Fix missing URIStart() for https downloads

[apt.git] / methods / https.cc

diff --git a/methods/https.cc b/methods/https.cc
index aa6786aa86fa9ad92f82ebbdccbfa9495485422c..3a5981b5824511a4cc9d9d74eb2f2bc3fbf475cd 100644 (file)
--- a/methods/https.cc
+++ b/methods/https.cc
@@ -3,64 +3,132 @@
 // $Id: http.cc,v 1.59 2004/05/08 19:42:35 mdz Exp $
 /* ######################################################################
 
 // $Id: http.cc,v 1.59 2004/05/08 19:42:35 mdz Exp $
 /* ######################################################################
 

-   HTTPS Acquire Method - This is the HTTPS aquire method for APT.
+   HTTPS Acquire Method - This is the HTTPS acquire method for APT.

    
    It uses libcurl
 
    ##################################################################### */
                                                                        /*}}}*/
 // Include Files                                                       /*{{{*/
    
    It uses libcurl
 
    ##################################################################### */
                                                                        /*}}}*/
 // Include Files                                                       /*{{{*/

+#include <config.h>
+

 #include <apt-pkg/fileutl.h>
 #include <apt-pkg/acquire-method.h>
 #include <apt-pkg/error.h>
 #include <apt-pkg/hashes.h>
 #include <apt-pkg/netrc.h>
 #include <apt-pkg/fileutl.h>
 #include <apt-pkg/acquire-method.h>
 #include <apt-pkg/error.h>
 #include <apt-pkg/hashes.h>
 #include <apt-pkg/netrc.h>

+#include <apt-pkg/configuration.h>
+#include <apt-pkg/macros.h>
+#include <apt-pkg/strutl.h>
+#include <apt-pkg/proxy.h>

 
 #include <sys/stat.h>
 #include <sys/time.h>
 
 #include <sys/stat.h>
 #include <sys/time.h>

-#include <utime.h>

 #include <unistd.h>
 #include <unistd.h>

-#include <signal.h>

 #include <stdio.h>
 #include <stdio.h>

-#include <errno.h>
-#include <string.h>

 #include <iostream>
 #include <iostream>

-#include <apti18n.h>

 #include <sstream>
 #include <sstream>

+#include <ctype.h>
+#include <stdlib.h>

 
 

-#include "config.h"

 #include "https.h"
 
 #include "https.h"
 

+#include <apti18n.h>

                                                                        /*}}}*/
 using namespace std;
 
                                                                        /*}}}*/
 using namespace std;
 

+size_t
+HttpsMethod::parse_header(void *buffer, size_t size, size_t nmemb, void *userp)
+{
+   size_t len = size * nmemb;
+   HttpsMethod *me = (HttpsMethod *)userp;
+   std::string line((char*) buffer, len);
+   for (--len; len > 0; --len)
+      if (isspace(line[len]) == 0)
+      {
+        ++len;
+        break;
+      }
+   line.erase(len);
+
+   if (line.empty() == true)
+   {
+      if (me->Server->Result != 416 && me->Server->StartPos != 0)
+        ;
+      else if (me->Server->Result == 416 && me->Server->Size == me->File->FileSize())
+      {
+         me->Server->Result = 200;
+        me->Server->StartPos = me->Server->Size;
+        // the actual size is not important for https as curl will deal with it
+        // by itself and e.g. doesn't bother us with transport-encoding…
+        me->Server->JunkSize = std::numeric_limits<unsigned long long>::max();
+      }
+      else
+        me->Server->StartPos = 0;
+
+      me->File->Truncate(me->Server->StartPos);
+      me->File->Seek(me->Server->StartPos);
+   }
+   else if (me->Server->HeaderLine(line) == false)
+      return 0;
+
+   return size*nmemb;
+}
+

 size_t 
 HttpsMethod::write_data(void *buffer, size_t size, size_t nmemb, void *userp)
 {
    HttpsMethod *me = (HttpsMethod *)userp;
 size_t 
 HttpsMethod::write_data(void *buffer, size_t size, size_t nmemb, void *userp)
 {
    HttpsMethod *me = (HttpsMethod *)userp;

