Add Utility/UriParser

[vharron]

Summary

I need a simple URI parser for parsing connect://hostname:port strings passed to "platform connect"

[clayborg]

Looks fine.

[vharron]

Changed path portion of sscanf to 2047 characters (2049 buffer length - terminating null - leading /)

[ovyalov]

LGTM

source/Utility/UriParser.cpp

37

indentation.

gtest/gtest.xcodeproj/project.pbxproj

 // !$*UTF8*$!
 {
   archiveVersion = 1;
   classes = {
   };
   objectVersion = 46;
   objects = {
 
 /* Begin PBXFileReference section */
     236ED33319D49076008CA7D7 /* ThreadStateCoordinatorTest.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = ThreadStateCoordinatorTest.cpp; sourceTree = "<group>"; };
     236ED33419D49081008CA7D7 /* Makefile */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.make; path = Makefile; sourceTree = "<group>"; };
     236ED33619D490B0008CA7D7 /* Makefile.rules */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = text; path = Makefile.rules; sourceTree = "<group>"; };
+    33064C981A5C7A1A0033D415 /* UriParserTest.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; name = UriParserTest.cpp; path = Utility/UriParserTest.cpp; sourceTree = "<group>"; };
+    33064C9D1A5C7AC90033D415 /* do-gtest.py */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = text.script.python; path = "do-gtest.py"; sourceTree = "<group>"; };
     338C47F41A1E67B900B46077 /* Makefile */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.make; name = Makefile; path = Utility/Makefile; sourceTree = "<group>"; };
     338C47F51A1E67B900B46077 /* StringExtractorTest.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; name = StringExtractorTest.cpp; path = Utility/StringExtractorTest.cpp; sourceTree = "<group>"; };
 /* End PBXFileReference section */
 
 /* Begin PBXGroup section */
     236ED32F19D4901D008CA7D7 /* unittest */ = {
       isa = PBXGroup;
       children = (
236ED33519D49098008CA7D7 /* make */ = {
         236ED33619D490B0008CA7D7 /* Makefile.rules */,
       );
       path = make;
       sourceTree = "<group>";
     };
     23CDD8EE19D4790700461DDC = {
       isa = PBXGroup;
       children = (
+        33064C9D1A5C7AC90033D415 /* do-gtest.py */,
         236ED33519D49098008CA7D7 /* make */,
         236ED32F19D4901D008CA7D7 /* unittest */,
       );
       sourceTree = "<group>";
     };
     338C47F31A1E677900B46077 /* Utility */ = {
       isa = PBXGroup;
       children = (
+        33064C981A5C7A1A0033D415 /* UriParserTest.cpp */,
         338C47F41A1E67B900B46077 /* Makefile */,
         338C47F51A1E67B900B46077 /* StringExtractorTest.cpp */,
       );
       name = Utility;
       sourceTree = "<group>";
     };
 /* End PBXGroup section */
 

gtest/unittest/Utility/Makefile

 THIS_FILE_DIR := $(shell dirname $(realpath $(lastword $(MAKEFILE_LIST))))/
 
 LEVEL := $(realpath $(THIS_FILE_DIR)../../make)
 
 CFLAGS_EXTRAS := -D__STDC_LIMIT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_CONSTANT_MACROS
 ENABLE_THREADS := YES
 CXX_SOURCES := $(wildcard *.cpp) \
-  $(realpath $(LEVEL)/../../source/Utility/StringExtractor.cpp)
+  $(realpath $(LEVEL)/../../source/Utility/StringExtractor.cpp) \
+  $(realpath $(LEVEL)/../../source/Utility/UriParser.cpp)
 MAKE_DSYM := NO
 
 OS := $(shell uname -s)
 # $(info OS $(OS))
 ifeq ($(OS),Linux)
     LD_EXTRAS := -lncurses -ldl
 endif
 
