keep compressed indexes in a low-cost format

Downloading and storing are two different operations were different
compression types can be preferred. For downloading we provide the
choice via Acquire::CompressionTypes::Order as there is a choice to
be made between download size and speed – and limited by whats available
in the repository.

Storage on the other hand has all compressions currently supported by
apt available and to reduce runtime of tools accessing these files the
compression type should be a low-cost format in terms of decompression.

apt traditionally stores its indexes uncompressed on disk, but has
options to keep them compressed. Now that apt downloads additional files
we also deal with files which simply can't be stored uncompressed as
they are just too big (like Contents for apt-file). Traditionally they
are downloaded in a low-cost format (gz) as repositories do not provide
other formats, but there might be even lower-cost formats and for
download we could introduce higher-cost in the repositories.

Downloading an entire index potentially requires recompression to
another format, so an update takes potentially longer – but big files
are usually updated via pdiffs which has to de- and re-compress anyhow
and does it on the fly anyhow, so there is no extra time needed and in
general it seems to be benefitial to invest the time in update to save
time later on file access.

6 files changed, 124 insertions, 108 deletions

diff --git a/apt-pkg/acquire-item.cc b/apt-pkg/acquire-item.cc
index 82ccaff64..7c7a204c4 100644
--- a/apt-pkg/acquire-item.cc
+++ b/apt-pkg/acquire-item.cc
@@ -85,32 +85,16 @@ static std::string GetKeepCompressedFileName(std::string file, IndexTarget const
    if (Target.KeepCompressed == false)
       return file;
 
-   std::string const CompressionTypes = Target.Option(IndexTarget::COMPRESSIONTYPES);
-   if (CompressionTypes.empty() == false)
+   std::string const KeepCompressedAs = Target.Option(IndexTarget::KEEPCOMPRESSEDAS);
+   if (KeepCompressedAs.empty() == false)
    {
-      std::string const ext = CompressionTypes.substr(0, CompressionTypes.find(' '));
+      std::string const ext = KeepCompressedAs.substr(0, KeepCompressedAs.find(' '));
       if (ext != "uncompressed")
 	 file.append(".").append(ext);
    }
    return file;
 }
 									/*}}}*/
-static std::string GetCompressedFileName(IndexTarget const &Target, std::string const &Name, std::string const &Ext) /*{{{*/
-{
-   if (Ext.empty() || Ext == "uncompressed")
-      return Name;
-
-   // do not reverify cdrom sources as apt-cdrom may rewrite the Packages
-   // file when its doing the indexcopy
-   if (Target.URI.substr(0,6) == "cdrom:")
-      return Name;
-
-   // adjust DestFile if its compressed on disk
-   if (Target.KeepCompressed == true)
-      return Name + '.' + Ext;
-   return Name;
-}
-									/*}}}*/
 static std::string GetMergeDiffsPatchFileName(std::string const &Final, std::string const &Patch)/*{{{*/
 {
    // rred expects the patch as $FinalFile.ed.$patchname.gz
@@ -326,7 +310,7 @@ std::string pkgAcqDiffIndex::GetFinalFilename() const
 std::string pkgAcqIndex::GetFinalFilename() const
 {
    std::string const FinalFile = GetFinalFileNameFromURI(Target.URI);
-   return GetCompressedFileName(Target, FinalFile, CurrentCompressionExtension);
+   return GetKeepCompressedFileName(FinalFile, Target);
 }
 std::string pkgAcqMetaSig::GetFinalFilename() const
 {
@@ -379,7 +363,30 @@ bool pkgAcqTransactionItem::TransactionState(TransactionStates const state)
       case TransactionCommit:
 	 if(PartialFile.empty() == false)
 	 {
-	    if (PartialFile != DestFile)
+	    bool sameFile = (PartialFile == DestFile);
+	    // we use symlinks on IMS-Hit to avoid copies
+	    if (RealFileExists(DestFile))
+	    {
+	       struct stat Buf;
+	       if (lstat(PartialFile.c_str(), &Buf) != -1)
+	       {
+		  if (S_ISLNK(Buf.st_mode) && Buf.st_size > 0)
+		  {
+		     char partial[Buf.st_size + 1];
+		     ssize_t const sp = readlink(PartialFile.c_str(), partial, Buf.st_size);
+		     if (sp == -1)
+			_error->Errno("pkgAcqTransactionItem::TransactionState-sp", _("Failed to readlink %s"), PartialFile.c_str());
+		     else
+		     {
+			partial[sp] = '\0';
+			sameFile = (DestFile == partial);
+		     }
+		  }
+	       }
+	       else
+		  _error->Errno("pkgAcqTransactionItem::TransactionState-stat", _("Failed to stat %s"), PartialFile.c_str());
+	    }
+	    if (sameFile == false)
 	    {
 	       // ensure that even without lists-cleanup all compressions are nuked
 	       std::string FinalFile = GetFinalFileNameFromURI(Target.URI);
@@ -2621,20 +2628,6 @@ void pkgAcqIndex::Failed(string const &Message,pkgAcquire::MethodConfig const *
       TransactionManager->AbortTransaction();
 }
 									/*}}}*/
-// AcqIndex::ReverifyAfterIMS - Reverify index after an ims-hit		/*{{{*/
-void pkgAcqIndex::ReverifyAfterIMS()
-{
-   // update destfile to *not* include the compression extension when doing
-   // a reverify (as its uncompressed on disk already)
-   DestFile = GetCompressedFileName(Target, GetPartialFileNameFromURI(Target.URI), CurrentCompressionExtension);
-
-   // copy FinalFile into partial/ so that we check the hash again
-   string FinalFile = GetFinalFilename();
-   Stage = STAGE_DECOMPRESS_AND_VERIFY;
-   Desc.URI = "copy:" + FinalFile;
-   QueueURI(Desc);
-}
-									/*}}}*/
 // AcqIndex::Done - Finished a fetch					/*{{{*/
 // ---------------------------------------------------------------------
 /* This goes through a number of states.. On the initial fetch the
@@ -2651,107 +2644,93 @@ void pkgAcqIndex::Done(string const &Message,
    switch(Stage) 
    {
       case STAGE_DOWNLOAD:
-         StageDownloadDone(Message, Hashes, Cfg);
+         StageDownloadDone(Message);
          break;
       case STAGE_DECOMPRESS_AND_VERIFY:
-         StageDecompressDone(Message, Hashes, Cfg);
+         StageDecompressDone();
          break;
    }
 }
 									/*}}}*/
 // AcqIndex::StageDownloadDone - Queue for decompress and verify	/*{{{*/
-void pkgAcqIndex::StageDownloadDone(string const &Message, HashStringList const &,
-                                    pkgAcquire::MethodConfig const * const)
+void pkgAcqIndex::StageDownloadDone(string const &Message)
 {
+   Local = true;
    Complete = true;
 