+   size_t buffer_size = size * nmemb;
+   // we don't need to count the junk here, just drop anything we get as
+   // we don't always know how long it would be, e.g. in chunked encoding.
+   if (me->Server->JunkSize != 0)
+      return buffer_size;
+
+   if (me->ReceivedData == false)
+   {
+      me->URIStart(me->Res);
+      me->ReceivedData = true;
+   }

 
 

-   if(me->File->Write(buffer, size*nmemb) != true)
+   if(me->File->Write(buffer, buffer_size) != true)

       return false;
 
       return false;
 

-   return size*nmemb;
+   return buffer_size;

 }
 
 }
 

-int 
-HttpsMethod::progress_callback(void *clientp, double dltotal, double dlnow, 
-                             double ultotal, double ulnow)
+int
+HttpsMethod::progress_callback(void *clientp, double dltotal, double /*dlnow*/,
+                             double /*ultotal*/, double /*ulnow*/)

 {
    HttpsMethod *me = (HttpsMethod *)clientp;
    if(dltotal > 0 && me->Res.Size == 0) {
 {
    HttpsMethod *me = (HttpsMethod *)clientp;
    if(dltotal > 0 && me->Res.Size == 0) {

-      me->Res.Size = (unsigned long)dltotal;
-      me->URIStart(me->Res);
+      me->Res.Size = (unsigned long long)dltotal;

    }
    return 0;
 }
 
    }
    return 0;
 }
 

+// HttpsServerState::HttpsServerState - Constructor                    /*{{{*/
+HttpsServerState::HttpsServerState(URI Srv,HttpsMethod * /*Owner*/) : ServerState(Srv, NULL)
+{
+   TimeOut = _config->FindI("Acquire::https::Timeout",TimeOut);
+   Reset();
+}
+                                                                       /*}}}*/
+

 void HttpsMethod::SetupProxy()                                         /*{{{*/
 {
    URI ServerName = Queue->Uri;
 
 void HttpsMethod::SetupProxy()                                         /*{{{*/
 {
    URI ServerName = Queue->Uri;
 

+   // Determine the proxy setting
+   AutoDetectProxy(ServerName);
+
+   // Curl should never read proxy settings from the environment, as
+   // we determine which proxy to use.  Do this for consistency among
+   // methods and prevent an environment variable overriding a
+   // no-proxy ("DIRECT") setting in apt.conf.
+   curl_easy_setopt(curl, CURLOPT_PROXY, "");
+

    // Determine the proxy setting - try https first, fallback to http and use env at last
    string UseProxy = _config->Find("Acquire::https::Proxy::" + ServerName.Host,
                                   _config->Find("Acquire::http::Proxy::" + ServerName.Host).c_str());
    // Determine the proxy setting - try https first, fallback to http and use env at last
    string UseProxy = _config->Find("Acquire::https::Proxy::" + ServerName.Host,
                                   _config->Find("Acquire::http::Proxy::" + ServerName.Host).c_str());


@@ -79,7 +147,14 @@ void HttpsMethod::SetupProxy()                                      /*{{{*/
       if (getenv("no_proxy") != 0 && CheckDomainList(ServerName.Host,getenv("no_proxy")) == true)
         return;
    } else {
       if (getenv("no_proxy") != 0 && CheckDomainList(ServerName.Host,getenv("no_proxy")) == true)
         return;
    } else {

-      const char* result = getenv("http_proxy");
+      const char* result = getenv("https_proxy");
+      // FIXME: Fall back to http_proxy is to remain compatible with
+      // existing setups and behaviour of apt.conf.  This should be
+      // deprecated in the future (including apt.conf).  Most other
+      // programs do not fall back to http proxy settings and neither
+      // should Apt.
+      if (result == NULL)
+         result = getenv("http_proxy");

