($ch);
curl_close($ch);

# at this point your file is not complete and corrupted

fclose($fh);

# now you can use your file;

read_file('/tmp/foo');

?>
up
1
mcknight at chek dot com
10 years ago
when specifing the file for either CURLOPT_COOKIEFILE or CURLOPT_COOKIEJAR you may need to use the full file path instead of just the relative path.
up
1
Jakub Horky <jakub dot php at horky dot net>
12 years ago
A bit more documentation (without minimum version numbers):

- CURLOPT_WRITEFUNCTION
- CURLOPT_HEADERFUNCTION
  Pass a function which will be called to write data or headers respectively. The callback function prototype:

long write_callback (resource ch, string data)

The ch argument is CURL session handle. The data argument is data received. Note that its size is variable. When writing data, as much data as possible will be returned in all invokes. When writing headers, exactly one complete header line is returned for better parsing.
The function must return number of bytes actually taken care of. If that amount differs from the amount passed to this function, an error will occur.

- CURLOPT_READFUNCTION
  Pass a function which will be called to read data. The callback function prototype:

string read_callback (resource ch, resource fd, long length)

The ch argument is CURL session handle. The fd argument is file descriptor passed to CURL by CURLOPT_INFILE option. The length argument is maximum length which can be returned.
The function must return string containing the data which were read. If length of the data is more than maximum length, it will be truncated to maximum length. Returning anything else than a string means an EOF.

[Note: there is more callbacks implemented in current cURL library but they aren't unfortunately implemented in php curl interface yet.]
up
1
paul at zgtec dot com
13 years ago
To make a POST in multipart/form-data mode
this worked for me, the " \n" at the end of the variables was very important on my OS X server.

<?php

$file
= "file_to_upload.txt";
$submit_url = "http://www.example.com/upload_page.php";

$formvars = array("cc"=>"us \n");
$formvars[variable_1] = "bla bla \n";
$formvars[variable_2] = "bla bla \n";
$formvars[variable_3] = "bla bla \n";
$formvars[variable_4] = "bla bla \n";
$formvars[upfile] = "@$file"; // "@" causes cURL to send as file and not string (I believe)

    // init curl handle
   
$ch = curl_init($submit_url);
   
curl_setopt($ch, CURLOPT_COOKIEJAR, "my_cookies.txt");  //initiates cookie file if needed
   
curl_setopt($ch, CURLOPT_COOKIEFILE, "my_cookies.txt");  // Uses cookies from previous session if exist
   
curl_setopt($ch, CURLOPT_REFERER, "http://www.example.net");  //if server needs to think this post came from elsewhere
   
curl_setopt($ch, CURLOPT_VERBOSE, 1);
   
curl_setopt($ch, CURLOPT_HEADER, 1);
   
curl_setopt($ch, CURLOPT_FOLLOWLOCATION,1); // follow redirects recursively
   
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
   
curl_setopt($ch, CURLOPT_POSTFIELDS, $formvars);

   
// perform post
   
echo $pnp_result_page = curl_exec($ch);
   
curl_close ($ch);

?>
up
1
fil at rezox dot com
15 years ago
If you want to connect to a secure server for posting info/reading info, you need to make cURL with the openSSL options. Then the sequence is nearly identical to the previous example (except http_S_://, and possibly add the useragent):

<?php
curl_setopt
($ch, CURLOPT_URL,"https://example.com");
//some sites only accept your request if your browser looks legit, so send a useragent profile...
curl_setopt ($ch, CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)");
?>
up
1
shiplu at programmer dot net
2 years ago
CURLOPT_POST should be set before CURLOPT_POSTFIELDS. Otherwise you might encounter 411 Length required error.

Following code generates "411 Length Required" on nginx/1.1.15
<?php
curl_setopt
($ch, CURLOPT_POSTFIELDS, $postfields);
curl_setopt ($ch, CURLOPT_POST, 1);
?>

But this one works.

<?php
curl_setopt
($ch, CURLOPT_POST, 1);
curl_setopt ($ch, CURLOPT_POSTFIELDS, $postfields);
?>
up
1
ericbianchetti at gmail dot com
6 years ago
if you need to send a SOAP string that is the CURL you must use :

<?php
$ch
= curl_init();
curl_setopt($ch, CURLOPT_URL, XML_POST_URL);
curl_setopt ($ch, CURLOPT_HTTPHEADER, array('SOAPAction: ""'));   
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, FALSE);
curl_setopt($ch, CURLOPT_POSTFIELDS, XML_PAYLOAD);
curl_setopt($ch, CURLOPT_HEADER, 0);

$output = curl_exec($ch);
?>

Note : Having based my snipet on Chemo demonstration (oscommerce user know who he is), XML_POST_URL and XML_PAYLOAD where defined as constant with define().

The point is : at the opposite of .xml , SOAP must send the header 'SOAPAction: ""' that can be a valid URI, an empty string (that is here) or nothing ('SOAPAction: '). The later case baing not accepted by all server, the second one indicating the target is the URI used to post the SOAP.
http://www.w3.org/TR/2000/NOTE-SOAP-20000508/#_Toc478383528
up
1
jID
7 years ago
if you use
<?php
curl_setopt
($ch, CURLOPT_INTERFACE, "XXX.XXX.XXX.XXX");
?>
to specify IP adress for request, sometimes you need to get list of all your IP's.

ifconfig command will output something like:

rl0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
    options=8<VLAN_MTU>
    inet 82.146.XXX.XXX netmask 0xffffffff broadcast 82.146.XXX.XXX
    inet 78.24.XXX.XXX netmask 0xffffffff broadcast 78.24.XXX.XXX
    inet 82.146.XXX.XXX netmask 0xffffffff broadcast 82.146.XXX.XXX
    inet 82.146.XXX.XXX netmask 0xffffffff broadcast 82.146.XXX.XXX
    inet 82.146.XXX.XXX netmask 0xffffffff broadcast 82.146.XXX.XXX
    inet 78.24.XXX.XXX netmask 0xffffffff broadcast 78.24.XXX.XXX
    inet 78.24.XXX.XXX netmask 0xffffffff broadcast 78.24.XXX.XXX
    ether XX:XX:XX:XX:XX:XX
    media: Ethernet autoselect (100baseTX <full-duplex>)
    status: active
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> mtu 16384
tun0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1500
    Opened by PID 564
tun1: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1500
    Opened by PID 565
    Opened by PID 565

My solution for FreeBSD 6 and PHP 5 was:
<?php
  ob_start
();
 
$ips=array();
 
$ifconfig=system("ifconfig");
  echo
$ifconfig;
 
