dd support for the LANGUAGE environment variable

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

diff --git a/apt-pkg/contrib/strutl.cc b/apt-pkg/contrib/strutl.cc
index 37d263794b42b9c2049096068e2b3b99124371bd..3bbaf5f3045dc6d3464325dbe75179ab8c34b093 100644 (file)
--- a/apt-pkg/contrib/strutl.cc
+++ b/apt-pkg/contrib/strutl.cc
@@ -15,10 +15,6 @@
    ##################################################################### */
                                                                        /*}}}*/
 // Includes                                                            /*{{{*/
    ##################################################################### */
                                                                        /*}}}*/
 // Includes                                                            /*{{{*/

-#ifdef __GNUG__
-#pragma implementation "apt-pkg/strutl.h"
-#endif
-

 #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>


@@ -28,6 +24,7 @@
 #include <ctype.h>
 #include <string.h>
 #include <stdio.h>
 #include <ctype.h>
 #include <string.h>
 #include <stdio.h>

+#include <algorithm>

 #include <unistd.h>
 #include <regex.h>
 #include <errno.h>
 #include <unistd.h>
 #include <regex.h>
 #include <errno.h>


@@ -46,9 +43,10 @@ bool UTF8ToCodeset(const char *codeset, const string &orig, string *dest)
 {
   iconv_t cd;
   const char *inbuf;
 {
   iconv_t cd;
   const char *inbuf;

-  char *inptr, *outbuf, *outptr;
-  size_t insize, outsize;
-  
+  char *inptr, *outbuf;
+  size_t insize, bufsize;
+  dest->clear();
+

   cd = iconv_open(codeset, "UTF-8");
   if (cd == (iconv_t)(-1)) {
      // Something went wrong
   cd = iconv_open(codeset, "UTF-8");
   if (cd == (iconv_t)(-1)) {
      // Something went wrong


@@ -58,22 +56,49 @@ bool UTF8ToCodeset(const char *codeset, const string &orig, string *dest)
      else
        perror("iconv_open");
      
      else
        perror("iconv_open");
      

-     // Clean the destination string
-     *dest = "";
-     

      return false;
   }
 
      return false;
   }
 

-  insize = outsize = orig.size();
+  insize = bufsize = orig.size();

   inbuf = orig.data();
   inptr = (char *)inbuf;
   inbuf = orig.data();
   inptr = (char *)inbuf;

-  outbuf = new char[insize+1];
-  outptr = outbuf;
+  outbuf = new char[bufsize];
+  size_t lastError = -1;

 
 

-  iconv(cd, &inptr, &insize, &outptr, &outsize);
-  *outptr = '\0';
+  while (insize != 0)
+  {
+     char *outptr = outbuf;
+     size_t outsize = bufsize;
+     size_t const err = iconv(cd, &inptr, &insize, &outptr, &outsize);
+     dest->append(outbuf, outptr - outbuf);
+     if (err == (size_t)(-1))
+     {
+       switch (errno)
+       {
+       case EILSEQ:
+          insize--;
+          inptr++;
+          // replace a series of unknown multibytes with a single "?"
+          if (lastError != insize) {
+             lastError = insize - 1;
+             dest->append("?");
+          }
+          break;
+       case EINVAL:
+          insize = 0;
+          break;
+       case E2BIG:
+          if (outptr == outbuf)
+          {
+             bufsize *= 2;
+             delete[] outbuf;
+             outbuf = new char[bufsize];
+          }
+          break;
+       }
+     }
+  }

 
 

-  *dest = outbuf;

   delete[] outbuf;
   
   iconv_close(cd);
   delete[] outbuf;
   
   iconv_close(cd);


@@ -307,13 +332,13 @@ string SizeToStr(double Size)
    {
       if (ASize < 100 && I != 0)
       {
    {
       if (ASize < 100 && I != 0)
       {

-         sprintf(S,"%.1f%c",ASize,Ext[I]);
+         sprintf(S,"%'.1f%c",ASize,Ext[I]);

         break;
       }
       
       if (ASize < 10000)
       {
         break;
       }
       
       if (ASize < 10000)
       {

-         sprintf(S,"%.0f%c",ASize,Ext[I]);
+         sprintf(S,"%'.0f%c",ASize,Ext[I]);

         break;
       }
       ASize /= 1000.0;
         break;
       }
       ASize /= 1000.0;


@@ -334,23 +359,27 @@ string TimeToStr(unsigned long Sec)
    {
       if (Sec > 60*60*24)
       {
    {
       if (Sec > 60*60*24)
       {