       UseProxy = result == NULL ? "" : result;
    }
 
       UseProxy = result == NULL ? "" : result;
    }
 


@@ -90,6 +165,11 @@ void HttpsMethod::SetupProxy()                                      /*{{{*/
       if (Proxy.Port != 1)
         curl_easy_setopt(curl, CURLOPT_PROXYPORT, Proxy.Port);
       curl_easy_setopt(curl, CURLOPT_PROXY, Proxy.Host.c_str());
       if (Proxy.Port != 1)
         curl_easy_setopt(curl, CURLOPT_PROXYPORT, Proxy.Port);
       curl_easy_setopt(curl, CURLOPT_PROXY, Proxy.Host.c_str());

+      if (Proxy.User.empty() == false || Proxy.Password.empty() == false)
+      {
+         curl_easy_setopt(curl, CURLOPT_PROXYUSERNAME, Proxy.User.c_str());
+         curl_easy_setopt(curl, CURLOPT_PROXYPASSWORD, Proxy.Password.c_str());
+      }

    }
 }                                                                      /*}}}*/
 // HttpsMethod::Fetch - Fetch an item                                  /*{{{*/
    }
 }                                                                      /*}}}*/
 // HttpsMethod::Fetch - Fetch an item                                  /*{{{*/


@@ -98,13 +178,12 @@ void HttpsMethod::SetupProxy()                                     /*{{{*/
    depth. */
 bool HttpsMethod::Fetch(FetchItem *Itm)
 {
    depth. */
 bool HttpsMethod::Fetch(FetchItem *Itm)
 {

-   stringstream ss;

    struct stat SBuf;
    struct curl_slist *headers=NULL;  
    char curl_errorstr[CURL_ERROR_SIZE];
    struct stat SBuf;
    struct curl_slist *headers=NULL;  
    char curl_errorstr[CURL_ERROR_SIZE];

-   long curl_responsecode;

    URI Uri = Itm->Uri;
    string remotehost = Uri.Host;
    URI Uri = Itm->Uri;
    string remotehost = Uri.Host;

+   ReceivedData = false;

 
    // TODO:
    //       - http::Pipeline-Depth
 
    // TODO:
    //       - http::Pipeline-Depth


@@ -118,13 +197,18 @@ bool HttpsMethod::Fetch(FetchItem *Itm)
 
    // callbacks
    curl_easy_setopt(curl, CURLOPT_URL, static_cast<string>(Uri).c_str());
 
    // callbacks
    curl_easy_setopt(curl, CURLOPT_URL, static_cast<string>(Uri).c_str());

+   curl_easy_setopt(curl, CURLOPT_HEADERFUNCTION, parse_header);
+   curl_easy_setopt(curl, CURLOPT_WRITEHEADER, this);

    curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_data);
    curl_easy_setopt(curl, CURLOPT_WRITEDATA, this);
    curl_easy_setopt(curl, CURLOPT_PROGRESSFUNCTION, progress_callback);
    curl_easy_setopt(curl, CURLOPT_PROGRESSDATA, this);
    curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_data);
    curl_easy_setopt(curl, CURLOPT_WRITEDATA, this);
    curl_easy_setopt(curl, CURLOPT_PROGRESSFUNCTION, progress_callback);
    curl_easy_setopt(curl, CURLOPT_PROGRESSDATA, this);

+   // options

    curl_easy_setopt(curl, CURLOPT_NOPROGRESS, false);
    curl_easy_setopt(curl, CURLOPT_NOPROGRESS, false);

-   curl_easy_setopt(curl, CURLOPT_FAILONERROR, true);

    curl_easy_setopt(curl, CURLOPT_FILETIME, true);
    curl_easy_setopt(curl, CURLOPT_FILETIME, true);

+   // only allow curl to handle https, not the other stuff it supports
+   curl_easy_setopt(curl, CURLOPT_PROTOCOLS, CURLPROTO_HTTPS);
+   curl_easy_setopt(curl, CURLOPT_REDIR_PROTOCOLS, CURLPROTO_HTTPS);

 
    // SSL parameters are set by default to the common (non mirror-specific) value
    // if available (or a default one) and gets overload by mirror-specific ones.
 