$ifconfig=ob_get_contents();
 
ob_end_clean();
 
$ifconfig=explode(chr(10), $ifconfig);
  for (
$i=0; $i<count($ifconfig); $i++) {
   
$t=explode(" ", $ifconfig[$i]);
    if (
$t[0]=="\tinet") {
     
array_push($ips, $t[1]);
    }    
  }
  for (
$i=0; $i<count($ips); $i++) {
    echo
$ips[$i]."\n";
  }
?>

You will get list of IP adresses in $ips array, like:
82.146.XXX.XXX
78.24.XXX.XXX
82.146.XXX.XXX
82.146.XXX.XXX
82.146.XXX.XXX
78.24.XXX.XXX
78.24.XXX.XXX
up
1
andrabr at gmail dot com
8 years ago
This is very clear in hindsight, but it still cost me several hours:

<?php curl_setopt($session, CURLOPT_HTTPPROXYTUNNEL, 1); ?>

means that you will tunnel THROUGH the proxy, as in "your communications will go as if the proxy is NOT THERE".

Why do you care? - Well, if you are trying to use, say, Paros, to debug HTTP between your cURL and the server, with CURLOPT_HTTPPROXYTUNNEL set to TRUE Paros will not see or log your traffic thus defeating the purpose and driving you nuts.

There are other cases, of course, where this option is extremely useful...
up
1
mcbreen at gmail dot com
8 years ago
If you are getting the following error:

SSL: certificate subject name 'example.com' does not match target host name 'example.net'

Then you can set the following option to get around it:

<?php curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, FALSE); ?>
up
1
vincent at ludden dot nl
8 years ago
Please note that the CURLOPT_INTERFACE setting only accepts IP addresses and hostnames of the local machine. It is not meant to send a URL to a specific IP address.
up
1
franciscocha at gmail dot com
6 years ago
Example how to connect to FTPES (FTP explicit SSL). This script will connect to any FTPES server and out put the list of directories.

<?php
    $username
= 'username';
   
$password = 'password';
   
$url = 'example.com';
   
$ftp_server = "ftp://" . $username . ":" . $password . "@" . $url;
   
    echo
"Starting CURL.\n";
   
$ch = curl_init();
    echo
"Set CURL URL.\n";
   
   
//curl FTP
   
curl_setopt($ch, CURLOPT_URL, $ftp_server);
   
   
//For Debugging
    //curl_setopt($ch, CURLOPT_VERBOSE, TRUE);   
   
    //SSL Settings
   
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, FALSE);
   
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, FALSE);
   
curl_setopt($ch, CURLOPT_FTP_SSL, CURLFTPSSL_TRY);
   
   
//List FTP files and directories
   
curl_setopt($ch, CURLOPT_FTPLISTONLY, TRUE);
   
   
//Output to curl_exec
   
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);

    echo
"Executing CURL.\n";
   
$output = curl_exec($ch);
   
curl_close($ch);
    echo
"Closing CURL.\n";
    echo
$output . "\n";

  
$files = explode("\n", $output);
  
print_r($files);
?>
up
2
phpdocs at neothermic dot com
4 months ago
In today's changed TLS landscape, you might eventually encounter servers which only operate on TLS 1.1 or TLS 1.2, and will actively refuse TLS 1.0 connections.

If you're relying on CURLOPT_SSLVERSION's default setting, you'll never make a successful connection to the server.

You can solve this by setting CURLOPT_SSLVERSION to either 5 (for CURL_SSLVERSION_TLSv1_1) or 6 (for CURL_SSLVERSION_TLSv1_2) .
up
1
joey
1 month ago
Sorry, I made a mistake. For validating cookie entries it is best to use at least:

/^([^\t]+\t){6}[^\t]+$/

There was not enough space for me to put in the rationale for not using persistent storage with cookies but it should be obvious. It's YAGNI for most scenarios. In this case at best it complicate things, at the worst you perform an operation using the wrong cookie session. It can also increase the chance of failure, waste resources, reduce performance and create mess in the file system.

The plus of persistent is that In some cases it may be used to accelerate across processes but not many people actually need that and when they do there tend to be better options such as using memcached.

If someone can edit and merge the comments it would be appreciated.
up
1
joey
1 month ago
It is important that anyone working with cURL and PHP keep in mind that not all of the CURLOPT and CURLINFO constants are documented. I always recommend reading the cURL documentation directly as it sometimes contains better information. The cURL API in tends to be fubar as well so do not expect things to be where you would normally logically look for them.

curl is especially difficult to work with when it comes to cookies. So I will talk about what I found with PHP 5.6 and curl 7.26.

If you want to manage cookies in memory without using files including reading, writing and clearing custom cookies then continue reading.

To start with, the way to enable in memory only cookies associated with a cURL handle you should use:

    curl_setopt($curl, CURLOPT_COOKIEFILE, "");

cURL likes to use magic strings in options as special commands. Rather than having an option to enable the cookie engine in memory it uses a magic string to do that. Although vaguely the documentation here mentions this however most people like me wouldn't even read that because a COOKIEFILE is the complete opposite of what we want.

To get the cookies for a curl handle you can use:

    curl_getinfo($curl, CURLINFO_COOKIELIST);

This will give an array containing a string for each cookie. It is tab delimited and unfortunately you will have to parse it yourself if you want to do anything beyond copying the cookies.

To clear the in memory cookies for a cURL handle you can use:

    curl_setopt($curl, CURLOPT_COOKIELIST, "ALL");

This is a magic string. There are others in the cURL documentation. If a magic string isn't used, this field should take a cookie in the same string format as in getinfo for the cookielist constant. This can be used to delete individual cookies although it's not the most elegant API for doing so.

For copying cookies I recommend using curl_share_init.

You can also copy cookies from one handle to another like so:

    foreach(curl_getinfo($curl_a, CURLINFO_COOKIELIST) as $cookie_line)
        curl_setopt($curl, CURLOPT_COOKIELIST, $cookie_line);

An inelegant way to delete a cookie would be to skip the one you don't want.

I only recommend using COOKIELIST with magic strings because the cookie format is not secure or stable. You can inject tabs into at least path and name so it becomes impossible to parse reliably. If you must parse this then to keep it secure I recommend prohibiting more than 6 tabs in the content which probably isn't a big loss to most people.

