Merge pull request #10269 from ClickHouse/path-optional-scheme

Continuation of #6256

src/Functions/URL/path.cpp

@@ -1,35 +1,15 @@
 #include <Functions/FunctionFactory.h>
 #include <Functions/FunctionStringToString.h>
 #include "FunctionsURL.h"
+#include "path.h"
 #include <common/find_symbols.h>
 
+
 namespace DB
 {
 
-struct ExtractPath
-{
-    static size_t getReserveLengthForElement() { return 25; }
-
-    static void execute(Pos data, size_t size, Pos & res_data, size_t & res_size)
-    {
-        res_data = data;
-        res_size = 0;
-
-        Pos pos = data;
-        Pos end = pos + size;
-
-        if (end != (pos = find_first_symbols<'/'>(pos, end)) && pos[1] == '/' && end != (pos = find_first_symbols<'/'>(pos + 2, end)))
-        {
-            Pos query_string_or_fragment = find_first_symbols<'?', '#'>(pos, end);
-
-            res_data = pos;
-            res_size = query_string_or_fragment - res_data;
-        }
-    }
-};
-
 struct NamePath { static constexpr auto name = "path"; };
-using FunctionPath = FunctionStringToString<ExtractSubstringImpl<ExtractPath>, NamePath>;
+using FunctionPath = FunctionStringToString<ExtractSubstringImpl<ExtractPath<false>>, NamePath>;
 
 void registerFunctionPath(FunctionFactory & factory)
 {

src/Functions/URL/path.h

@@ -0,0 +1,56 @@
+#pragma once
+
+#include <common/find_symbols.h>
+#include <Functions/URL/FunctionsURL.h>
+
+
+namespace DB
+{
+
+template <bool with_query_string>
+struct ExtractPath
+{
+    static size_t getReserveLengthForElement() { return 25; }
+
+    static void execute(Pos data, size_t size, Pos & res_data, size_t & res_size)
+    {
+        res_data = data;
+        res_size = 0;
+
+        Pos pos = data;
+        Pos end = pos + size;
+
+        /// We support URLs with and without schema:
+        /// 1. http://host/path
+        /// 2. host/path
+        /// We search for first slash and if there is subsequent slash, then skip and repeat search for the next slash.
+
+        pos = find_first_symbols<'/'>(pos, end);
+        if (end == pos)
+            return;
+
+        /// Note that strings are zero-terminated.
+        bool has_subsequent_slash = pos[1] == '/';
+        if (has_subsequent_slash)
+        {
+            /// Search for next slash.
+            pos = find_first_symbols<'/'>(pos + 2, end);
+            if (end == pos)
+                return;
+        }
+
+        res_data = pos;
+
+        if constexpr (with_query_string)
+        {
+            res_size = end - res_data;
+        }
+        else
+        {
+            Pos query_string_or_fragment = find_first_symbols<'?', '#'>(pos, end);
+            res_size = query_string_or_fragment - res_data;
+        }
+    }
+};
+
+}

src/Functions/URL/pathFull.cpp

@@ -1,33 +1,14 @@
 #include <Functions/FunctionFactory.h>
 #include <Functions/FunctionStringToString.h>
 #include "FunctionsURL.h"
+#include "path.h"
 #include <common/find_symbols.h>
 
 namespace DB
 {
 
-struct ExtractPathFull
-{
-    static size_t getReserveLengthForElement() { return 30; }
-
-    static void execute(const Pos data, const size_t size, Pos & res_data, size_t & res_size)
-    {
-        res_data = data;
-        res_size = 0;
-
-        Pos pos = data;
-        Pos end = pos + size;
-
-        if (end != (pos = find_first_symbols<'/'>(pos, end)) && pos[1] == '/' && end != (pos = find_first_symbols<'/'>(pos + 2, end)))
-        {
-            res_data = pos;
-            res_size = end - res_data;
-        }
-    }
-};
-
 struct NamePathFull { static constexpr auto name = "pathFull"; };
-using FunctionPathFull = FunctionStringToString<ExtractSubstringImpl<ExtractPathFull>, NamePathFull>;
+using FunctionPathFull = FunctionStringToString<ExtractSubstringImpl<ExtractPath<true>>, NamePathFull>;
 
 void registerFunctionPathFull(FunctionFactory & factory)
 {

tests/queries/0_stateless/01199_url_functions_path_without_schema_yiurule.reference

@@ -0,0 +1,2 @@
+/a/b/c
+/?query=hello world+foo+bar

tests/queries/0_stateless/01199_url_functions_path_without_schema_yiurule.sql

@@ -0,0 +1,2 @@
+SELECT path('www.example.com:443/a/b/c') AS Path;
+SELECT decodeURLComponent(materialize(pathFull('www.example.com/?query=hello%20world+foo%2Bbar'))) AS Path;