A lot of fixes and API changes

ext/dom/lexbor/lexbor/url/url.c

@@ -40,14 +40,6 @@ typedef enum {
 }
 lxb_url_map_type_t;
 
-typedef enum {
-    LXB_URL_HOST_OPT_UNDEF       = 0 << 0,
-    LXB_URL_HOST_OPT_NOT_SPECIAL = 1 << 0,
-    LXB_URL_HOST_OPT_DECODE      = 1 << 1,
-    LXB_URL_HOST_OPT_IDNA        = 1 << 2
-}
-lxb_url_host_opt_t;
-
 typedef struct {
     lexbor_mraw_t *mraw;
     lexbor_str_t  *str;
@@ -599,11 +591,6 @@ lxb_url_opaque_host_parse(lxb_url_parser_t *parser, const lxb_char_t *data,
                           const lxb_char_t *end, lxb_url_host_t *host,
                           lexbor_mraw_t *mraw);
 
-static lxb_status_t
-lxb_url_percent_decode(const lxb_char_t *data, const lxb_char_t *end,
-                       lexbor_str_t *str, lexbor_mraw_t *mraw,
-                       lxb_url_host_opt_t *opt);
-
 static const lxb_char_t *
 lxb_url_path_part_by_index(const lxb_url_t *url, size_t index,
                            size_t *out_length);
@@ -3961,7 +3948,7 @@ lxb_url_opaque_host_parse(lxb_url_parser_t *parser, const lxb_char_t *data,
                                               LXB_URL_MAP_C0, false);
 }
 
-static lxb_status_t
+lxb_status_t
 lxb_url_percent_decode(const lxb_char_t *data, const lxb_char_t *end,
                        lexbor_str_t *str, lexbor_mraw_t *mraw,
                        lxb_url_host_opt_t *opt)
@@ -4523,6 +4510,93 @@ lxb_url_serialize(const lxb_url_t *url, lexbor_serialize_cb_f cb, void *ctx,
     return LXB_STATUS_OK;
 }
 
+lxb_status_t
+lxb_url_serialize_unicode(lxb_unicode_idna_t *idna, const lxb_url_t *url, lexbor_serialize_cb_f cb, void *ctx,
+		                  bool exclude_fragment)
+{
+    lxb_status_t status;
+    const lexbor_str_t *str;
+    lxb_char_t *p;
+    lxb_char_t buf[LXB_URL_BUFFER_NUM_SIZE];
+
+    static const lexbor_str_t colon_str = lexbor_str(":");
+    static const lexbor_str_t dsol_str = lexbor_str("//");
+    static const lexbor_str_t at_str = lexbor_str("@");
+    static const lexbor_str_t dt_str = lexbor_str("/.");
+    static const lexbor_str_t qm_str = lexbor_str("?");
+    static const lexbor_str_t hs_str = lexbor_str("#");
+
+    /* Scheme. */
+
+    str = &url->scheme.name;
+
+    lexbor_serialize_write(cb, str->data, str->length, ctx, status);
+    lexbor_serialize_write(cb, colon_str.data, colon_str.length, ctx, status);
+
+    /* Host. */
+
+    if (url->host.type != LXB_URL_HOST_TYPE__UNDEF) {
+        lexbor_serialize_write(cb, dsol_str.data, dsol_str.length, ctx, status);
+
+        if (lxb_url_includes_credentials(url)) {
+            lexbor_serialize_write(cb, url->username.data, url->username.length,
+                                   ctx, status);
+
+            if (url->password.length != 0) {
+                lexbor_serialize_write(cb, colon_str.data, colon_str.length,
+                                       ctx, status);
+                lexbor_serialize_write(cb, url->password.data,
+                                       url->password.length, ctx, status);
+            }
+
+            lexbor_serialize_write(cb, at_str.data, at_str.length, ctx, status);
+        }
+
+        status = lxb_url_serialize_host_unicode(idna, &url->host, cb, ctx);
+        if (status != LXB_STATUS_OK) {
+            return status;
+        }
+
+        if (url->has_port) {
+            lexbor_serialize_write(cb, colon_str.data, colon_str.length,
+                                   ctx, status);
+
+            p = buf + lexbor_conv_int64_to_data((int64_t) url->port,
+                                                buf, LXB_URL_BUFFER_NUM_SIZE);
+
+            lexbor_serialize_write(cb, buf, p - buf, ctx, status);
+        }
+    }
+    else if (!url->path.opaque && url->path.str.length > 1) {
+        str = &url->path.str;
+
+        if (str->data[0] == '/' && str->data[1] == '/') {
+            lexbor_serialize_write(cb, dt_str.data, dt_str.length, ctx, status);
+        }
+    }
+
+    status = lxb_url_serialize_path(&url->path, cb, ctx);
+    if (status != LXB_STATUS_OK) {
+        return status;
+    }
+
+    if (url->query.data != NULL) {
+        lexbor_serialize_write(cb, qm_str.data, qm_str.length,
+                               ctx, status);
+        lexbor_serialize_write(cb, url->query.data, url->query.length,
+                               ctx, status);
+    }
+
+    if (!exclude_fragment && url->fragment.data != NULL) {
+        lexbor_serialize_write(cb, hs_str.data, hs_str.length,
+                               ctx, status);
+        lexbor_serialize_write(cb, url->fragment.data, url->fragment.length,
+                               ctx, status);
+    }
+
+    return LXB_STATUS_OK;
+}
+
 lxb_status_t
 lxb_url_serialize_scheme(const lxb_url_t *url,
                          lexbor_serialize_cb_f cb, void *ctx)

ext/dom/lexbor/lexbor/url/url.h

@@ -404,6 +404,10 @@ LXB_API lxb_status_t
 lxb_url_serialize(const lxb_url_t *url, lexbor_serialize_cb_f cb, void *ctx,
                   bool exclude_fragment);
 