    // SSL parameters are set by default to the common (non mirror-specific) value
    // if available (or a default one) and gets overload by mirror-specific ones.


@@ -143,13 +227,11 @@ bool HttpsMethod::Fetch(FetchItem *Itm)
    curl_easy_setopt(curl, CURLOPT_SSL_VERIFYPEER, peer_verify);
 
    // ... and hostname against cert CN or subjectAltName
    curl_easy_setopt(curl, CURLOPT_SSL_VERIFYPEER, peer_verify);
 
    // ... and hostname against cert CN or subjectAltName

-   int default_verify = 2;

    bool verify = _config->FindB("Acquire::https::Verify-Host",true);
    knob = "Acquire::https::"+remotehost+"::Verify-Host";
    verify = _config->FindB(knob.c_str(),verify);
    bool verify = _config->FindB("Acquire::https::Verify-Host",true);
    knob = "Acquire::https::"+remotehost+"::Verify-Host";
    verify = _config->FindB(knob.c_str(),verify);

-   if (!verify)
-      default_verify = 0;
-   curl_easy_setopt(curl, CURLOPT_SSL_VERIFYHOST, verify);
+   int const default_verify = (verify == true) ? 2 : 0;
+   curl_easy_setopt(curl, CURLOPT_SSL_VERIFYHOST, default_verify);

 
    // Also enforce issuer of server certificate using its cert
    string issuercert = _config->Find("Acquire::https::IssuerCert","");
 
    // Also enforce issuer of server certificate using its cert
    string issuercert = _config->Find("Acquire::https::IssuerCert","");


@@ -199,6 +281,7 @@ bool HttpsMethod::Fetch(FetchItem *Itm)
       if (_config->FindB("Acquire::https::No-Store",
                _config->FindB("Acquire::http::No-Store",false)) == true)
         headers = curl_slist_append(headers,"Cache-Control: no-store");
       if (_config->FindB("Acquire::https::No-Store",
                _config->FindB("Acquire::http::No-Store",false)) == true)
         headers = curl_slist_append(headers,"Cache-Control: no-store");

+      stringstream ss;

       ioprintf(ss, "Cache-Control: max-age=%u", _config->FindI("Acquire::https::Max-Age",
                _config->FindI("Acquire::http::Max-Age",0)));
       headers = curl_slist_append(headers, ss.str().c_str());
       ioprintf(ss, "Cache-Control: max-age=%u", _config->FindI("Acquire::https::Max-Age",
                _config->FindI("Acquire::http::Max-Age",0)));
       headers = curl_slist_append(headers, ss.str().c_str());


@@ -219,7 +302,7 @@ bool HttpsMethod::Fetch(FetchItem *Itm)
    curl_easy_setopt(curl, CURLOPT_USERAGENT,
        _config->Find("Acquire::https::User-Agent",
                _config->Find("Acquire::http::User-Agent",
    curl_easy_setopt(curl, CURLOPT_USERAGENT,
        _config->Find("Acquire::https::User-Agent",
                _config->Find("Acquire::http::User-Agent",

-                       "Debian APT-CURL/1.0 ("VERSION")").c_str()).c_str());
+                       "Debian APT-CURL/1.0 (" PACKAGE_VERSION ")").c_str()).c_str());

 
    // set timeout
    int const timeout = _config->FindI("Acquire::https::Timeout",
 
    // set timeout
    int const timeout = _config->FindI("Acquire::https::Timeout",


@@ -240,17 +323,31 @@ bool HttpsMethod::Fetch(FetchItem *Itm)
       curl_easy_setopt(curl, CURLOPT_VERBOSE, true);
 