A the absolute minimum for validation I would suggest:

    /^([^\t]+\t){5}[^\t]+$/D

Here is the format:

    #define SEP  "\t"  /* Tab separates the fields */

    char *my_cookie =
      "example.com"    /* Hostname */
      SEP "FALSE"      /* Include subdomains */
      SEP "/"          /* Path */
      SEP "FALSE"      /* Secure */
      SEP "0"          /* Expiry in epoch time format. 0 == Session */
      SEP "foo"        /* Name */
      SEP "bar";       /* Value */
up
2
Martin K.
2 years ago
If you only want to enable cookie handling and you don't need to save the cookies for a separate session, just set CURLOPT_COOKIEFILE to an empty string.  I was given the advice to use php://memory but that did not seem to have the same effect.

Although this is stated in the documentation I thought it was worth reiterating since it cause me so much trouble.
up
1
Aaron Wells
1 year ago
If you use cURL to fetch user-supplied URLs (for instance, in a web-based RSS aggregator), be aware of the risk of server-side request forgery (SSRF). This is an attack where the user takes advantage of the fact that cURL requests are sent from the web server itself, to reach network locations they wouldn't be able to reach from outside the network.

For instance, they could enter a "http://localhost" URL, and access things on the web server via "localhost". Or, "ftp://localhost". cURL supports a lot of protocols!

If you are using CURLOPT_FOLLOWLOCATION, the malicious URL could be in a redirect from the original request. cURL also will follow redirect headers to other protocols! (303 See Other; Location: ftp://localhost).

So if you're using cURL with user-supplied URLs, at the very least use CURLOPT_PROTOCOLS (which also sets CURLOPT_REDIR_PROTOCOLS), and either disable CURLOPT_FOLLOWLOCATION or use the "SafeCurl" library to safely follow redirects.
up
0
mikko dot rantalainen at peda dot net
3 months ago
If you need to do DELETE request, use CURLOPT_CUSTOMREQUEST with "DELETE" and use CURLOPT_POSTFIELDS for parameters. Do not put request parameters into the URL (GET-like) or bad things will happen (at least Apache+mod_php does not like such requests).
up
1
lachlan at radelaide dot net
2 years ago
For those using CURLAUTH_NTLM, it may come to no surprise  that NTLM request will fail if you have set CURLOPT_FORBID_REUSE to true.

This is because NTLM authorisation is connect-based, not request-based. If the connection is not kept alive and re-used, cURL can never complete the request.

You may notice this if you get a 401 status code or max out the number of redirects.
up
1
Joan
2 years ago
Using CURLOPT_NOPROXY to avoid using the proxy for some urls is very convenient.
For example when the page is trying to look for itself.
The parameter can be found at least in version 5.5.7, (probably earlier)
Unfortunately it's not present on debian wheezy (5.4.4) but it will be on jessie (it's already there)

A related bug: https://bugs.php.net/bug.php?id=53543
up
2
Joey Hewitt
4 years ago
Note that if you put a certificate chain in a PEM file, the certificates need to be ordered so that each certificate is followed by its issuer (i.e., root last.)

Source: http://publib.boulder.ibm.com/tividd/td/ITIM/SC32-1493-00/en_US/HTML/im451_config09.htm
up
2
ellert at _removeme_ vankoperen dot nl
4 years ago
If you are using curl to do a soap request and consistently get the following error back:
The server cannot service the request because the media type is unsupported.
You are sending the Content-type of soap 1.2 to a 1.1 server.
Soap 1.1 needs Content-Type: text/xml;
Soap 1.2 should have Content-Type: application/soap+xml;
up
0
alvaro at demogracia dot com
7 months ago
With the legacy file upload feature, Curl sends the file name of the actual file and there isn't a documented way to change that behaviour. If you aren't able to use the CURLFile class there's a workaround that apparently works: append "; filename=" after the value (and make sure it comes after "type=").

<?php

$postfields
= array(
   
'document[]' => '@/tmp/upload_qm5ee8FBxi; type=image/jpeg; filename=Kitten.jpg',
);
up
2
v dot tverdun at gmail dot com
4 years ago
Make sure to set keys for array if passing to CURLOPT_POSTFIELDS.

<?php
//This can cause errors
$data = array('bar');

//Use this instead
$data = array('foo' => 'bar');

curl_setopt(CURLOPT_POSTFIELDS, $data);
?>
up
1
xuyan83121 at gmail dot com
3 years ago
when use curl_multi_exec, the CURLOPT_TIMEOUT need curl version newer than 7.21.2.
this can be found at curl changelog.
up
0
White Gandalf
1 year ago
to complement shiplu's  comment on the neccessary option sequence of CURLOPT_POST before CURLOPT_POSTFIELDS:

The crux is not some error on nginx, but that nothing at all will be send over the line by curl. Parameters set by a "CURLOPT_POSTFIELDS" option setting will be completely ignored, as long as no "CURLOPT_POST" has been encountered beforehand: Neigther the Content-Type header will be set/generated accordingly nor Content-Length nor any data will be send in the body.

When using curl_setopt_array, the sequence in the array matters as well.
up
2
john dot david dot steele at gmail dot com
5 years ago
A note on the way Curl posts files...

<?php
  curl_setopt
($ch, CURLOPT_POSTFIELDS, array('file' => '@/path/to/file.ext');
?>

will post the FULL PATH of the file in the filename field:

Content-Disposition: form-data; name="file"; filename="/path/to/file.ext"

Whereas typical browser behavior only sends the filename:

Content-Disposition: form-data; name="file"; filename="file.ext"

Workaround:
<?php
  curl_setopt
($ch, CURLOPT_POSTFIELDS, array('file' => '@file.ext');
 
$cwd = getcwd();
 
chdir('/path/to/');
 
$receivedData = curl_exec($ch);
 
chdir($cwd);
?>
up
0
Martin K.
2 years ago
If you need to read page contents in between file downloads, while still using the same curl handle, you'll probably need this code:
<?php
    curl_setopt
($handle, CURLOPT_FILE, fopen('php://stdout','w'));   // 'php://output' didn't work for me
   
curl_setopt($handle, CURLOPT_RETURNTRANSFER, true);  // using CURLOPT_FILE sets this to false automatically
?>
up
1
jeffb at nospam dot videx dot com
3 years ago
When trying to pass a multi-dimensional array to CURLOPT_POSTFIELDS, first run it through http_build_query().  That will get rid of the Array to String conversion notice.
up
2
sam at def dot reyssi dot net
5 years ago
Be careful when setting the CURLOPT_POSTFIELDS setting using an array. The array used to set the POST fields must only contain scalar values. Multidimentional arrays or objects lacking a __toString implementation will cause Curl to error.

If there is a need to send non-scalar values using a POST request, consider serializing them before transmission.

