X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/4860d40db4a78d8c2fd20d0b7b75136d749050eb..67abc78356eca7b033aed920a6852b226fe42438:/include/wx/uri.h diff --git a/include/wx/uri.h b/include/wx/uri.h index a2a16b2421..6b9b9ace15 100644 --- a/include/wx/uri.h +++ b/include/wx/uri.h @@ -1,31 +1,30 @@ ///////////////////////////////////////////////////////////////////////////// -// Name: uri.h -// Purpose: wxURI - Class for parsing URIs +// Name: wx/uri.h +// Purpose: wxURI - Class for parsing URIs // Author: Ryan Norton -// Modified By: +// Vadim Zeitlin (UTF-8 URI support, many other changes) // Created: 07/01/2004 // RCS-ID: $Id$ -// Licence: wxWindows +// Copyright: (c) 2004 Ryan Norton +// 2008 Vadim Zeitlin +// Licence: wxWindows Licence ///////////////////////////////////////////////////////////////////////////// #ifndef _WX_URI_H_ #define _WX_URI_H_ -#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA) -#pragma interface "uri.h" -#endif - #include "wx/defs.h" #include "wx/object.h" #include "wx/string.h" +#include "wx/arrstr.h" // Host Type that the server component can be enum wxURIHostType { - wxURI_REGNAME, // Host is a normal register name (www.mysite.com etc.) - wxURI_IPV4ADDRESS, // Host is a version 4 ip address (192.168.1.100) + wxURI_REGNAME, // Host is a normal register name (www.mysite.com etc.) + wxURI_IPV4ADDRESS, // Host is a version 4 ip address (192.168.1.100) wxURI_IPV6ADDRESS, // Host is a version 6 ip address [aa:aa:aa:aa::aa:aa]:5050 - wxURI_IPVFUTURE // Host is a future ip address (wxURI is unsure what kind) + wxURI_IPVFUTURE // Host is a future ip address (wxURI is unsure what kind) }; // Component Flags @@ -49,92 +48,125 @@ enum wxURIFlags // Generic class for parsing URIs. // -// Originally based off of RFC 2396 - then -// extended to meet the newer RFC 2396.bis -// specifications. +// See RFC 3986 class WXDLLIMPEXP_BASE wxURI : public wxObject { public: wxURI(); wxURI(const wxString& uri); - wxURI(const wxURI& uri); - - virtual ~wxURI(); - - const wxChar* Create(const wxString& uri); - - bool HasScheme() const { return (m_fields & wxURI_SCHEME) == wxURI_SCHEME; } - bool HasUserInfo() const { return (m_fields & wxURI_USERINFO) == wxURI_USERINFO; } - bool HasServer() const { return (m_fields & wxURI_SERVER) == wxURI_SERVER; } - bool HasPort() const { return (m_fields & wxURI_PORT) == wxURI_PORT; } - bool HasPath() const { return (m_fields & wxURI_PATH) == wxURI_PATH; } - bool HasQuery() const { return (m_fields & wxURI_QUERY) == wxURI_QUERY; } - bool HasFragment() const { return (m_fields & wxURI_FRAGMENT) == wxURI_FRAGMENT; } - - const wxString& GetScheme() const { return m_scheme; } - const wxString& GetPath() const { return m_path; } - const wxString& GetQuery() const { return m_query; } - const wxString& GetFragment() const { return m_fragment; } - const wxString& GetPort() const { return m_port; } - const wxString& GetUserInfo() const { return m_userinfo; } - const wxString& GetServer() const { return m_server; } - const wxURIHostType& GetHostType() const { return m_hostType; } - - //Note that the following two get functions are explicitly depreciated by RFC 2396 + + // default copy ctor, assignment operator and dtor are ok + + bool Create(const wxString& uri); + + wxURI& operator=(const wxString& string) + { + Create(string); + return *this; + } + + bool operator==(const wxURI& uri) const; + + // various accessors + + bool HasScheme() const { return (m_fields & wxURI_SCHEME) != 0; } + bool HasUserInfo() const { return (m_fields & wxURI_USERINFO) != 0; } + bool HasServer() const { return (m_fields & wxURI_SERVER) != 0; } + bool HasPort() const { return (m_fields & wxURI_PORT) != 0; } + bool HasPath() const { return (m_fields & wxURI_PATH) != 0; } + bool HasQuery() const { return (m_fields & wxURI_QUERY) != 0; } + bool HasFragment() const { return (m_fields & wxURI_FRAGMENT) != 0; } + + const wxString& GetScheme() const { return m_scheme; } + const wxString& GetPath() const { return m_path; } + const wxString& GetQuery() const { return m_query; } + const wxString& GetFragment() const { return m_fragment; } + const wxString& GetPort() const { return m_port; } + const wxString& GetUserInfo() const { return m_userinfo; } + const wxString& GetServer() const { return m_server; } + wxURIHostType GetHostType() const { return m_hostType; } + + // these functions only work if the user information part of the URI is in + // the usual (but insecure and hence explicitly recommended against by the + // RFC) "user:password" form wxString GetUser() const; wxString