WIP local deb install

[apt.git] / apt-pkg / contrib / strutl.cc

diff --git a/apt-pkg/contrib/strutl.cc b/apt-pkg/contrib/strutl.cc
index 2e2bb5ebcdfdfb985c427d2f16d403730b9f67aa..2100ee47b3c84dcc8efe313724b12a312239f7ee 100644 (file)
--- a/apt-pkg/contrib/strutl.cc
+++ b/apt-pkg/contrib/strutl.cc
@@ -15,14 +15,20 @@
    ##################################################################### */
                                                                        /*}}}*/
 // Includes                                                            /*{{{*/
    ##################################################################### */
                                                                        /*}}}*/
 // Includes                                                            /*{{{*/

+#include <config.h>
+

 #include <apt-pkg/strutl.h>
 #include <apt-pkg/fileutl.h>
 #include <apt-pkg/error.h>
 
 #include <apt-pkg/strutl.h>
 #include <apt-pkg/fileutl.h>
 #include <apt-pkg/error.h>
 

-#include <apti18n.h>
-    
+#include <stddef.h>
+#include <stdlib.h>
+#include <time.h>
+#include <string>
+#include <vector>

 #include <ctype.h>
 #include <string.h>
 #include <ctype.h>
 #include <string.h>

+#include <sstream>

 #include <stdio.h>
 #include <algorithm>
 #include <unistd.h>
 #include <stdio.h>
 #include <algorithm>
 #include <unistd.h>


@@ -31,11 +37,34 @@
 #include <stdarg.h>
 #include <iconv.h>
 
 #include <stdarg.h>
 #include <iconv.h>
 

-#include "config.h"
-
-using namespace std;
+#include <apti18n.h>

                                                                        /*}}}*/
                                                                        /*}}}*/

+using namespace std;

 
 

+// Strip - Remove white space from the front and back of a string       /*{{{*/
+// ---------------------------------------------------------------------
+namespace APT {
+   namespace String {
+std::string Strip(const std::string &s)
+{
+   size_t start = s.find_first_not_of(" \t\n");
+   // only whitespace
+   if (start == string::npos)
+      return "";
+   size_t end = s.find_last_not_of(" \t\n");
+   return s.substr(start, end-start+1);
+}
+
+bool Endswith(const std::string &s, const std::string &end)
+{
+   if (end.size() > s.size())
+      return false;
+   return (s.substr(s.size() - end.size(), s.size()) == end);
+}
+
+}
+}
+                                                                       /*}}}*/

 // UTF8ToCodeset - Convert some UTF-8 string for some codeset          /*{{{*/
 // ---------------------------------------------------------------------
 /* This is handy to use before display some information for enduser  */
 // UTF8ToCodeset - Convert some UTF-8 string for some codeset          /*{{{*/
 // ---------------------------------------------------------------------
 /* This is handy to use before display some information for enduser  */


@@ -116,14 +145,20 @@ char *_strstrip(char *String)
 
    if (*String == 0)
       return String;
 
    if (*String == 0)
       return String;

-
+   return _strrstrip(String);
+}
+                                                                       /*}}}*/
+// strrstrip - Remove white space from the back of a string    /*{{{*/
+// ---------------------------------------------------------------------
+char *_strrstrip(char *String)
+{

    char *End = String + strlen(String) - 1;
    for (;End != String - 1 && (*End == ' ' || *End == '\t' || *End == '\n' ||
                               *End == '\r'); End--);
    End++;
    *End = 0;
    return String;
    char *End = String + strlen(String) - 1;
    for (;End != String - 1 && (*End == ' ' || *End == '\t' || *End == '\n' ||
                               *End == '\r'); End--);
    End++;
    *End = 0;
    return String;

-};
+}

                                                                        /*}}}*/
 // strtabexpand - Converts tabs into 8 spaces                          /*{{{*/
 // ---------------------------------------------------------------------
                                                                        /*}}}*/
 // strtabexpand - Converts tabs into 8 spaces                          /*{{{*/
 // ---------------------------------------------------------------------


@@ -179,14 +214,14 @@ bool ParseQuoteWord(const char *&String,string &Res)
    {
       if (*C == '"')
       {
    {
       if (*C == '"')
       {

-        for (C++; *C != 0 && *C != '"'; C++);
-        if (*C == 0)
+        C = strchr(C + 1, '"');
+        if (C == NULL)

            return false;
       }
       if (*C == '[')
       {
            return false;
       }
       if (*C == '[')
       {