<?php
$ch
= curl_init('http://host.example.com');

// Data to post
$multiDimensional = array(
  
'name' = 'foo',
  
'data' = array(1,2,3,4),
  
'value' = 'bar'
);

// Will error
curl_setopt($ch, CURLOPT_POSTFIELDS, $multiDimensional);

// Data to post
$postData = array(
 
'name' = 'foo',
 
'data' = serialize(array(1,2,3,4)),
 
'value' = 'bar'
);

// Will not error
curl_setopt($ch, CURLOPT_POSTFIELDS, $postData);
?>
up
0
Manasi
2 years ago
CURLAUTH_ANY is not an alias for CURLAUTH_NTLM. I had to specify CURLAUTH_NTLM for a Windows authenticated URL
up
0
S.F.
3 years ago
I spent a couple of days trying to upload a file using a curl post.

The problem I ran into was the filename had an '@' in the middle of it.  It turned out that at least on my system if I encoded the file path using the quoted_printable_encode() function the upload works.

I'm posting this in the hopes that it will help someone else, and for my own future reference.

Code:
<?php

$filepath
= '/tmp/test@example.txt';
$postdata['file'] = '@' . quoted_printable_encode($filepath);

//... supporting code.

$result = curl_exec($ch);

?>
I'm not exactly sure why this works when escaping the '@' doesn't work but it does for me.

If anyone can offer insight into why this works or a better way to handle the '@' symbol in a filename when using curl to upload I would love to hear it.

Thanks
up
2
Stan van de Burgt
6 years ago
If you get a "failed creating formpost data" upon curl_exec() when POSTing a form, check if one of the field values starts with the @ character.

Took me an hour or so to find out as I wanted to post a @reply tweet to twitter which typically start with @screenname.
up
2
Tyranoweb
6 years ago
There is a function to send POST data in page with five parameters :

$post must be an array
$page is the page where POST datas will be send.
$n must be true to continue if they are php redirection (Location: )
$session must be define true if you want to use cookies
$referer must be a link to get a wrong referer or only to have a referer.

<?php
function curl_data_post($post, $page, $n, $session, $referer)
    {
        if(!
is_array($post))
        {
         return
false;
        }
       
       
$DATA_POST = curl_init();
       
curl_setopt($DATA_POST, CURLOPT_RETURNTRANSFER, true);
       
curl_setopt($DATA_POST, CURLOPT_URL, $page);
       
curl_setopt($DATA_POST, CURLOPT_POST, true);
        if(
$n)
        {
        
curl_setopt($DATA_POST, CURLOPT_FOLLOWLOCATION, true);
        }
        if(
$session)
        {
        
curl_setopt($DATA_POST, CURLOPT_COOKIEFILE, 'cookiefile.txt');
        
curl_setopt($DATA_POST, CURLOPT_COOKIEJAR, 'cookiefile.txt');
        }
       
        if(
$referer)
        {
        
curl_setopt($DATA_POST, CURLOPT_REFERER, $referer);
        }
       
       
curl_setopt($DATA_POST, CURLOPT_POSTFIELDS, $post);
       
$data = curl_exec($DATA_POST);
        if(
$data == false)
        {
         echo
'Warning : ' . curl_error($DATA_POST);
        
curl_close($DATA_POST);
         return
false;
        }
        else
        {
        
curl_close($DATA_POST);
         return
$data;
        }
    }
?>
up
2
Victor Jerlin
6 years ago
Seems like some options not mentioned on this page, but listed on http://curl.haxx.se/libcurl/c/curl_easy_setopt.html is actually supported.

I was happy to see that I could actually use CURLOPT_FTP_CREATE_MISSING_DIRS even from PHP.
up
1
gskluzacek at gmail dot com
5 years ago
FYI... unless you specifically set the user agent, no user agent will be sent in your request as there is no default value like some of the other options.

As others have said, not sending a user agent may cause you to not  get the results that you expected, e.g., 0 byte length content, different content, etc.
up
0
juozaspo at gmail dot com
3 years ago
I've created an example that gets the file on url passed to script and outputs it to the browser.

<?php
//get the file (e.g. image) and output it to the browser
$ch = curl_init(); //open curl handle
curl_setopt($ch, CURLOPT_URL, $_GET['url']); //set an url
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); //do not output directly, use variable
curl_setopt($ch, CURLOPT_BINARYTRANSFER, 1); //do a binary transfer
curl_setopt($ch, CURLOPT_FAILONERROR, 1); //stop if an error occurred
$file=curl_exec($ch); //store the content in variable
if(!curl_errno($ch))
{
   
//send out headers and output
   
header ("Content-type: ".curl_getinfo($ch, CURLINFO_CONTENT_TYPE)."");
   
header ("Content-Length: ".curl_getinfo($ch, CURLINFO_CONTENT_LENGTH_DOWNLOAD)."");
    echo
$file;
} else echo
'Curl error: ' . curl_error($ch);
curl_close($ch); //close curl handle
?>

p.s. Make sure that there're no new lines before and after code or script may not work.
up
1
scy-phpmanual at scytale dot name
5 years ago
In order to reset CURLOPT_HTTPHEADER, set it to array(). The cURL C API says you should set it to NULL, but that doesn’t work in the PHP wrapper.
up
1
prohfesor at gmail dot com
5 years ago
This function helps to parse netscape cookie file, generated by cURL into cookie array:

<?php
 
function _curl_parse_cookiefile($file) {
   
$aCookies = array();
   
$aLines = file($file);
    foreach(
$aLines as $line){
      if(
'#'==$line{0})
        continue;
     
$arr = explode("\t", $line);
      if(isset(
$arr[5]) && isset($arr[6]))
       
$aCookies[$arr[5]] = $arr[6];
    }
   
    return
$aCookies;
  }
?>
up
1
kavih7 at yahoo dot com
5 years ago
When POSTing with cURL, my POSTs were magically being converted to GETs and I debugged it until finding the issue. I was setting the CURLOPT_MUTE option. Not sure why this conflicts, since the documentation doesn't specify as such. Anyways, if your $_POST is empty, make sure you aren't setting CURLOPT_MUTE.

Cheers!
up
0
mw+php dot net at lw-systems dot de
4 years ago
The description of the use of the CURLOPT_POSTFIELDS option should be emphasize, that using POST with HTTP/1.1 with cURL implies the use of a "Expect: 100-continue" header. Some web servers will not understand the handling of chunked transfer of post data.

To disable this behavior one must disable the use of the "Expect:" header with

    curl_setopt($ch, CURLOPT_HTTPHEADER,array("Expect:"));
up
1
patryk at do dot not dot spam dot com
5 years ago
CURLOPT_HTTPPROXYTUNNEL means curl will use CONNECT method of the HTTP protocol to make a tunnel through a proxy, which is most likely not the one you want to do.
up
0
dotpointer at gmail dot com
4 years ago
About CURLOPT_COOKIEJAR and CURLOPT_COOKIEFILE, and which / how to use.