 include $(LEVEL)/Makefile.rules

gtest/unittest/Utility/UriParserTest.cpp

+#include "gtest/gtest.h"
+#include "Utility/UriParser.h"
+
+namespace
+{
+    class UriParserTest: public ::testing::Test
+    {
+    };
+}
+
+// result strings (scheme/hostname/port/path) passed into UriParser::Parse
+// are initialized to kAsdf so we can verify that they are unmodified if the 
+// URI is invalid
+static const char* kAsdf = "asdf";
+
+class UriTestCase
+{
+public:
+    UriTestCase(const char* uri, const char* scheme, const char* hostname, int port, const char* path) :
+      m_uri(uri),
+      m_result(true),
+      m_scheme(scheme),
+      m_hostname(hostname),
+      m_port(port),
+      m_path(path)
+    {
+    }
+
+    UriTestCase(const char* uri) :
+      m_uri(uri),
+      m_result(false),
+      m_scheme(kAsdf),
+      m_hostname(kAsdf),
+      m_port(1138),
+      m_path(kAsdf)
+    {
+    }
+
+    const char* m_uri;
+    bool        m_result;
+    const char* m_scheme;
+    const char* m_hostname;
+    int         m_port;
+    const char* m_path;
+};
+
+#define VALIDATE \
+    std::string scheme(kAsdf); \
+    std::string hostname(kAsdf); \
+    int port(1138); \
+    std::string path(kAsdf); \
+    EXPECT_EQ (testCase.m_result, UriParser::Parse(testCase.m_uri, scheme, hostname, port, path)); \
+    EXPECT_STREQ (testCase.m_scheme, scheme.c_str()); \
+    EXPECT_STREQ (testCase.m_hostname, hostname.c_str()); \
+    EXPECT_EQ (testCase.m_port, port); \
+    EXPECT_STREQ (testCase.m_path, path.c_str());
+
+TEST_F (UriParserTest, Minimal)
+{
+    const UriTestCase testCase("x://y", "x", "y", 0, "/");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, MinimalPort)
+{
+    const UriTestCase testCase("x://y:1", "x", "y", 1, "/");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, MinimalPath)
+{
+    const UriTestCase testCase("x://y/", "x", "y", 0, "/");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, MinimalPortPath)
+{
+    const UriTestCase testCase("x://y:1/", "x", "y", 1, "/");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, TypicalPortPath)
+{
+    const UriTestCase testCase("connect://192.168.100.132:5432/", "connect", "192.168.100.132", 5432, "/");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, SchemeHostSeparator)
+{
+    const UriTestCase testCase("x:/y");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, SchemeHostSeparator2)
+{
+    const UriTestCase testCase("x:y");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, SchemeHostSeparator3)
+{
+    const UriTestCase testCase("x//y");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, SchemeHostSeparator4)
+{
+    const UriTestCase testCase("x/y");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, BadPort)
+{
+    const UriTestCase testCase("x://y:a/");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, BadPort2)
+{
+    const UriTestCase testCase("x://y:5432a/");
+    VALIDATE
+}
+
+TEST_F (UriParserTest, Empty)
+{
+    const UriTestCase testCase("");
+    VALIDATE
+}
+