    // error handling
       curl_easy_setopt(curl, CURLOPT_VERBOSE, true);
 
    // error handling

+   curl_errorstr[0] = '\0';

    curl_easy_setopt(curl, CURLOPT_ERRORBUFFER, curl_errorstr);
 
    curl_easy_setopt(curl, CURLOPT_ERRORBUFFER, curl_errorstr);
 

+   // If we ask for uncompressed files servers might respond with content-
+   // negotiation which lets us end up with compressed files we do not support,
+   // see 657029, 657560 and co, so if we have no extension on the request
+   // ask for text only. As a sidenote: If there is nothing to negotate servers
+   // seem to be nice and ignore it.
+   if (_config->FindB("Acquire::https::SendAccept", _config->FindB("Acquire::http::SendAccept", true)) == true)
+   {
+      size_t const filepos = Itm->Uri.find_last_of('/');
+      string const file = Itm->Uri.substr(filepos + 1);
+      if (flExtension(file) == file)
+        headers = curl_slist_append(headers, "Accept: text/*");
+   }
+

    // if we have the file send an if-range query with a range header
    if (stat(Itm->DestFile.c_str(),&SBuf) >= 0 && SBuf.st_size > 0)
    {
    // if we have the file send an if-range query with a range header
    if (stat(Itm->DestFile.c_str(),&SBuf) >= 0 && SBuf.st_size > 0)
    {

-      char Buf[1000];
-      sprintf(Buf,"Range: bytes=%li-\r\nIf-Range: %s\r\n",
-             (long)SBuf.st_size - 1,
-             TimeRFC1123(SBuf.st_mtime).c_str());
-      headers = curl_slist_append(headers, Buf);
-   } 
+      std::string Buf;
+      strprintf(Buf, "Range: bytes=%lli-", (long long) SBuf.st_size);
+      headers = curl_slist_append(headers, Buf.c_str());
+      strprintf(Buf, "If-Range: %s", TimeRFC1123(SBuf.st_mtime).c_str());
+      headers = curl_slist_append(headers, Buf.c_str());
+   }

    else if(Itm->LastModified > 0)
    {
       curl_easy_setopt(curl, CURLOPT_TIMECONDITION, CURL_TIMECOND_IFMODSINCE);
    else if(Itm->LastModified > 0)
    {
       curl_easy_setopt(curl, CURLOPT_TIMECONDITION, CURL_TIMECOND_IFMODSINCE);


@@ -259,71 +356,101 @@ bool HttpsMethod::Fetch(FetchItem *Itm)
 
    // go for it - if the file exists, append on it
    File = new FileFd(Itm->DestFile, FileFd::WriteAny);
 
    // go for it - if the file exists, append on it
    File = new FileFd(Itm->DestFile, FileFd::WriteAny);

-   if (File->Size() > 0)
-      File->Seek(File->Size() - 1);
-   
+   Server = new HttpsServerState(Itm->Uri, this);
+

    // keep apt updated
    Res.Filename = Itm->DestFile;
 
    // get it!
    CURLcode success = curl_easy_perform(curl);
    // keep apt updated
    Res.Filename = Itm->DestFile;
 
    // get it!
    CURLcode success = curl_easy_perform(curl);

-   curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &curl_responsecode);

 
 

-   long curl_servdate;
-   curl_easy_getinfo(curl, CURLINFO_FILETIME, &curl_servdate);
+   // If the server returns 200 OK but the If-Modified-Since condition is not
+   // met, CURLINFO_CONDITION_UNMET will be set to 1
+   long curl_condition_unmet = 0;
+   curl_easy_getinfo(curl, CURLINFO_CONDITION_UNMET, &curl_condition_unmet);
+
+   File->Close();
+   curl_slist_free_all(headers);