-        for (C++; *C != 0 && *C != ']'; C++);
-        if (*C == 0)
+        C = strchr(C + 1, ']');
+        if (C == NULL)

            return false;
       }
    }
            return false;
       }
    }


@@ -271,7 +306,7 @@ bool ParseCWord(const char *&String,string &Res)
 string QuoteString(const string &Str, const char *Bad)
 {
    string Res;
 string QuoteString(const string &Str, const char *Bad)
 {
    string Res;

-   for (string::const_iterator I = Str.begin(); I != Str.end(); I++)
+   for (string::const_iterator I = Str.begin(); I != Str.end(); ++I)

    {
       if (strchr(Bad,*I) != 0 || isprint(*I) == 0 || 
          *I == 0x25 || // percent '%' char
    {
       if (strchr(Bad,*I) != 0 || isprint(*I) == 0 || 
          *I == 0x25 || // percent '%' char


@@ -298,7 +333,7 @@ string DeQuoteString(string::const_iterator const &begin,
                        string::const_iterator const &end)
 {
    string Res;
                        string::const_iterator const &end)
 {
    string Res;

-   for (string::const_iterator I = begin; I != end; I++)
+   for (string::const_iterator I = begin; I != end; ++I)

    {
       if (*I == '%' && I + 2 < end &&
          isxdigit(I[1]) && isxdigit(I[2]))
    {
       if (*I == '%' && I + 2 < end &&
          isxdigit(I[1]) && isxdigit(I[2]))


@@ -396,7 +431,7 @@ string TimeToStr(unsigned long Sec)
                                                                        /*}}}*/
 // SubstVar - Substitute a string for another string                   /*{{{*/
 // ---------------------------------------------------------------------
                                                                        /*}}}*/
 // SubstVar - Substitute a string for another string                   /*{{{*/
 // ---------------------------------------------------------------------

-/* This replaces all occurances of Subst with Contents in Str. */
+/* This replaces all occurrences of Subst with Contents in Str. */

 string SubstVar(const string &Str,const string &Subst,const string &Contents)
 {
    string::size_type Pos = 0;
 string SubstVar(const string &Str,const string &Subst,const string &Contents)
 {
    string::size_type Pos = 0;


@@ -632,7 +667,7 @@ string LookupTag(const string &Message,const char *Tag,const char *Default)
 {
    // Look for a matching tag.
    int Length = strlen(Tag);
 {
    // Look for a matching tag.
    int Length = strlen(Tag);

-   for (string::const_iterator I = Message.begin(); I + Length < Message.end(); I++)
+   for (string::const_iterator I = Message.begin(); I + Length < Message.end(); ++I)

    {
       // Found the tag
       if (I[Length] == ':' && stringcasecmp(I,I+Length,Tag) == 0)
    {
       // Found the tag
       if (I[Length] == ':' && stringcasecmp(I,I+Length,Tag) == 0)


@@ -640,14 +675,14 @@ string LookupTag(const string &Message,const char *Tag,const char *Default)
         // Find the end of line and strip the leading/trailing spaces
         string::const_iterator J;
         I += Length + 1;
         // Find the end of line and strip the leading/trailing spaces
         string::const_iterator J;
         I += Length + 1;

-        for (; isspace(*I) != 0 && I < Message.end(); I++);
-        for (J = I; *J != '\n' && J < Message.end(); J++);
-        for (; J > I && isspace(J[-1]) != 0; J--);
+        for (; isspace(*I) != 0 && I < Message.end(); ++I);
+        for (J = I; *J != '\n' && J < Message.end(); ++J);
+        for (; J > I && isspace(J[-1]) != 0; --J);

         
         return string(I,J);
       }
       
         
         return string(I,J);
       }
       

-      for (; *I != '\n' && I < Message.end(); I++);
+      for (; *I != '\n' && I < Message.end(); ++I);

    }   
    
    // Failed to find a match
    }   
    
    // Failed to find a match


@@ -692,14 +727,16 @@ int StringToBool(const string &Text,int Default)
    year 2000 complient and timezone neutral */
 string TimeRFC1123(time_t Date)
 {
    year 2000 complient and timezone neutral */
 string TimeRFC1123(time_t Date)
 {