-        sprintf(S,"%lid %lih%lim%lis",Sec/60/60/24,(Sec/60/60) % 24,(Sec/60) % 60,Sec % 60);
+        //d means days, h means hours, min means minutes, s means seconds
+        sprintf(S,_("%lid %lih %limin %lis"),Sec/60/60/24,(Sec/60/60) % 24,(Sec/60) % 60,Sec % 60);

         break;
       }
       
       if (Sec > 60*60)
       {
         break;
       }
       
       if (Sec > 60*60)
       {

-        sprintf(S,"%lih%lim%lis",Sec/60/60,(Sec/60) % 60,Sec % 60);
+        //h means hours, min means minutes, s means seconds
+        sprintf(S,_("%lih %limin %lis"),Sec/60/60,(Sec/60) % 60,Sec % 60);

         break;
       }
       
       if (Sec > 60)
       {
         break;
       }
       
       if (Sec > 60)
       {

-        sprintf(S,"%lim%lis",Sec/60,Sec % 60);
+        //min means minutes, s means seconds
+        sprintf(S,_("%limin %lis"),Sec/60,Sec % 60);

         break;
       }
         break;
       }

-      
-      sprintf(S,"%lis",Sec);
+
+      //s means seconds
+      sprintf(S,_("%lis"),Sec);

       break;
    }
    
       break;
    }
    


@@ -386,6 +415,17 @@ string SubstVar(string Str,const struct SubstVar *Vars)
    return Str;
 }
                                                                        /*}}}*/
    return Str;
 }
                                                                        /*}}}*/

+// OutputInDepth - return a string with separator multiplied with depth /*{{{*/
+// ---------------------------------------------------------------------
+/* Returns a string with the supplied separator depth + 1 times in it */
+std::string OutputInDepth(const unsigned long Depth, const char* Separator)
+{
+   std::string output = "";
+   for(unsigned long d=Depth+1; d > 0; d--)
+      output.append(Separator);
+   return output;
+}
+                                                                       /*}}}*/

 // URItoFileName - Convert the uri into a unique file name             /*{{{*/
 // ---------------------------------------------------------------------
 /* This converts a URI into a safe filename. It quotes all unsafe characters
 // URItoFileName - Convert the uri into a unique file name             /*{{{*/
 // ---------------------------------------------------------------------
 /* This converts a URI into a safe filename. It quotes all unsafe characters


@@ -462,9 +502,9 @@ string Base64Encode(const string &S)
    return Final;
 }
                                                                        /*}}}*/
    return Final;
 }
                                                                        /*}}}*/

-// stringcmp - Arbitary string compare                                 /*{{{*/
+// stringcmp - Arbitrary string compare                                        /*{{{*/

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

-/* This safely compares two non-null terminated strings of arbitary 
+/* This safely compares two non-null terminated strings of arbitrary 

    length */
 int stringcmp(const char *A,const char *AEnd,const char *B,const char *BEnd)
 {
    length */
 int stringcmp(const char *A,const char *AEnd,const char *B,const char *BEnd)
 {


@@ -520,7 +560,7 @@ int stringcmp(string::const_iterator A,string::const_iterator AEnd,
 }
 #endif
                                                                        /*}}}*/
 }
 #endif
                                                                        /*}}}*/

-// stringcasecmp - Arbitary case insensitive string compare            /*{{{*/
+// stringcasecmp - Arbitrary case insensitive string compare           /*{{{*/

 // ---------------------------------------------------------------------
 /* */
 int stringcasecmp(const char *A,const char *AEnd,const char *B,const char *BEnd)
 // ---------------------------------------------------------------------
 /* */
 int stringcasecmp(const char *A,const char *AEnd,const char *B,const char *BEnd)


@@ -661,11 +701,24 @@ string TimeRFC1123(time_t Date)
 // ---------------------------------------------------------------------
 /* This pulls full messages from the input FD into the message buffer. 
    It assumes that messages will not pause during transit so no
 // ---------------------------------------------------------------------
 /* This pulls full messages from the input FD into the message buffer. 
    It assumes that messages will not pause during transit so no

-   fancy buffering is used. */
+   fancy buffering is used.
+
+   In particular: this reads blocks from the input until it believes
+   that it's run out of input text.  Each block is terminated by a
+   double newline ('\n' followed by '\n').  As noted below, there is a
+   bug in this code: it assumes that all the blocks have been read if
+   it doesn't see additional text in the buffer after the last one is
+   parsed, which will cause it to lose blocks if the last block
+   coincides with the end of the buffer.
+ */

 bool ReadMessages(int Fd, vector<string> &List)
 {
    char Buffer[64000];
    char *End = Buffer;
 bool ReadMessages(int Fd, vector<string> &List)
 {
    char Buffer[64000];
    char *End = Buffer;