- CURLOPT_COOKIEJAR is used when cURL is reading cookie data from disk.

- CURLOPT_COOKIEFILE is used when cURL is writing the cookie data to disk.

So you need to specify both (and set the same file location on both) when working with sessions for example.
up
0
wonderfish+php at gmail dot com
4 years ago
As of at least PHP 5.3.9, if you are continuing to use a cURL session handle after downloading a file and closing the file handle, you will need to change CURLOPT_FILE back to stdout, and cannot count simply on a side effect of CURLOPT_RETURNTRANSFER to do so, even if you are setting it.  For example:

<?php
$ch
= curl_init();
$fh = fopen('/path/to/stored/file/example_file.dat', 'w');
curl_setopt($ch, CURLOPT_FILE, $fh);
curl_setopt($ch, CURLOPT_URL, 'http://example.com/example_file.dat');
curl_exec($ch);
fflush($fh);
fclose($fh);

//must reset cURL file handle. Not doing so will cause a warning to be
//thrown and for cURL to default to output regardless
//for our example, we'll set return transfer.
curl_setopt($ch, CURLOPT_FILE, fopen('php://stdout', 'w'));
curl_setopt($ch, CURLOPT_RETURNTRANSFER, TRUE);
curl_setopt($ch, CURLOPT_URL, 'http://example.org/index.html');
$html = curl_exec($ch); //this will now work
?>
up
1
obones_remove_me at free dot fr
6 years ago
For those of you wondering how to specify the content-type for a file uploaded via curl, the syntax is as follows:

<?php

$data
= array('file' => '@/home/user/test.png;type=image/png');

?>

Simply adding a semicolon with the type= at the end.
Note that this has been reported not to work in all versions of PHP and I have done the following tests:

5.2.6 (libcurl 7.18.2) : Does not work
5.2.13 (libcurl 7.20.0) : Works just fine

So it might be worth updating your installation of PHP and/or libcurl if you want to be able to use this syntax
up
0
chsa at ten dot ch
4 years ago
Sending a post file upload across a squid proxy, the request was rejected by the proxy. In the error page returned it provided among other possible causes:"Expect:" feature is being asked from a HTTP/one.zero.
Solution: Add the option <?php curl_setopt($cl,CURLOPT_HTTPHEADER,array("Expect:")); ?>. This will remove the expect http header.
up
2
Salil Kothadia
7 years ago
In PHP5, for the "CURLOPT_POSTFIELDS" option, we can use:

<?php
$ch
= curl_init($URI);
$Post = http_build_query($PostData);
curl_setopt($ch, CURLOPT_POSTFIELDS, $Post);
$Output = curl_exec($ch);
curl_close($ch);
?>
up
2
mavook at gmail dot com
8 years ago
If you try to upload file to a server, you need do CURLOPT_POST first and then fill CURLOPT_POSTFIELDS.

<?php
curl_setopt
($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, $postvars);
// ^^ This will post multipart/form-data
?>
up
0
m dot ghadam at gmail dot com
4 years ago
When you set ($ch, curlopt_post, 1) , after you have posted your data with curl_exec , you need to set ($ch, curlopt_post, 0), Otherwise all your subsequent requests seems as a post with no postdata and some reverse proxy servers send 500 or 403 error for these case ( access denied or forbidden )!
up
1
Andrew
6 years ago
I noticed that if you want to get current cookie file after curl_exec() - you need to close current curl handle (like it said in manual), but if you want cookies to be dumped to file after any curl_exec (without curl_close) you can:

<?php
#call it normally
$ch = curl_init();
curl_setopt($ch, CURLOPT_HEADER, 0);
curl_setopt($ch, CURLOPT_COOKIEFILE, "cookiefile");
curl_setopt($ch, CURLOPT_COOKIEJAR, "cookiefile");
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
curl_setopt($ch, CURLOPT_URL, 'http://www.example.com/');
$result1 = curl_exec($ch);

#and then  make a temp copy
$ch_temp=curl_copy_handle(ch);
curl_close($ch);
$ch=$ch_temp;
?>

Only this way, if you close $ch_temp - cookies wont be dumped.
up
1
ashooner - - gmail , com
7 years ago
When passing CURLOPT_POSTFIELDS a url-encoded string in order to use Content-Type: application/x-www-form-urlencoded, you can pass a string directly:
<?php
curl_setopt
(CURLOPT_POSTFIELDS, 'field1=value&field2=value2');
?>

rather than passing the string in an array, as in fred at themancan dot com's example.
up
2
alfredoaguirre dot v at gmail dot com
9 years ago
Seems that CURLOPT_RETURNTRANSFER Option set to TRUE, returns a "1" when the transaction returns a blank page.

I think is for eliminate the FALSE to can be with a blank page as return
up
2
paczor
9 years ago
How to get rid of response after POST: just add callback function for returned data (CURLOPT_WRITEFUNCTION) and make this function empty.

<?php
function curlHeaderCallback($ch, $strHeader) {
}
curl_setopt($ch, CURLOPT_WRITEFUNCTION, 'curlHeaderCallback');
?>
up
0
shailesh4all at gmail dot com
5 years ago
Hi,
Anyone who is interested in submitting their information by post to HTTPS site (e.g. payment gateway) where https page needs basic authentication before submitting the information. below code will be helpful.

<?php
$submit_url
= "https://sitename/process.php";

$curl = curl_init();

curl_setopt($curl, CURLOPT_HTTPAUTH, CURLAUTH_BASIC ) ;
curl_setopt($curl, CURLOPT_USERPWD, "username:password");
curl_setopt($curl, CURLOPT_SSLVERSION,3);
curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, FALSE);
curl_setopt($curl, CURLOPT_SSL_VERIFYHOST, 2);
curl_setopt($curl, CURLOPT_HEADER, true);
curl_setopt($curl, CURLOPT_POST, true);
curl_setopt($curl, CURLOPT_POSTFIELDS, $params );
curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
curl_setopt($curl, CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)");
curl_setopt($curl, CURLOPT_URL, $submit_url);

$data = split("text/html", curl_exec($curl) );
$temp = split("\r\n", $data[1]) ;

$result = unserialize( $temp[2] ) ;

print_r($result);
curl_close($curl);
?>
up
0
sjungwirth at google mail dot com
6 years ago
I couldn't find a way to force a curl request to go to a particular IP address, but you can do it with fsockopen:
<?php
    $ip
= '123.45.67.89';
   