GetPassword() const; - - wxString BuildURI() const; - wxString BuildUnescapedURI() const; + + + // combine all URI components into a single string + // + // BuildURI() returns the real URI suitable for use with network libraries, + // for example, while BuildUnescapedURI() returns a string suitable to be + // shown to the user. + wxString BuildURI() const { return DoBuildURI(&wxURI::Nothing); } + wxString BuildUnescapedURI() const { return DoBuildURI(&wxURI::Unescape); } + + // the escaped URI should contain only ASCII characters, including possible + // escape sequences + static wxString Unescape(const wxString& escapedURI); + void Resolve(const wxURI& base, int flags = wxURI_STRICT); bool IsReference() const; - wxURI& operator = (const wxURI& uri); - wxURI& operator = (const wxString& string); - bool operator == (const wxURI& uri) const; - - static wxString Unescape (const wxString& szEscapedURI); - protected: - wxURI& Assign(const wxURI& uri); - void Clear(); - const wxChar* Parse (const wxChar* uri); - const wxChar* ParseAuthority (const wxChar* uri); - const wxChar* ParseScheme (const wxChar* uri); - const wxChar* ParseUserInfo (const wxChar* uri); - const wxChar* ParseServer (const wxChar* uri); - const wxChar* ParsePort (const wxChar* uri); - const wxChar* ParsePath (const wxChar* uri, - bool bReference = false, - bool bNormalize = true); - const wxChar* ParseQuery (const wxChar* uri); - const wxChar* ParseFragment (const wxChar* uri); - - - static bool ParseH16(const wxChar*& uri); - static bool ParseIPv4address(const wxChar*& uri); - static bool ParseIPv6address(const wxChar*& uri); - static bool ParseIPvFuture(const wxChar*& uri); - - static void Normalize(wxChar* uri, bool bIgnoreLeads = false); - static void UpTree(const wxChar* uristart, const wxChar*& uri); - - static wxChar TranslateEscape(const wxChar* s); - static void Escape (wxString& s, const wxChar& c); - static bool IsEscape(const wxChar*& uri); - - static wxChar CharToHex(const wxChar& c); - - static bool IsUnreserved (const wxChar& c); - static bool IsReserved (const wxChar& c); - static bool IsGenDelim (const wxChar& c); - static bool IsSubDelim (const wxChar& c); - static bool IsHex(const wxChar& c); - static bool IsAlpha(const wxChar& c); - static bool IsDigit(const wxChar& c); + // common part of BuildURI() and BuildUnescapedURI() + wxString DoBuildURI(wxString (*funcDecode)(const wxString&)) const; + + // function which returns its argument unmodified, this is used by + // BuildURI() to tell DoBuildURI() that nothing needs to be done with the + // URI components + static wxString Nothing(const wxString& value) { return value; } + + bool Parse(const char* uri); + + const char* ParseAuthority (const char* uri); + const char* ParseScheme (const char* uri); + const char* ParseUserInfo (const char* uri); + const char* ParseServer (const char* uri); + const char* ParsePort (const char* uri); + const char* ParsePath (const char* uri); + const char* ParseQuery (const char* uri); + const char* ParseFragment (const char* uri); + + + static bool ParseH16(const char*& uri); + static bool ParseIPv4address(const char*& uri); + static bool ParseIPv6address(const char*& uri); + static bool ParseIPvFuture(const char*& uri); + + // should be called with i pointing to '%', returns the encoded character + // following it or -1 if invalid and advances i past it (so that it points + // to the last character consumed on return) + static int DecodeEscape(wxString::const_iterator& i); + + // append next character pointer to by p to the string in an escaped form + // and advance p past it + // + // if the next character is '%' and it's followed by 2 hex digits, they are + // not escaped (again) by this function, this allows to keep (backwards- + // compatible) ambiguity about the input format to wxURI::Create(): it can + // be either already escaped or not + void AppendNextEscaped(wxString& s, const char *& p); + + // convert hexadecimal digit to its value; return -1 if c isn't valid + static int CharToHex(char c); + + // split an URI path string in its component segments (including empty and + // "." ones, no post-processing is done) + static wxArrayString SplitInSegments(const wxString& path); + + // various URI grammar helpers + static bool IsUnreserved(char c); + static bool IsReserved(char c); + static bool IsGenDelim(char c); + static bool IsSubDelim(char c); + static bool IsHex(char c); + static bool IsAlpha(char c); + static bool IsDigit(char c); + static bool IsEndPath(char c); wxString m_scheme; wxString m_path;