lldb.xcodeproj/project.pbxproj

/* Begin PBXBuildFile section */
     26F5C32D10F3DFDD009D5894 /* libtermcap.dylib in Frameworks */ = {isa = PBXBuildFile; fileRef = 26F5C32B10F3DFDD009D5894 /* libtermcap.dylib */; };
     26F73062139D8FDB00FD51C7 /* History.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 26F73061139D8FDB00FD51C7 /* History.cpp */; };
     26FFC19914FC072100087D58 /* AuxVector.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 26FFC19314FC072100087D58 /* AuxVector.cpp */; };
     26FFC19A14FC072100087D58 /* AuxVector.h in Headers */ = {isa = PBXBuildFile; fileRef = 26FFC19414FC072100087D58 /* AuxVector.h */; };
     26FFC19B14FC072100087D58 /* DYLDRendezvous.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 26FFC19514FC072100087D58 /* DYLDRendezvous.cpp */; };
     26FFC19C14FC072100087D58 /* DYLDRendezvous.h in Headers */ = {isa = PBXBuildFile; fileRef = 26FFC19614FC072100087D58 /* DYLDRendezvous.h */; };
     26FFC19D14FC072100087D58 /* DynamicLoaderPOSIXDYLD.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 26FFC19714FC072100087D58 /* DynamicLoaderPOSIXDYLD.cpp */; };
     26FFC19E14FC072100087D58 /* DynamicLoaderPOSIXDYLD.h in Headers */ = {isa = PBXBuildFile; fileRef = 26FFC19814FC072100087D58 /* DynamicLoaderPOSIXDYLD.h */; };
+    33064C9A1A5C7A330033D415 /* UriParser.cpp in CopyFiles */ = {isa = PBXBuildFile; fileRef = 33064C991A5C7A330033D415 /* UriParser.cpp */; };
+    33064C9C1A5C7A490033D415 /* UriParser.h in CopyFiles */ = {isa = PBXBuildFile; fileRef = 33064C9B1A5C7A490033D415 /* UriParser.h */; };
     3FDFDDBD199C3A06009756A7 /* FileAction.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 3FDFDDBC199C3A06009756A7 /* FileAction.cpp */; };
     3FDFDDBF199D345E009756A7 /* FileCache.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 3FDFDDBE199D345E009756A7 /* FileCache.cpp */; };
     3FDFDDC6199D37ED009756A7 /* FileSystem.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 3FDFDDC5199D37ED009756A7 /* FileSystem.cpp */; };
     3FDFE52C19A2917A009756A7 /* HostInfoMacOSX.mm in Sources */ = {isa = PBXBuildFile; fileRef = 3FDFE52B19A2917A009756A7 /* HostInfoMacOSX.mm */; };
     3FDFE53119A292F0009756A7 /* HostInfoPosix.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 3FDFE53019A292F0009756A7 /* HostInfoPosix.cpp */; };
     3FDFE53319A29304009756A7 /* HostInfoPosix.h in Headers */ = {isa = PBXBuildFile; fileRef = 3FDFE53219A29304009756A7 /* HostInfoPosix.h */; };
     3FDFE53519A29327009756A7 /* HostInfoBase.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 3FDFE53419A29327009756A7 /* HostInfoBase.cpp */; };
     3FDFE56C19AF9C44009756A7 /* HostProcessPosix.cpp in Sources */ = {isa = PBXBuildFile; fileRef = 3FDFE56A19AF9C44009756A7 /* HostProcessPosix.cpp */; };
/* Begin PBXCopyFilesBuildPhase section */
     };
     AF90106415AB7D2900FF120D /* CopyFiles */ = {
       isa = PBXCopyFilesBuildPhase;
       buildActionMask = 8;
       dstPath = "$(DEVELOPER_DIR)/usr/share/man/man1/";
       dstSubfolderSpec = 0;
       files = (
         AF90106515AB7D3600FF120D /* lldb.1 in CopyFiles */,
+        33064C9C1A5C7A490033D415 /* UriParser.h in CopyFiles */,
+        33064C9A1A5C7A330033D415 /* UriParser.cpp in CopyFiles */,
       );
       runOnlyForDeploymentPostprocessing = 1;
     };
     AFF87C85150FF5CC000E1742 /* CopyFiles */ = {
       isa = PBXCopyFilesBuildPhase;
       buildActionMask = 8;
       dstPath = /Developer/Library/Lockdown/ServiceAgents/;
       dstSubfolderSpec = 0;
/* Begin PBXFileReference section */
     26FA4315130103F400E71120 /* FileSpec.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; name = FileSpec.h; path = include/lldb/Host/FileSpec.h; sourceTree = "<group>"; };
     26FA43171301048600E71120 /* FileSpec.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = FileSpec.cpp; sourceTree = "<group>"; };
     26FFC19314FC072100087D58 /* AuxVector.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = AuxVector.cpp; sourceTree = "<group>"; };
     26FFC19414FC072100087D58 /* AuxVector.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = AuxVector.h; sourceTree = "<group>"; };
     26FFC19514FC072100087D58 /* DYLDRendezvous.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = DYLDRendezvous.cpp; sourceTree = "<group>"; };
     26FFC19614FC072100087D58 /* DYLDRendezvous.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = DYLDRendezvous.h; sourceTree = "<group>"; };
     26FFC19714FC072100087D58 /* DynamicLoaderPOSIXDYLD.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = DynamicLoaderPOSIXDYLD.cpp; sourceTree = "<group>"; };
     26FFC19814FC072100087D58 /* DynamicLoaderPOSIXDYLD.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = DynamicLoaderPOSIXDYLD.h; sourceTree = "<group>"; };
+    33064C991A5C7A330033D415 /* UriParser.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; name = UriParser.cpp; path = source/Utility/UriParser.cpp; sourceTree = "<group>"; };
+    33064C9B1A5C7A490033D415 /* UriParser.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; name = UriParser.h; path = source/Utility/UriParser.h; sourceTree = "<group>"; };
     3F5E8AF31A40D4A500A73232 /* PipeBase.h */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.c.h; name = PipeBase.h; path = include/lldb/Host/PipeBase.h; sourceTree = "<group>"; };
     3FDFD6C3199C396E009756A7 /* FileAction.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; name = FileAction.h; path = include/lldb/Target/FileAction.h; sourceTree = "<group>"; };
     3FDFDDBC199C3A06009756A7 /* FileAction.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; name = FileAction.cpp; path = source/Target/FileAction.cpp; sourceTree = "<group>"; };
     3FDFDDBE199D345E009756A7 /* FileCache.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; name = FileCache.cpp; path = source/Host/common/FileCache.cpp; sourceTree = "<group>"; };
     3FDFDDC0199D34E2009756A7 /* FileCache.h */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.c.h; name = FileCache.h; path = include/lldb/Host/FileCache.h; sourceTree = "<group>"; };
     3FDFDDC1199D34E2009756A7 /* FileSystem.h */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.c.h; name = FileSystem.h; path = include/lldb/Host/FileSystem.h; sourceTree = "<group>"; };
     3FDFDDC5199D37ED009756A7 /* FileSystem.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = FileSystem.cpp; sourceTree = "<group>"; };
     3FDFE52B19A2917A009756A7 /* HostInfoMacOSX.mm */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.objcpp; name = HostInfoMacOSX.mm; path = source/Host/macosx/HostInfoMacOSX.mm; sourceTree = "<group>"; };
266DFE9013FD64D200D0C574 /* OperatingSystem */ = {
         2698699715E6CBD0002415FF /* Python */,
       );
       path = OperatingSystem;
       sourceTree = "<group>";
     };
     2682F168115ED9C800CCFF99 /* Utility */ = {
       isa = PBXGroup;
       children = (
+        33064C9B1A5C7A490033D415 /* UriParser.h */,
+        33064C991A5C7A330033D415 /* UriParser.cpp */,
         AF9107E91685709A00DBCD3C /* ARM64_GCC_Registers.h */,
         26CF992414428766001E4138 /* AnsiTerminal.h */,
         26F996A7119B79C300412154 /* ARM_DWARF_Registers.h */,
         26ECA04213665FED008D1F18 /* ARM_DWARF_Registers.cpp */,
         264A12FF137252C700875C42 /* ARM64_DWARF_Registers.h */,
         264A12FE137252C700875C42 /* ARM64_DWARF_Registers.cpp */,
         26F996A8119B79C300412154 /* ARM_GCC_Registers.h */,
         264723A511FA076E00DE380C /* CleanUp.h */,

