Get URL parts

The CURLU handle stores the individual parts of a URL and the application can extract those pieces individually from the handle at any time. If they are set.

The second argument to curl_url_get() specifies which part you want extracted. They are all extracted as null-terminated char * data, so you pass a pointer to such a variable.

char *host;
rc = curl_url_get(h, CURLUPART_HOST, &host, 0);

char *scheme;
rc = curl_url_get(h, CURLUPART_SCHEME, &scheme, 0);

char *user;
rc = curl_url_get(h, CURLUPART_USER, &user, 0);

char *password;
rc = curl_url_get(h, CURLUPART_PASSWORD, &password, 0);

char *port;
rc = curl_url_get(h, CURLUPART_PORT, &port, 0);

char *path;
rc = curl_url_get(h, CURLUPART_PATH, &path, 0);

char *query;
rc = curl_url_get(h, CURLUPART_QUERY, &query, 0);

char *fragment;
rc = curl_url_get(h, CURLUPART_FRAGMENT, &fragment, 0);

char *zoneid;
rc = curl_url_get(h, CURLUPART_ZONEID, &zoneid, 0);

Remember to free the returned string with curl_free() when you are done with it.

Extracted parts are not URL decoded unless the user asks for it with the CURLU_URLDECODE flag.

URL parts

The different parts are named from their roles in the URL. Imagine a URL that looks like this:

http://joe:7Hbz@example.com:8080/images?id=5445#footer

When this URL is parsed by curl, it stores the different components like this:

Zone ID

The one thing that might stick out a little is the Zone id. It is an extra qualifier that can be used for IPv6 numerical addresses, and only for such addresses. It is used like this, where it is set to eth0:

http://[2a04:4e42:e00::347%25eth0]/

For this URL, curl extracts:

Asking for any other component returns non-zero as they are missing.

Empty parts

When asked to retrieve the query or the fragment from a URL, curl_url_get() by default treats those parts as missing when there is no content after the separator character itself.

For example if the URL looks like this:

http://example.com?#

By using the flag CURLU_GET_EMPTY, curl_url_get() instead returns a zero-length string to indicate that the separator was present but that there was no data immediately following it.

The path component is treated differently. It will always return at least "/", even if not present in the original URL (for example in https://example.com). It is therefore never missing.

When parsing a full URL, the hostname cannot be missing unless explicitly permitted with CURLU_NO_AUTHORITY. If this option is set, the host component may be missing.