-   struct tm Conv = *gmtime(&Date);
-   char Buf[300];
+   struct tm Conv;
+   if (gmtime_r(&Date, &Conv) == NULL)
+      return "";

 
 

+   char Buf[300];

    const char *Day[] = {"Sun","Mon","Tue","Wed","Thu","Fri","Sat"};
    const char *Month[] = {"Jan","Feb","Mar","Apr","May","Jun","Jul",
                           "Aug","Sep","Oct","Nov","Dec"};
 
    const char *Day[] = {"Sun","Mon","Tue","Wed","Thu","Fri","Sat"};
    const char *Month[] = {"Jan","Feb","Mar","Apr","May","Jun","Jul",
                           "Aug","Sep","Oct","Nov","Dec"};
 

-   sprintf(Buf,"%s, %02i %s %i %02i:%02i:%02i GMT",Day[Conv.tm_wday],
+   snprintf(Buf, sizeof(Buf), "%s, %02i %s %i %02i:%02i:%02i GMT",Day[Conv.tm_wday],

           Conv.tm_mday,Month[Conv.tm_mon],Conv.tm_year+1900,Conv.tm_hour,
           Conv.tm_min,Conv.tm_sec);
    return Buf;
           Conv.tm_mday,Month[Conv.tm_mon],Conv.tm_year+1900,Conv.tm_hour,
           Conv.tm_min,Conv.tm_sec);
    return Buf;


@@ -749,7 +786,8 @@ bool ReadMessages(int Fd, vector<string> &List)
       // Look for the end of the message
       for (char *I = Buffer; I + 1 < End; I++)
       {
       // Look for the end of the message
       for (char *I = Buffer; I + 1 < End; I++)
       {

-        if (I[0] != '\n' || I[1] != '\n')
+        if (I[1] != '\n' ||
+              (I[0] != '\n' && strncmp(I, "\r\n\r\n", 4) != 0))

            continue;
         
         // Pull the message out
            continue;
         
         // Pull the message out


@@ -757,7 +795,7 @@ bool ReadMessages(int Fd, vector<string> &List)
         PartialMessage += Message;
 
         // Fix up the buffer
         PartialMessage += Message;
 
         // Fix up the buffer

-        for (; I < End && *I == '\n'; I++);
+        for (; I < End && (*I == '\n' || *I == '\r'); ++I);

         End -= I-Buffer;        
         memmove(Buffer,I,End-Buffer);
         I = Buffer;
         End -= I-Buffer;        
         memmove(Buffer,I,End-Buffer);
         I = Buffer;


@@ -893,7 +931,7 @@ bool FTPMDTMStrToTime(const char* const str,time_t &time)
                                                                        /*}}}*/
 // StrToTime - Converts a string into a time_t                         /*{{{*/
 // ---------------------------------------------------------------------
                                                                        /*}}}*/
 // StrToTime - Converts a string into a time_t                         /*{{{*/
 // ---------------------------------------------------------------------

-/* This handles all 3 populare time formats including RFC 1123, RFC 1036
+/* This handles all 3 popular time formats including RFC 1123, RFC 1036

    and the C library asctime format. It requires the GNU library function
    'timegm' to convert a struct tm in UTC to a time_t. For some bizzar
    reason the C library does not provide any such function :< This also
    and the C library asctime format. It requires the GNU library function
    'timegm' to convert a struct tm in UTC to a time_t. For some bizzar
    reason the C library does not provide any such function :< This also


@@ -902,24 +940,23 @@ bool StrToTime(const string &Val,time_t &Result)
 {
    struct tm Tm;
    char Month[10];
 {
    struct tm Tm;
    char Month[10];

-   const char *I = Val.c_str();
-   
+

    // Skip the day of the week
    // Skip the day of the week

-   for (;*I != 0  && *I != ' '; I++);
-   
+   const char *I = strchr(Val.c_str(), ' ');
+

    // Handle RFC 1123 time
    Month[0] = 0;
    // Handle RFC 1123 time
    Month[0] = 0;

-   if (sscanf(I," %d %3s %d %d:%d:%d GMT",&Tm.tm_mday,Month,&Tm.tm_year,
+   if (sscanf(I," %2d %3s %4d %2d:%2d:%2d GMT",&Tm.tm_mday,Month,&Tm.tm_year,