source/Utility/CMakeLists.txt

 set(LLVM_NO_RTTI 1)
 
 add_lldb_library(lldbUtility
   ARM_DWARF_Registers.cpp
   ARM64_DWARF_Registers.cpp
   KQueue.cpp
   PseudoTerminal.cpp
   Range.cpp
   SharingPtr.cpp
   StringExtractor.cpp
   StringExtractorGDBRemote.cpp
   StringLexer.cpp
   TimeSpecTimeout.cpp
+  UriParser.cpp
   )

source/Utility/UriParser.h

+//===-- UriParser.h ---------------------------------------------*- C++ -*-===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef utility_UriParser_h_
+#define utility_UriParser_h_
+
+// C Includes
+// C++ Includes
+#include <string>
+
+// Other libraries and framework includes
+// Project includes
+
+class UriParser
+{
+public:
+    static bool Parse(const char* uri,
+        std::string& scheme,
+        std::string& hostname,
+        int& port,
+        std::string& path
+        );
+};
+
+#endif  // utility_UriParser_h_

source/Utility/UriParser.cpp

+//===-- UriParser.cpp -------------------------------------------*- C++ -*-===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "Utility/UriParser.h"
+
+// C Includes
+#include <stdlib.h>
+
+// C++ Includes
+// Other libraries and framework includes
+// Project includes
+
+//----------------------------------------------------------------------
+// UriParser::Parse
+//----------------------------------------------------------------------
+bool
+UriParser::Parse(const char* uri,
+    std::string& scheme,
+    std::string& hostname,
+    int& port,
+    std::string& path
+    )
+{
+    char scheme_buf[100] = {0};
+    char hostname_buf[256] = {0};
+    char port_buf[11] = {0}; // 10==strlen(2^32)
+    char path_buf[2049] = {'/', 0};
+  
+    bool ok = false;
+         if (4==sscanf(uri, "%99[^:/]://%255[^/:]:%[^/]/%2048s", scheme_buf, hostname_buf, port_buf, path_buf+1)) { ok = true; }
+    else if (3==sscanf(uri, "%99[^:/]://%255[^/:]:%[^/]", scheme_buf, hostname_buf, port_buf)) { ok = true; }
+    else if (3==sscanf(uri, "%99[^:/]://%255[^/]/%2048s", scheme_buf, hostname_buf, path_buf+1)) { ok = true; }
+    else if (2==sscanf(uri, "%99[^:/]://%255[^/]", scheme_buf, hostname_buf)) { ok = true; }
+
+    char* end = port_buf;
+    int port_tmp = strtoul(port_buf, &end, 10);
+    if (*end != 0)
+    {
+        // there are invalid characters in port_buf
+        return false;
+    }
+
+    if (ok)
+    {
+        scheme.assign(scheme_buf);
+        hostname.assign(hostname_buf);
+        port = port_tmp;
+        path.assign(path_buf);
+    }
+    return ok;
+}
+