$fp = fsockopen($ip, 80, $errno, $errstr, 5);
   
$result = '';
    if (!
$fp) {
        echo
"$errstr ($errno)<br />\n";
    } else {
       
$out = "GET /path/to/the/file.ext HTTP/1.1\r\n";
       
$out .= "Host: www.exampl.com\r\n";
       
$out .= "Connection: Close\r\n\r\n";
       
fwrite($fp, $out);
        while (!
feof($fp)) {
           
$result .= fgets($fp, 128);
        }
       
fclose($fp);
    }
?>
I needed it to test the response from a set of servers behind a load balancer.
up
0
clint at fewbar dot com
6 years ago
If you have turned on conditional gets on a curl handle, and then for a subsequent request, you don't have a good setting for CURLOPT_TIMEVALUE , you can disable If-Modified-Since checking with:

<?php

$ch
= curl_init();
curl_setopt($ch, CURLOPT_URL, $foo);
curl_setopt($ch, CURLOPT_TIMEVALUE, filemtime($foo_path));
curl_setopt($ch, CURLOPT_TIMECONDITION, CURLOPT_TIMECOND_IFMODIFIEDSINCE);
curl_exec($ch);
// Reuse same curl handle
curl_setopt($ch, CURLOPT_URL, $bar);
curl_setopt($ch, CURLOPT_TIMEVALUE, null); // don't know mtime
curl_setopt($ch, CURLOPT_TIMECONDITION, 0); // set it to 0, turns it off
curl_exec($ch);

?>
up
2
php at miggy dot org
9 years ago
Note that if you want to use a proxy and use it as a _cache_, you'll have to do:

<?php curl_setopt($ch, CURLOPT_HTTPHEADER, array("Pragma: ")); ?>

else by default Curl puts a "Pragma: no-cache" header in and thus force cache misses for all requests.
up
0
mikeseth at gmail dot com
6 years ago
When using CURLOPT_FILE, pass it the file handle that is open for write only (eg fopen('blahblah', 'w+')). If you also open the file for reading (eg fopen('blahblah', 'rw')), curl will fail with error 23.
up
2
skyogre __at__ yandex __dot__ ru
10 years ago
There is really a problem of transmitting $_POST data with curl in php 4+ at least.
I improved the encoding function by Alejandro Moreno to work properly with mulltidimensional arrays.

<?php
function data_encode($data, $keyprefix = "", $keypostfix = "") {
 
assert( is_array($data) );
 
$vars=null;
  foreach(
$data as $key=>$value) {
    if(
is_array($value)) $vars .= data_encode($value, $keyprefix.$key.$keypostfix.urlencode("["), urlencode("]"));
    else
$vars .= $keyprefix.$key.$keypostfix."=".urlencode($value)."&";
  }
  return
$vars;
}

curl_setopt($ch, CURLOPT_POSTFIELDS, substr(data_encode($_POST), 0, -1) );

?>
up
2
phpnet at andywaite dot com
10 years ago
After setting CURLOPT_FILE, you may want want to revert back to the normal behaviour of displaying the results. This can be achieved using:

<?php
$fp
= fopen ("php://output", "w") or die("Unable to open stdout for writing.\n");
curl_setopt($ch, CURLOPT_FILE, $fp);
?>
up
2
michaeledwards.com
10 years ago
Problems can occur if you mix CURLOPT_URL with a 'Host:' header in CURLOPT_HEADERS on redirects because cURL will combine the host you explicitly stated in the 'Host:' header with the host from the Location: header of the redirect response.

In short, don't do this:

<?php
$host
= "www.example.com";
$url = "http://$host/";

$headers = array("Host: $host");

$ch = curl_init();

curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);

Do
this instead:

$host = "www.example.com";
$url = "http://$host/";

$ch = curl_init();

curl_setopt($ch, CURLOPT_URL, $url);
?>
up
0
OPALA
7 years ago
To fetch (or submit data to) multiple pages during one session,use this:

<?php
$ch
= curl_init();
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, FALSE);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, TRUE);
curl_setopt($ch, CURLOPT_COOKIESESSION, TRUE);
curl_setopt($ch, CURLOPT_HEADER, 0);
curl_setopt($ch, CURLOPT_COOKIEFILE, "cookiefile");
curl_setopt($ch, CURLOPT_COOKIEJAR, "cookiefile");
curl_setopt($ch, CURLOPT_COOKIE, session_name() . '=' . session_id());
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);

curl_setopt($ch, CURLOPT_URL, 'http://example.com/page1.php');
$result1 = curl_exec($ch);

curl_setopt($ch, CURLOPT_URL, 'http://example.com/page2.php');
$result2 = curl_exec($ch);

curl_close($ch);
?>
up
0
charles at tastik dot net
8 years ago
FYI,

Anyone trying to connect to .NET with CURL to send a simple XML post, pay attention to the following. This will save you hours! There is a previous note that I saw either on this page, or somewhere else on this site that explains the correct way to specify the header option is to create an array, then reference the array from the CURLOPT.

ie.  Do something like this:

<?php
// Req. HTTP Header Values
$header[] = "Content-type: text/xml";

// Target URL
$sendTo = "http://www.example.com";

// Post Data
$post = "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n<root>\n....etc, etc,";

// Create CURL Connection
$ch = curl_init();
curl_setopt($ch, CURLOPT_USERAGENT, 'XtraDoh xAgent');
curl_setopt($ch, CURLOPT_URL, $sendTo);
curl_setopt($ch, CURLOPT_TIMEOUT, 900);
curl_setopt($ch, CURLOPT_CONNECTIONTIMEOUT, 30);
curl_setopt($ch, CURLOPT_FAILONERROR, false);
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, 0);
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, 0);
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_HTTPHEADER, $header);
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, $post);
?>

Notice the HTTPHEADER, $header above. I have not been able to get .NET to properly read the HTTP header as specified (in this case as text/xml) when using the following:

<?php
curl_setopt
($ch, CURLOPT_HTTPHEADER, array('Content-Type'=>'text/xml'));
?>

Although this may work when working with other PHP, IIS, or even PHP, Apache, it does not (at least in my experience) work with .NET, IIS.
up
0
michael sky
8 years ago
if you are trying to connect to 'https://...' and after that want to work with POST data - that's the way:

<?php
$curl
= curl_init();
curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, FALSE);
curl_setopt($curl, CURLOPT_HEADER, 0);
curl_setopt($curl, CURLOPT_POST, true);
curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
curl_setopt($curl, CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)");
curl_setopt($curl, CURLOPT_FOLLOWLOCATION, true);
curl_setopt($curl, CURLOPT_COOKIEFILE, "cookiefile");
curl_setopt($curl, CURLOPT_COOKIEJAR, "cookiefile"); # SAME cookiefile
curl_setopt($curl, CURLOPT_URL, "url1"); # this is where you first time connect - GET method authorization in my case, if you have POST - need to edit code a bit
$xxx = curl_exec($curl);

curl_setopt($curl, CURLOPT_URL, "url2"); # this is where you are requesting POST-method form results (working with secure connection using cookies after auth)
curl_setopt($curl, CURLOPT_POSTFIELDS, "var1=value&var2=value&var3=value&"); # form params that'll be used to get form results
$xxx = curl_exec($curl);

curl_close ($curl);
echo
$xxx;
?>
up
0
me
12 years ago
Just a reminder: When setting your CURLOPT_POSTFIELDS remember to replace the spaces in your values with %20
up
0
Anonymous
13 years ago
beware that not all cURLlib constants are supported under php :
e.g. CURLOPT_PROGRESSFUNCTION or CURLOPT_WRITEDATA are not supported.

CURLOPT_WRITEFUNCTION, although undocumented is supported. It takes the name of a user_defined function.
the function should take two arguments (the curl handle, and the inputdata) and return the length of the written data
e.g.

<?php
function myPoorProgressFunc($ch,$str){
global
$fd;
$len = fwrite($fd,$str);
print(
"#");
return
$len;
}

curl_setopt($ch,CURLOPT_WRITEFUNCTION,"myPoorProgressFunc");
?>

Also be aware that CURLOPT_WRITEFUNCTION  does NOT take the CURLOPT_FILE as a parameter!
in curl lib it would take CURLOPT_WRITEDATA but this is not supported by php; that's why I use "global $fd;" in my exemple function.

CURLOPT_HEADERFUNCTION works the same, and is guaranteed to receive complete header lines as input!

Hope this helps

Ivan
up
0
bvwj at swbell dot net
14 years ago
To collect cookies recieved with a request, set CURLOPT_COOKIEJAR "cookieFileName".  Then use CURLOPT_COOKIEFILE "cookieFileName" to recall them in subsequent transactions.
up
0
dan dot polansky at seznam dot cz
15 years ago
I used to download www pages to my script and one of the pages was different in MS explorer and different, when I downloaded it. Namely, information, I was really interested in was missing. That was because the server on the other bank of the river was looking at who is downloading the page. Everything got fixed when I pretended I was MSIE. It is done with curl. Here is a function, that you may use in similar situation

<?php
function download_pretending($url,$user_agent) {
  
$ch = curl_init();
  
curl_setopt ($ch, CURLOPT_URL, $url);
  
curl_setopt ($ch, CURLOPT_USERAGENT, $user_agent);
  
curl_setopt ($ch, CURLOPT_HEADER, 0);
  
curl_setopt ($ch, CURLOPT_RETURNTRANSFER, 1);
  
$result = curl_exec ($ch);
  
curl_close ($ch);
   return
$result;
}
?>
up
-1
mfuhrman at enetarch dot net
2 years ago
Here is an example of sending JSON data as via POST.

index.php
------
$data = array
(
  "command" => "isInstalled",
  "params" => Array
  (
    "1" => "3",
    "2" => "4",
  )
);

$data_string = "json=" . json_encode($data) . "&";
$ch = curl_init("http://localhose/parrot.php");
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, $data_string);
$results = curl_exec($ch);

print ($results);
?>
=====

parrot.php
------
<?
print_r ($_REQUEST);
?>
up
-2
dorphalsig at gmail dot com
7 years ago
This may not be a surprise for many, but I know I bled my eyes out trying to implement this in php. And when I knew it was this simple, I really felt extremely stupid. So I put this just so google will save somebody some time in the future.

PHP NTLM AUTH

Make sure you have the 'curl' extension loaded
now just do...

<?php
curl_setopt
($ch,CURLAUTH_NTLM);
curl_setopt($ch,CURLOPT_USERPWD,"$username:$password");
?>

and just continue to use curl in the ordinary fashion.
up
-3
Anonymous
14 years ago
To make a POST in multipart/form-data mode

(to upload a file for example) you can use

<?php curl_setopt($ch,CURLOPT_POSTFIELDS,$post); ?>

where $post is an array :

<?php
$post
['key1'] = 'data1';
//  like a text field in a POST
$post['file1'] = '@filename1'
// upload filename1
?>

For more informations see the

curl_formparse man page.
up
-4
maxime dot elomari at gmail dot com
1 year ago
If you get an error with the error code 35 saying "Unknown SSL protocol error in connection to ...", maybe you are using the wrongs ciphers.

Try to precise a bunch of ciphers as below:

$arrayCiphers = array(
                'DHE-RSA-AES256-SHA',
        'DHE-DSS-AES256-SHA',
        'AES256-SHA:KRB5-DES-CBC3-MD5',
        'KRB5-DES-CBC3-SHA',
        'EDH-RSA-DES-CBC3-SHA',
        'EDH-DSS-DES-CBC3-SHA',
        'DES-CBC3-SHA:DES-CBC3-MD5',
        'DHE-RSA-AES128-SHA',
        'DHE-DSS-AES128-SHA',
        'AES128-SHA:RC2-CBC-MD5',
        'KRB5-RC4-MD5:KRB5-RC4-SHA',
        'RC4-SHA:RC4-MD5:RC4-MD5',
        'KRB5-DES-CBC-MD5',
        'KRB5-DES-CBC-SHA',
        'EDH-RSA-DES-CBC-SHA',
        'EDH-DSS-DES-CBC-SHA:DES-CBC-SHA',
        'DES-CBC-MD5:EXP-KRB5-RC2-CBC-MD5',
        'EXP-KRB5-DES-CBC-MD5',
        'EXP-KRB5-RC2-CBC-SHA',
        'EXP-KRB5-DES-CBC-SHA',
        'EXP-EDH-RSA-DES-CBC-SHA',
        'EXP-EDH-DSS-DES-CBC-SHA',
        'EXP-DES-CBC-SHA',
        'EXP-RC2-CBC-MD5',
        'EXP-RC2-CBC-MD5',
        'EXP-KRB5-RC4-MD5',
        'EXP-KRB5-RC4-SHA',
        'EXP-RC4-MD5:EXP-RC4-MD5'
);

curl_setopt($ch, CURLOPT_SSL_CIPHER_LIST, implode(':', $arrayCiphers));

Worked for me, could work for you!