              &Tm.tm_hour,&Tm.tm_min,&Tm.tm_sec) != 6)
    {
       // Handle RFC 1036 time
              &Tm.tm_hour,&Tm.tm_min,&Tm.tm_sec) != 6)
    {
       // Handle RFC 1036 time

-      if (sscanf(I," %d-%3s-%d %d:%d:%d GMT",&Tm.tm_mday,Month,
+      if (sscanf(I," %2d-%3s-%3d %2d:%2d:%2d GMT",&Tm.tm_mday,Month,

                 &Tm.tm_year,&Tm.tm_hour,&Tm.tm_min,&Tm.tm_sec) == 6)
         Tm.tm_year += 1900;
       else
       {
         // asctime format
                 &Tm.tm_year,&Tm.tm_hour,&Tm.tm_min,&Tm.tm_sec) == 6)
         Tm.tm_year += 1900;
       else
       {
         // asctime format

-        if (sscanf(I," %3s %d %d:%d:%d %d",Month,&Tm.tm_mday,
+        if (sscanf(I," %3s %2d %2d:%2d:%2d %4d",Month,&Tm.tm_mday,

                    &Tm.tm_hour,&Tm.tm_min,&Tm.tm_sec,&Tm.tm_year) != 6)
         {
            // 'ftp' time
                    &Tm.tm_hour,&Tm.tm_min,&Tm.tm_sec,&Tm.tm_year) != 6)
         {
            // 'ftp' time


@@ -934,6 +971,8 @@ bool StrToTime(const string &Val,time_t &Result)
    Tm.tm_isdst = 0;
    if (Month[0] != 0)
       Tm.tm_mon = MonthConv(Month);
    Tm.tm_isdst = 0;
    if (Month[0] != 0)
       Tm.tm_mon = MonthConv(Month);

+   else
+      Tm.tm_mon = 0; // we don't have a month, so pick something

    Tm.tm_year -= 1900;
    
    // Convert to local time and then to GMT
    Tm.tm_year -= 1900;
    
    // Convert to local time and then to GMT


@@ -968,6 +1007,34 @@ bool StrToNum(const char *Str,unsigned long &Res,unsigned Len,unsigned Base)
    return true;
 }
                                                                        /*}}}*/
    return true;
 }
                                                                        /*}}}*/

+// StrToNum - Convert a fixed length string to a number                        /*{{{*/
+// ---------------------------------------------------------------------
+/* This is used in decoding the crazy fixed length string headers in 
+   tar and ar files. */
+bool StrToNum(const char *Str,unsigned long long &Res,unsigned Len,unsigned Base)
+{
+   char S[30];
+   if (Len >= sizeof(S))
+      return false;
+   memcpy(S,Str,Len);
+   S[Len] = 0;
+   
+   // All spaces is a zero
+   Res = 0;
+   unsigned I;
+   for (I = 0; S[I] == ' '; I++);
+   if (S[I] == 0)
+      return true;
+   
+   char *End;
+   Res = strtoull(S,&End,Base);
+   if (End == S)
+      return false;
+   
+   return true;
+}
+                                                                       /*}}}*/
+

 // Base256ToNum - Convert a fixed length binary to a number             /*{{{*/
 // ---------------------------------------------------------------------
 /* This is used in decoding the 256bit encoded fixed length fields in
 // Base256ToNum - Convert a fixed length binary to a number             /*{{{*/
 // ---------------------------------------------------------------------
 /* This is used in decoding the 256bit encoded fixed length fields in


@@ -1068,9 +1135,11 @@ bool TokSplitString(char Tok,char *Input,char **List,
    also, but the advantage is that we have an iteratable vector */
 vector<string> VectorizeString(string const &haystack, char const &split)
 {
    also, but the advantage is that we have an iteratable vector */
 vector<string> VectorizeString(string const &haystack, char const &split)
 {

+   vector<string> exploded;
+   if (haystack.empty() == true)
+      return exploded;

    string::const_iterator start = haystack.begin();
    string::const_iterator end = start;
    string::const_iterator start = haystack.begin();
    string::const_iterator end = start;

-   vector<string> exploded;

    do {
       for (; end != haystack.end() && *end != split; ++end);
       exploded.push_back(string(start, end));
    do {
       for (; end != haystack.end() && *end != split; ++end);
       exploded.push_back(string(start, end));


@@ -1079,6 +1148,37 @@ vector<string> VectorizeString(string const &haystack, char const &split)
    return exploded;
 }
                                                                        /*}}}*/
    return exploded;
 }
                                                                        /*}}}*/