+   // Represents any left-over from the previous iteration of the
+   // parse loop.  (i.e., if a message is split across the end
+   // of the buffer, it goes here)
+   string PartialMessage;

    
    while (1)
    {
    
    while (1)
    {


@@ -693,6 +746,7 @@ bool ReadMessages(int Fd, vector<string> &List)
         
         // Pull the message out
         string Message(Buffer,I-Buffer);
         
         // Pull the message out
         string Message(Buffer,I-Buffer);

+        PartialMessage += Message;

 
         // Fix up the buffer
         for (; I < End && *I == '\n'; I++);
 
         // Fix up the buffer
         for (; I < End && *I == '\n'; I++);


@@ -700,10 +754,32 @@ bool ReadMessages(int Fd, vector<string> &List)
         memmove(Buffer,I,End-Buffer);
         I = Buffer;
         
         memmove(Buffer,I,End-Buffer);
         I = Buffer;
         

-        List.push_back(Message);
+        List.push_back(PartialMessage);
+        PartialMessage.clear();

       }
       }

-      if (End == Buffer)
-        return true;
+      if (End != Buffer)
+       {
+         // If there's text left in the buffer, store it
+         // in PartialMessage and throw the rest of the buffer
+         // away.  This allows us to handle messages that
+         // are longer than the static buffer size.
+         PartialMessage += string(Buffer, End);
+         End = Buffer;
+       }
+      else
+       {
+         // BUG ALERT: if a message block happens to end at a
+         // multiple of 64000 characters, this will cause it to
+         // terminate early, leading to a badly formed block and
+         // probably crashing the method.  However, this is the only
+         // way we have to find the end of the message block.  I have
+         // an idea of how to fix this, but it will require changes
+         // to the protocol (essentially to mark the beginning and
+         // end of the block).
+         //
+         //  -- dburrows 2008-04-02
+         return true;
+       }

 
       if (WaitFd(Fd) == false)
         return false;
 
       if (WaitFd(Fd) == false)
         return false;


@@ -924,6 +1000,24 @@ bool TokSplitString(char Tok,char *Input,char **List,
    return true;
 }
                                                                        /*}}}*/
    return true;
 }
                                                                        /*}}}*/

+// ExplodeString - Split a string up into a vector                     /*{{{*/
+// ---------------------------------------------------------------------
+/* This can be used to split a given string up into a vector, so the
+   propose is the same as in the method above and this one is a bit slower
+   also, but the advantage is that we an iteratable vector */
+vector<string> ExplodeString(string const &haystack, char const &split)
+{
+   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));
+      start = end + 1;
+   } while (end != haystack.end() && (++end) != haystack.end());
+   return exploded;
+}
+                                                                       /*}}}*/

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


@@ -994,11 +1088,26 @@ void ioprintf(ostream &out,const char *format,...)
    va_start(args,format);
    
    // sprintf the description
    va_start(args,format);
    
    // sprintf the description

-   char S[400];
+   char S[4096];

    vsnprintf(S,sizeof(S),format,args);
    out << S;
 }
                                                                        /*}}}*/
    vsnprintf(S,sizeof(S),format,args);
    out << S;
 }
                                                                        /*}}}*/

+// 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,...) 
+{
+   va_list args;
+   va_start(args,format);
+   
+   // sprintf the description
+   char S[4096];
+   vsnprintf(S,sizeof(S),format,args);
+   out = string(S);
+}
+                                                                       /*}}}*/

 // safe_snprintf - Safer snprintf                                      /*{{{*/
 // ---------------------------------------------------------------------
 /* This is a snprintf that will never (ever) go past 'End' and returns a
 // safe_snprintf - Safer snprintf                                      /*{{{*/
 // ---------------------------------------------------------------------
 /* This is a snprintf that will never (ever) go past 'End' and returns a


@@ -1022,6 +1131,17 @@ char *safe_snprintf(char *Buffer,char *End,const char *Format,...)
 }
                                                                        /*}}}*/
 
 }
                                                                        /*}}}*/
 

+// tolower_ascii - tolower() function that ignores the locale          /*{{{*/
+// ---------------------------------------------------------------------
+/* */
+int tolower_ascii(int c)
+{
+   if (c >= 'A' and c <= 'Z')
+      return c + 32;
+   return c;
+}
+                                                                       /*}}}*/
+

 // CheckDomainList - See if Host is in a , seperate list               /*{{{*/
 // ---------------------------------------------------------------------
 /* The domain list is a comma seperate list of domains that are suffix
 // CheckDomainList - See if Host is in a , seperate list               /*{{{*/
 // ---------------------------------------------------------------------
 /* The domain list is a comma seperate list of domains that are suffix