+LXB_API lxb_status_t
+lxb_url_serialize_unicode(lxb_unicode_idna_t *idna, const lxb_url_t *url, lexbor_serialize_cb_f cb, void *ctx,
+						  bool exclude_fragment);
+
 LXB_API lxb_status_t
 lxb_url_serialize_scheme(const lxb_url_t *url,
                          lexbor_serialize_cb_f cb, void *ctx);
@@ -462,6 +466,23 @@ lxb_url_serialize_fragment(const lxb_url_t *url,
 LXB_API lxb_url_t *
 lxb_url_clone(lexbor_mraw_t *mraw, lxb_url_t *url);
 
+/*
+ * Below are auxiliary functions.
+ */
+
+typedef enum {
+	LXB_URL_HOST_OPT_UNDEF       = 0 << 0,
+	LXB_URL_HOST_OPT_NOT_SPECIAL = 1 << 0,
+	LXB_URL_HOST_OPT_DECODE      = 1 << 1,
+	LXB_URL_HOST_OPT_IDNA        = 1 << 2
+}
+lxb_url_host_opt_t;
+
+LXB_API lxb_status_t
+lxb_url_percent_decode(const lxb_char_t *data, const lxb_char_t *end,
+					   lexbor_str_t *str, lexbor_mraw_t *mraw,
+					   lxb_url_host_opt_t *opt);
+
 /*
  * Inline functions.
  */

ext/filter/logical_filters.c

@@ -620,14 +620,14 @@ void php_filter_validate_url(http://github.com/PHP_INPUT_FILTER_PARAM_DECL) /* {{{ */
 	}
 
 	zval scheme;
-	result = php_uri_get_scheme(internal_uri, &scheme);
+	result = php_uri_get_scheme(internal_uri, URI_COMPONENT_READ_RAW, &scheme);
 	if (result == FAILURE) {
 		php_uri_free(internal_uri);
 		RETURN_VALIDATION_FAILED
 	}
 
 	zval host;
-	result = php_uri_get_host(internal_uri, &host);
+	result = php_uri_get_host(internal_uri, URI_COMPONENT_READ_RAW, &host);
 	if (result == FAILURE) {
 		zval_ptr_dtor(&scheme);
 		php_uri_free(internal_uri);
@@ -669,7 +669,7 @@ void php_filter_validate_url(http://github.com/PHP_INPUT_FILTER_PARAM_DECL) /* {{{ */
 	}
 
 	zval path;
-	result = php_uri_get_path(internal_uri, &path);
+	result = php_uri_get_path(internal_uri, URI_COMPONENT_READ_RAW, &path);
 	if (result == FAILURE) {
 		php_uri_free(internal_uri);
 		zval_ptr_dtor(&scheme);
@@ -678,7 +678,7 @@ void php_filter_validate_url(http://github.com/PHP_INPUT_FILTER_PARAM_DECL) /* {{{ */
 	}
 
 	zval query;
-	result = php_uri_get_query(internal_uri, &query);
+	result = php_uri_get_query(internal_uri, URI_COMPONENT_READ_RAW, &query);
 	if (result == FAILURE) {
 		php_uri_free(internal_uri);
 		zval_ptr_dtor(&scheme);
@@ -707,14 +707,14 @@ void php_filter_validate_url(http://github.com/PHP_INPUT_FILTER_PARAM_DECL) /* {{{ */
 	zval_ptr_dtor(&query);
 
 	zval user;
-	result = php_uri_get_user(internal_uri, &user);
+	result = php_uri_get_user(internal_uri, URI_COMPONENT_READ_RAW, &user);
 	if (result == FAILURE) {
 		php_uri_free(internal_uri);
 		RETURN_VALIDATION_FAILED
 	}
 
 	zval password;
-	result = php_uri_get_password(internal_uri, &password);
+	result = php_uri_get_password(internal_uri, URI_COMPONENT_READ_RAW, &password);
 	if (result == FAILURE) {
 		php_uri_free(internal_uri);
 		zval_ptr_dtor(&user);

ext/openssl/xp_ssl.c

@@ -2627,7 +2627,7 @@ static char *php_openssl_get_url_name(const char *resourcename,
 	}
 
 	zval host_zv;
-	zend_result result = php_uri_get_host(internal_uri, &host_zv);
+	zend_result result = php_uri_get_host(internal_uri, URI_COMPONENT_READ_RAW, &host_zv);
 	if (result == SUCCESS && Z_TYPE(host_zv) == IS_STRING) {
 		const char * host = Z_STRVAL(host_zv);
 		char * url_name = NULL;