+// StringSplit - split a string into a string vector by token          /*{{{*/
+// ---------------------------------------------------------------------
+/* See header for details.
+ */
+vector<string> StringSplit(std::string const &s, std::string const &sep,
+                           unsigned int maxsplit)
+{
+   vector<string> split;
+   size_t start, pos;
+
+   // no seperator given, this is bogus
+   if(sep.size() == 0)
+      return split;
+
+   start = pos = 0;
+   while (pos != string::npos)
+   {
+      pos = s.find(sep, start);
+      split.push_back(s.substr(start, pos-start));
+      
+      // if maxsplit is reached, the remaining string is the last item
+      if(split.size() >= maxsplit)
+      {
+         split[split.size()-1] = s.substr(start);
+         break;
+      }
+      start = pos+sep.size();
+   }
+   return split;
+}
+                                                                       /*}}}*/

 // RegexChoice - Simple regex list/list matcher                                /*{{{*/
 // ---------------------------------------------------------------------
 /* */
 // RegexChoice - Simple regex list/list matcher                                /*{{{*/
 // ---------------------------------------------------------------------
 /* */


@@ -1089,7 +1189,7 @@ unsigned long RegexChoice(RxChoiceList *Rxs,const char **ListBegin,
       R->Hit = false;
 
    unsigned long Hits = 0;
       R->Hit = false;
 
    unsigned long Hits = 0;

-   for (; ListBegin != ListEnd; ListBegin++)
+   for (; ListBegin < ListEnd; ++ListBegin)

    {
       // Check if the name is a regex
       const char *I;
    {
       // Check if the name is a regex
       const char *I;


@@ -1139,34 +1239,50 @@ unsigned long RegexChoice(RxChoiceList *Rxs,const char **ListBegin,
    return Hits;
 }
                                                                        /*}}}*/
    return Hits;
 }
                                                                        /*}}}*/

-// ioprintf - C format string outputter to C++ iostreams               /*{{{*/
+// {str,io}printf - C format string outputter to C++ strings/iostreams /*{{{*/

 // ---------------------------------------------------------------------
 /* This is used to make the internationalization strings easier to translate
    and to allow reordering of parameters */
 // ---------------------------------------------------------------------
 /* This is used to make the internationalization strings easier to translate
    and to allow reordering of parameters */

-void ioprintf(ostream &out,const char *format,...) 
+static bool iovprintf(ostream &out, const char *format,
+                     va_list &args, ssize_t &size) {
+   char *S = (char*)malloc(size);
+   ssize_t const n = vsnprintf(S, size, format, args);
+   if (n > -1 && n < size) {
+      out << S;
+      free(S);
+      return true;
+   } else {
+      if (n > -1)
+        size = n + 1;
+      else
+        size *= 2;
+   }
+   free(S);
+   return false;
+}
+void ioprintf(ostream &out,const char *format,...)

 {
    va_list args;
 {
    va_list args;

-   va_start(args,format);
-   
-   // sprintf the description
-   char S[4096];
-   vsnprintf(S,sizeof(S),format,args);
-   out << S;
+   ssize_t size = 400;
+   while (true) {
+      va_start(args,format);
+      if (iovprintf(out, format, args, size) == true)
+        return;
+      va_end(args);
+   }

 }
 }

-                                                                       /*}}}*/
-// strprintf - C format string outputter to C++ strings                /*{{{*/
-// ---------------------------------------------------------------------
-/* This is used to make the internationalization strings easier to translate
-   and to allow reordering of parameters */
-void strprintf(string &out,const char *format,...) 
+void strprintf(string &out,const char *format,...)