P.S: Used with PHP 5.4 and cURL 7.26.0.
up
-3
dardyole at hotmail dot com
3 years ago
Another note addressing the issues with servers that have open_basedir and safe mode turned on. Such an issue spawns the following E_WARNING:

Warning: curl_setopt() [function.curl-setopt]: CURLOPT_FOLLOWLOCATION cannot be activated when safe_mode is enabled or an open_basedir is set

After looking through the notes, most of the proposed manual implementations were kind of clunky and in some cases just didn't work at all. Most importantly (in my case), was the behaviour of the 302 Header. Anyway, here's the code I ended up using which has worked well for me in all cases so far, it even addresses the issue that caused FOLLOWLOCATION to be turned off in some cases :)

EDIT: Unfortunately the code itself is deemed "too long" for PHP's note system. I've uploaded it to a few paste sites below so hopefully the links will live for a while at least.

http://pastebin.com/aaJtPy1j
http://pastie.org/7646116

Use it as a replacement for curl_exec. For example:

<?php
$ch
= curl_init("http://php.net");
var_dump(curl_exec_follow($ch, 9001));
curl_close($ch);
?>
up
-4
tychay at alumni dot caltech dot edu
14 years ago
CURLOPT_HTTPHEADER is NOT like the -H command line switch. The command line switch adds or replaces headers (much like the header() line in PHP, but for HTTP clients instead of servers), but the curl extension will eliminate the headers cURL sends by default.

For instance, your Authorization, Host, Referer, Pragma, and Accept headers which are normally written by default or by other CURLOPT_*'s.

Also, it might seem intuitive that this should accept an array hash of header->values, but this is not the case. It accepts an array of strings of the format "Header: Value", much like the -H command-line switch.

Hope this helps,

terry
up
-2
coding query
3 years ago
Sometimes we want to extract the HTML content of the remote website page, this technique is called as HTML scrapper. This article will discuss on how we can extract the HTML content of the remote webpage.
We can achieve HTML scrapper operation in 2 step operation:
Call to Remote Web Page and extract the HTML content.
Match the HTML tags using Regular Expression.
Call to Remote Web Page using PHP:
In PHP there are various ways we can call the remote webpage. But here we will be using CURL to achieve our operation.

<?php
$ch
= curl_init();
$timeout = 5; // set to zero for no timeout
curl_setopt ($ch, CURLOPT_URL, $url);
curl_setopt ($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt ($ch, CURLOPT_CONNECTTIMEOUT, $timeout);
$file_contents = curl_exec($ch);
curl_close($ch);

preg_match_all('/<span>[\\/\\(\\)-:<>\\w\\s]+< \\/span>/',$file_contents,$htmlContent);
?>
up
-4
etiennez0r at gmail dot com
2 years ago
// For those having message
// SSL connection timeout
curl_setopt($ch, CURLOPT_SSLVERSION, 1);
// solved for me
up
-9
gabriel dot rota at gmail dot com
2 years ago
php curl pass through proxy handle: big file, https, autentication

<?php
/*
* curl-pass-through-proxy.php
*
* Copyright 2014 Gabriel Rota <gabriel.rota@gmail.com>
*
* propose: php curl pass through proxy handle: big file, https, autentication
* example: curl-pass-through-proxy.php?url=precise/ubuntu-12.04.4-desktop-i386.iso
* limitation: don't work on binary if is enabled in php.ini the ;output_handler = ob_gzhandler
* licence: BSD
*
*/

 
$url = "http://releases.ubuntu.com/" . $_GET["url"]; // NOTE: this example don't use https
 
$credentials = "user:pwd";
 
$headers = array(
   
"GET ".$url." HTTP/1.1",
   
"Content-type: text/xml",
   
"Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
   
"Cache-Control: no-cache",
   
"Pragma: no-cache",
   
"Authorization: Basic " . base64_encode($credentials)
  );

  global
$filename; // used in fn_CURLOPT_HEADERFUNCTION setting download filename
 
$filename = substr($url, strrpos($url, "/")+1); // find last /

 
function fn_CURLOPT_WRITEFUNCTION($ch, $str){
   
$len = strlen($str);
    echo(
$str );
    return
$len;
  }

  function
fn_CURLOPT_HEADERFUNCTION($ch, $str){
    global
$filename;
   
$len = strlen($str);
   
header( $str );
   
//~ error_log("curl-pass-through-proxy:fn_CURLOPT_HEADERFUNCTION:str:".$str.PHP_EOL, 3, "/tmp/curl-pass-through-proxy.log");
   
if ( strpos($str, "application/x-iso9660-image") !== false ) {
     
header( "Content-Disposition: attachment; filename=\"$filename\"" ); // set download filename
   
}
    return
$len;
  }

 
$ch = curl_init(); // init curl resource
   
curl_setopt($ch, CURLOPT_URL,$url);
   
curl_setopt($ch, CURLOPT_RETURNTRANSFER, false); // a true curl_exec return content
   
curl_setopt($ch, CURLOPT_TIMEOUT, 600); // 60 second
   
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers); // login $url
   
curl_setopt($ch, CURLOPT_USERAGENT, $_SERVER['HTTP_USER_AGENT']);
   
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false); // don't check certificate
   
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false); // don't check certificate
   
curl_setopt($ch, CURLOPT_HEADER, false); // true Return the HTTP headers in string, no good with CURLOPT_HEADERFUNCTION
   
curl_setopt($ch, CURLOPT_BUFFERSIZE, 8192); // 8192 8k
   
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
   
curl_setopt($ch, CURLOPT_HEADERFUNCTION, "fn_CURLOPT_HEADERFUNCTION"); // handle received headers
   
curl_setopt($ch, CURLOPT_WRITEFUNCTION, 'fn_CURLOPT_WRITEFUNCTION'); // callad every CURLOPT_BUFFERSIZE

   
if ( ! curl_exec($ch) ) {
     
error_log( "curl-pass-through-proxy:Error:".curl_error($ch).PHP_EOL, 3, "/tmp/curl-pass-through-proxy.log" );
    }

 
curl_close($ch); // close curl resource

?>
up
-5
webmaster () stauceni.com
10 years ago
A little mistake, that took a half-day to fix it:
When specifing CURLOPT_COOKIEFILE or CURLOPT_COOKIEJAR options, don't forget to "chmod 777" that directory where cookie-file must be created.
up
-11
George
7 years ago
If you set CURLOPT_RESUME_FROM to resume the file, and then reuse the same Curl handle to download another file, you must reset the resume status by calling curl_setopt( $ch, CURLOPT_RESUME_FROM, 0 ). It will not reset, and will apply to all subsequent transfers even if the URL is the same.
To Top