 
    // cleanup
 
    // cleanup

-   if(success != 0) 
+   if (success != 0)

    {
       _error->Error("%s", curl_errorstr);
    {
       _error->Error("%s", curl_errorstr);

-      Fail();
+      unlink(File->Name().c_str());
+      return false;
+   }
+
+   // server says file not modified
+   if (Server->Result == 304 || curl_condition_unmet == 1)
+   {
+      unlink(File->Name().c_str());
+      Res.IMSHit = true;
+      Res.LastModified = Itm->LastModified;
+      Res.Size = 0;
+      URIDone(Res);

       return true;
    }
       return true;
    }

-   File->Close();
+   Res.IMSHit = false;

 
 

-   // Timestamp
-   struct utimbuf UBuf;
-   if (curl_servdate != -1) {
-       UBuf.actime = curl_servdate;
-       UBuf.modtime = curl_servdate;
-       utime(File->Name().c_str(),&UBuf);
+   if (Server->Result != 200 && // OK
+        Server->Result != 206 && // Partial
+        Server->Result != 416) // invalid Range
+   {
+      char err[255];
+      snprintf(err, sizeof(err) - 1, "HttpError%i", Server->Result);
+      SetFailReason(err);
+      _error->Error("%s", err);
+      // unlink, no need keep 401/404 page content in partial/
+      unlink(File->Name().c_str());
+      return false;

    }
 
    }
 

-   // check the downloaded result
-   struct stat Buf;
-   if (stat(File->Name().c_str(),&Buf) == 0)
+   struct stat resultStat;
+   if (unlikely(stat(File->Name().c_str(), &resultStat) != 0))

    {
    {

-      Res.Filename = File->Name();
-      Res.LastModified = Buf.st_mtime;
-      Res.IMSHit = false;
-      if (curl_responsecode == 304)
-      {
-        unlink(File->Name().c_str());
-        Res.IMSHit = true;
-        Res.LastModified = Itm->LastModified;
-        Res.Size = 0;
-        URIDone(Res);
-        return true;
-      }
-      Res.Size = Buf.st_size;
+      _error->Errno("stat", "Unable to access file %s", File->Name().c_str());
+      return false;
+   }
+   Res.Size = resultStat.st_size;
+
+   // invalid range-request
+   if (Server->Result == 416)
+   {
+      unlink(File->Name().c_str());
+      Res.Size = 0;
+      delete File;
+      Redirect(Itm->Uri);
+      return true;
+   }
+
+   // Timestamp
+   curl_easy_getinfo(curl, CURLINFO_FILETIME, &Res.LastModified);
+   if (Res.LastModified != -1)
+   {
+      struct timeval times[2];
+      times[0].tv_sec = Res.LastModified;
+      times[1].tv_sec = Res.LastModified;
+      times[0].tv_usec = times[1].tv_usec = 0;
+      utimes(File->Name().c_str(), times);

    }
    }

+   else
+      Res.LastModified = resultStat.st_mtime;

 
    // take hashes
    Hashes Hash;
    FileFd Fd(Res.Filename, FileFd::ReadOnly);
 
    // take hashes
    Hashes Hash;
    FileFd Fd(Res.Filename, FileFd::ReadOnly);

-   Hash.AddFD(Fd.Fd(), Fd.Size());
+   Hash.AddFD(Fd);

    Res.TakeHashes(Hash);
    Res.TakeHashes(Hash);

-   
+

    // keep apt updated
    URIDone(Res);
 
    // cleanup
    Res.Size = 0;
    delete File;
    // keep apt updated
    URIDone(Res);
 
    // cleanup
    Res.Size = 0;
    delete File;

-   curl_slist_free_all(headers);

 
    return true;
 
    return true;

-};
+}

 
 int main()
 {
 
 int main()
 {


@@ -335,4 +462,3 @@ int main()
    return Mth.Run();
 }
 
    return Mth.Run();
 }
 

-