 {
    va_list args;
 {
    va_list args;

-   va_start(args,format);
-   
-   // sprintf the description
-   char S[4096];
-   vsnprintf(S,sizeof(S),format,args);
-   out = string(S);
+   ssize_t size = 400;
+   std::ostringstream outstr;
+   while (true) {
+      va_start(args,format);
+      if (iovprintf(outstr, format, args, size) == true)
+        break;
+      va_end(args);
+   }
+   out = outstr.str();

 }
                                                                        /*}}}*/
 // safe_snprintf - Safer snprintf                                      /*{{{*/
 }
                                                                        /*}}}*/
 // safe_snprintf - Safer snprintf                                      /*{{{*/


@@ -1180,12 +1296,12 @@ char *safe_snprintf(char *Buffer,char *End,const char *Format,...)
    va_list args;
    int Did;
 
    va_list args;
    int Did;
 

-   va_start(args,Format);
-

    if (End <= Buffer)
       return End;
    if (End <= Buffer)
       return End;

-
+   va_start(args,Format);

    Did = vsnprintf(Buffer,End - Buffer,Format,args);
    Did = vsnprintf(Buffer,End - Buffer,Format,args);

+   va_end(args);
+

    if (Did < 0 || Buffer + Did > End)
       return End;
    return Buffer + Did;
    if (Did < 0 || Buffer + Did > End)
       return End;
    return Buffer + Did;


@@ -1200,11 +1316,11 @@ string StripEpoch(const string &VerStr)
       return VerStr;
    return VerStr.substr(i+1);
 }
       return VerStr;
    return VerStr.substr(i+1);
 }

-
+                                                                       /*}}}*/

 // tolower_ascii - tolower() function that ignores the locale          /*{{{*/
 // ---------------------------------------------------------------------
 /* This little function is the most called method we have and tries
 // tolower_ascii - tolower() function that ignores the locale          /*{{{*/
 // ---------------------------------------------------------------------
 /* This little function is the most called method we have and tries

-   therefore to do the absolut minimum - and is noteable faster than
+   therefore to do the absolut minimum - and is notable faster than

    standard tolower/toupper and as a bonus avoids problems with different
    locales - we only operate on ascii chars anyway. */
 int tolower_ascii(int const c)
    standard tolower/toupper and as a bonus avoids problems with different
    locales - we only operate on ascii chars anyway. */
 int tolower_ascii(int const c)


@@ -1215,14 +1331,14 @@ int tolower_ascii(int const c)
 }
                                                                        /*}}}*/
 
 }
                                                                        /*}}}*/
 

-// CheckDomainList - See if Host is in a , seperate list               /*{{{*/
+// CheckDomainList - See if Host is in a , separate list               /*{{{*/

 // ---------------------------------------------------------------------
 // ---------------------------------------------------------------------

-/* The domain list is a comma seperate list of domains that are suffix
+/* The domain list is a comma separate list of domains that are suffix

    matched against the argument */
 bool CheckDomainList(const string &Host,const string &List)
 {
    string::const_iterator Start = List.begin();
    matched against the argument */
 bool CheckDomainList(const string &Host,const string &List)
 {
    string::const_iterator Start = List.begin();

-   for (string::const_iterator Cur = List.begin(); Cur <= List.end(); Cur++)
+   for (string::const_iterator Cur = List.begin(); Cur <= List.end(); ++Cur)

    {
       if (Cur < List.end() && *Cur != ',')
         continue;
    {
       if (Cur < List.end() && *Cur != ',')
         continue;


@@ -1238,7 +1354,80 @@ bool CheckDomainList(const string &Host,const string &List)
    return false;
 }
                                                                        /*}}}*/
    return false;
 }
                                                                        /*}}}*/

+// strv_length - Return the length of a NULL-terminated string array   /*{{{*/
+// ---------------------------------------------------------------------
+/* */
+size_t strv_length(const char **str_array)
+{
+   size_t i;
+   for (i=0; str_array[i] != NULL; i++)
+      /* nothing */
+      ;
+   return i;
+}
+
+// DeEscapeString - unescape (\0XX and \xXX) from a string             /*{{{*/
+// ---------------------------------------------------------------------
+/* */
+string DeEscapeString(const string &input)
+{
+   char tmp[3];
+   string::const_iterator it;
+   string output;
+   for (it = input.begin(); it != input.end(); ++it)
+   {
+      // just copy non-escape chars
+      if (*it != '\\')
+      {
+         output += *it;
+         continue;
+      }
+
+      // deal with double escape
+      if (*it == '\\' && 
+          (it + 1 < input.end()) &&  it[1] == '\\')
+      {
+         // copy
+         output += *it;
+         // advance iterator one step further
+         ++it;
+         continue;
+      }
+        
+      // ensure we have a char to read
+      if (it + 1 == input.end())
+         continue;

 
 

