urlapi: CURLU_NO_AUTHORITY allows empty authority/host part

CURLU_NO_AUTHORITY is intended for use with unknown schemes (i.e. not
"file:///") to override cURL's default demand that an authority exists.

Closes #4349

5 files changed, 57 insertions, 11 deletions

diff --git a/docs/libcurl/curl_url_set.3 b/docs/libcurl/curl_url_set.3
index 43d741220..e2e2c1887 100644
--- a/docs/libcurl/curl_url_set.3
+++ b/docs/libcurl/curl_url_set.3
@@ -111,6 +111,12 @@ instead "guesses" which scheme that was intended based on the host name.  If
 the outermost sub-domain name matches DICT, FTP, IMAP, LDAP, POP3 or SMTP then
 that scheme will be used, otherwise it picks HTTP. Conflicts with the
 \fICURLU_DEFAULT_SCHEME\fP option which takes precedence if both are set.
+.IP CURLU_NO_AUTHORITY
+If set, skips authority checks. The RFC allows individual schemes to omit the
+host part (normally the only mandatory part of the authority), but libcurl
+cannot know whether this is permitted for custom schemes. Specifying the flag
+permits empty authority sections, similar to how file scheme is handled.
+
 .SH RETURN VALUE
 Returns a CURLUcode error value, which is CURLUE_OK (0) if everything went
 fine.
diff --git a/docs/libcurl/symbols-in-versions b/docs/libcurl/symbols-in-versions
index 9daad949f..ff0dfb238 100644
--- a/docs/libcurl/symbols-in-versions
+++ b/docs/libcurl/symbols-in-versions
@@ -779,6 +779,7 @@ CURLU_DISALLOW_USER             7.62.0
 CURLU_GUESS_SCHEME              7.62.0
 CURLU_NON_SUPPORT_SCHEME        7.62.0
 CURLU_NO_DEFAULT_PORT           7.62.0
+CURLU_NO_AUTHORITY              7.67.0
 CURLU_PATH_AS_IS                7.62.0
 CURLU_URLDECODE                 7.62.0
 CURLU_URLENCODE                 7.62.0
diff --git a/include/curl/urlapi.h b/include/curl/urlapi.h
index 0f2f152f1..f2d06770d 100644
--- a/include/curl/urlapi.h
+++ b/include/curl/urlapi.h
@@ -77,6 +77,8 @@ typedef enum {
 #define CURLU_URLENCODE (1<<7)          /* URL encode on set */
 #define CURLU_APPENDQUERY (1<<8)        /* append a form style part */
 #define CURLU_GUESS_SCHEME (1<<9)       /* legacy curl-style guessing */
+#define CURLU_NO_AUTHORITY (1<<10)      /* Allow empty authority when the
+                                           scheme is unknown. */
 
 typedef struct Curl_URL CURLU;
 
diff --git a/lib/urlapi.c b/lib/urlapi.c
index a0ee331da..6e414cf18 100644
--- a/lib/urlapi.c
+++ b/lib/urlapi.c
@@ -784,6 +784,7 @@ static CURLUcode seturl(const char *url, CURLU *u, unsigned int flags)
 
       if(junkscan(schemep))
         return CURLUE_MALFORMED_INPUT;
+
     }
     else {
       /* no scheme! */
@@ -804,11 +805,14 @@ static CURLUcode seturl(const char *url, CURLU *u, unsigned int flags)
       p++;
 
     len = p - hostp;
-    if(!len)
-      return CURLUE_MALFORMED_INPUT;
-
-    memcpy(hostname, hostp, len);
-    hostname[len] = 0;
+    if(len) {
+      memcpy(hostname, hostp, len);
+      hostname[len] = 0;
+    }
+    else {
+      if(!(flags & CURLU_NO_AUTHORITY))
+        return CURLUE_MALFORMED_INPUT;
+    }
 
     if((flags & CURLU_GUESS_SCHEME) && !schemep) {
       /* legacy curl-style guess based on host name */
@@ -889,9 +893,14 @@ static CURLUcode seturl(const char *url, CURLU *u, unsigned int flags)
     if(result)
       return result;
 
-    result = hostname_check(u, hostname);
-    if(result)
-      return result;
+    if(0 == strlen(hostname) && (flags & CURLU_NO_AUTHORITY)) {
+      /* Skip hostname check, it's allowed to be empty. */
+    }
+    else {
+      result = hostname_check(u, hostname);
+      if(result)
+        return result;
+    }
 
     u->host = strdup(hostname);
     if(!u->host)
@@ -1432,9 +1441,14 @@ CURLUcode curl_url_set(CURLU *u, CURLUPart what,
     }
 
     if(what == CURLUPART_HOST) {
-      if(hostname_check(u, (char *)newp)) {
-        free((char *)newp);
-        return CURLUE_MALFORMED_INPUT;
+      if(0 == strlen(newp) && (flags & CURLU_NO_AUTHORITY)) {
+        /* Skip hostname check, it's allowed to be empty. */
+      }
+      else {
+        if(hostname_check(u, (char *)newp)) {
+          free((char *)newp);
+          return CURLUE_MALFORMED_INPUT;
+        }
       }
     }
 
diff --git a/tests/libtest/lib1560.c b/tests/libtest/lib1560.c
index 85884474e..d161423e3 100644
--- a/tests/libtest/lib1560.c
+++ b/tests/libtest/lib1560.c
@@ -414,6 +414,18 @@ static struct urltestcase get_url_list[] = {
   {"tp://example.com/path/html",
    "tp://example.com/path/html",
    CURLU_NON_SUPPORT_SCHEME, 0, CURLUE_OK},
+  {"custom-scheme://host?expected=test-good",
+   "custom-scheme://host/?expected=test-good",
+   CURLU_NON_SUPPORT_SCHEME, 0, CURLUE_OK},
+  {"custom-scheme://?expected=test-bad",
+   "",
+   CURLU_NON_SUPPORT_SCHEME, 0, CURLUE_MALFORMED_INPUT},
+  {"custom-scheme://?expected=test-new-good",
+   "custom-scheme:///?expected=test-new-good",
+   CURLU_NON_SUPPORT_SCHEME | CURLU_NO_AUTHORITY, 0, CURLUE_OK},
+  {"custom-scheme://host?expected=test-still-good",
+   "custom-scheme://host/?expected=test-still-good",
+   CURLU_NON_SUPPORT_SCHEME | CURLU_NO_AUTHORITY, 0, CURLUE_OK},
   {NULL, NULL, 0, 0, 0}
 };
 
@@ -551,6 +563,17 @@ static struct setcase set_parts_list[] = {
    "scheme=ftp,",
    "ftp://example.com:80/",
    0, 0, CURLUE_OK, CURLUE_OK},
+  {"custom-scheme://host",
+   "host=\"\",",
+   "custom-scheme://host/",
+   CURLU_NON_SUPPORT_SCHEME, CURLU_NON_SUPPORT_SCHEME, CURLUE_OK,
+   CURLUE_MALFORMED_INPUT},
+  {"custom-scheme://host",
+   "host=\"\",",
+   "custom-scheme:///",
+   CURLU_NON_SUPPORT_SCHEME, CURLU_NON_SUPPORT_SCHEME | CURLU_NO_AUTHORITY,
+   CURLUE_OK, CURLUE_OK},
+
   {NULL, NULL, NULL, 0, 0, 0, 0}
 };