+      // read it
+      ++it;
+      switch (*it)
+      {
+         case '0':
+            if (it + 2 <= input.end()) {
+               tmp[0] = it[1];
+               tmp[1] = it[2];
+               tmp[2] = 0;
+               output += (char)strtol(tmp, 0, 8);
+               it += 2;
+            }
+            break;
+         case 'x':
+            if (it + 2 <= input.end()) {
+               tmp[0] = it[1];
+               tmp[1] = it[2];
+               tmp[2] = 0;
+               output += (char)strtol(tmp, 0, 16);
+               it += 2;
+            }
+            break;
+         default:
+            // FIXME: raise exception here?
+            break;
+      }
+   }
+   return output;
+}
+                                                                       /*}}}*/

 // URI::CopyFrom - Copy from an object                                 /*{{{*/
 // ---------------------------------------------------------------------
 /* This parses the URI into all of its components */
 // URI::CopyFrom - Copy from an object                                 /*{{{*/
 // ---------------------------------------------------------------------
 /* This parses the URI into all of its components */


@@ -1247,7 +1436,7 @@ void URI::CopyFrom(const string &U)
    string::const_iterator I = U.begin();
 
    // Locate the first colon, this separates the scheme
    string::const_iterator I = U.begin();
 
    // Locate the first colon, this separates the scheme

-   for (; I < U.end() && *I != ':' ; I++);
+   for (; I < U.end() && *I != ':' ; ++I);

    string::const_iterator FirstColon = I;
 
    /* Determine if this is a host type URI with a leading double //
    string::const_iterator FirstColon = I;
 
    /* Determine if this is a host type URI with a leading double //


@@ -1259,7 +1448,7 @@ void URI::CopyFrom(const string &U)
    /* Find the / indicating the end of the hostname, ignoring /'s in the
       square brackets */
    bool InBracket = false;
    /* Find the / indicating the end of the hostname, ignoring /'s in the
       square brackets */
    bool InBracket = false;

-   for (; SingleSlash < U.end() && (*SingleSlash != '/' || InBracket == true); SingleSlash++)
+   for (; SingleSlash < U.end() && (*SingleSlash != '/' || InBracket == true); ++SingleSlash)

    {
       if (*SingleSlash == '[')
         InBracket = true;
    {
       if (*SingleSlash == '[')
         InBracket = true;


@@ -1292,11 +1481,11 @@ void URI::CopyFrom(const string &U)
    I = FirstColon + 1;
    if (I > SingleSlash)
       I = SingleSlash;
    I = FirstColon + 1;
    if (I > SingleSlash)
       I = SingleSlash;

-   for (; I < SingleSlash && *I != ':'; I++);
+   for (; I < SingleSlash && *I != ':'; ++I);

    string::const_iterator SecondColon = I;
    
    // Search for the @ after the colon
    string::const_iterator SecondColon = I;
    
    // Search for the @ after the colon

-   for (; I < SingleSlash && *I != '@'; I++);
+   for (; I < SingleSlash && *I != '@'; ++I);

    string::const_iterator At = I;
    
    // Now write the host and user/pass
    string::const_iterator At = I;
    
    // Now write the host and user/pass


@@ -1370,9 +1559,12 @@ URI::operator string()
           
       if (User.empty() == false)
       {
           
       if (User.empty() == false)
       {

-        Res +=  User;
+        // FIXME: Technically userinfo is permitted even less
+        // characters than these, but this is not conveniently
+        // expressed with a blacklist.
+        Res += QuoteString(User, ":/?#[]@");

         if (Password.empty() == false)
         if (Password.empty() == false)

-           Res += ":" + Password;
+           Res += ":" + QuoteString(Password, ":/?#[]@");

         Res += "@";
       }
       
         Res += "@";
       }
       


@@ -1411,7 +1603,6 @@ string URI::SiteOnly(const string &URI)
    U.User.clear();
    U.Password.clear();
    U.Path.clear();
    U.User.clear();
    U.Password.clear();
    U.Path.clear();

-   U.Port = 0;

    return U;
 }
                                                                        /*}}}*/
    return U;
 }
                                                                        /*}}}*/


@@ -1423,7 +1614,6 @@ string URI::NoUserPassword(const string &URI)
    ::URI U(URI);
    U.User.clear();
    U.Password.clear();
    ::URI U(URI);
    U.User.clear();
    U.Password.clear();

-   U.Port = 0;

    return U;
 }
                                                                        /*}}}*/
    return U;
 }
                                                                        /*}}}*/