-   // Handle the unzipd case
-   std::string FileName = LookupTag(Message,"Alt-Filename");
-   if (FileName.empty() == false)
+   std::string const AltFilename = LookupTag(Message,"Alt-Filename");
+   std::string Filename = LookupTag(Message,"Filename");
+
+   // we need to verify the file against the current Release file again
+   // on if-modfied-since hit to avoid a stale attack against us
+   if(StringToBool(LookupTag(Message,"IMS-Hit"),false) == true)
    {
+      // copy FinalFile into partial/ so that we check the hash again
+      string const FinalFile = GetExistingFilename(GetFinalFileNameFromURI(Target.URI));
+      if (symlink(FinalFile.c_str(), DestFile.c_str()) != 0)
+	 _error->WarningE("pkgAcqIndex::StageDownloadDone", "Symlinking final file %s back to %s failed", FinalFile.c_str(), DestFile.c_str());
+      else
+      {
+	 EraseFileName = DestFile;
+	 Filename = DestFile;
+      }
       Stage = STAGE_DECOMPRESS_AND_VERIFY;
-      Local = true;
-      if (CurrentCompressionExtension != "uncompressed")
-	 DestFile.erase(DestFile.length() - (CurrentCompressionExtension.length() + 1));
-      Desc.URI = "copy:" + FileName;
+      Desc.URI = "store:" + Filename;
       QueueURI(Desc);
-      SetActiveSubprocess("copy");
+      SetActiveSubprocess(::URI(Desc.URI).Access);
       return;
    }
-   FileName = LookupTag(Message,"Filename");
-
+   // methods like file:// give us an alternative (uncompressed) file
+   else if (Target.KeepCompressed == false && AltFilename.empty() == false)
+   {
+      if (CurrentCompressionExtension != "uncompressed")
+	 DestFile.erase(DestFile.length() - (CurrentCompressionExtension.length() + 1));
+      Filename = AltFilename;
+   }
    // Methods like e.g. "file:" will give us a (compressed) FileName that is
    // not the "DestFile" we set, in this case we uncompress from the local file
-   if (FileName != DestFile && RealFileExists(DestFile) == false)
+   else if (Filename != DestFile && RealFileExists(DestFile) == false)
    {
-      Local = true;
-      if (Target.KeepCompressed == true)
-      {
-	 // but if we don't keep the uncompress we copy the compressed file first
-	 Stage = STAGE_DOWNLOAD;
-	 Desc.URI = "copy:" + FileName;
-	 QueueURI(Desc);
-	 SetActiveSubprocess("copy");
-	 return;
-      }
+      // symlinking ensures that the filename can be used for compression detection
+      // that is e.g. needed for by-hash which has no extension over file
+      if (symlink(Filename.c_str(),DestFile.c_str()) != 0)
+	 _error->WarningE("pkgAcqIndex::StageDownloadDone", "Symlinking file %s to %s failed", Filename.c_str(), DestFile.c_str());
       else
       {
-	 // symlinking ensures that the filename can be used for compression detection
-	 // that is e.g. needed for by-hash over file
-	 if (symlink(FileName.c_str(),DestFile.c_str()) != 0)
-	    _error->WarningE("pkgAcqIndex::StageDownloadDone", "Symlinking file %s to %s failed", FileName.c_str(), DestFile.c_str());
-	 else
-	 {
-	    EraseFileName = DestFile;
-	    FileName = DestFile;
-	 }
+	 EraseFileName = DestFile;
+	 Filename = DestFile;
       }
    }
-   else
-      EraseFileName = FileName;
-
-   // we need to verify the file against the current Release file again
-   // on if-modfied-since hit to avoid a stale attack against us
-   if(StringToBool(LookupTag(Message,"IMS-Hit"),false) == true)
-   {
-      // The files timestamp matches, reverify by copy into partial/
-      EraseFileName = "";
-      ReverifyAfterIMS();
-      return;
-   }
 
-   string decompProg = "store";
-   if (Target.KeepCompressed == true)
-   {
-      DestFile = "/dev/null";
-      EraseFileName.clear();
-   }
+   Stage = STAGE_DECOMPRESS_AND_VERIFY;
+   DestFile = GetKeepCompressedFileName(GetPartialFileNameFromURI(Target.URI), Target);
+   if (Filename != DestFile && flExtension(Filename) == flExtension(DestFile))
+      Desc.URI = "copy:" + Filename;
    else
+      Desc.URI = "store:" + Filename;
+   if (DestFile == Filename)
    {
       if (CurrentCompressionExtension == "uncompressed")
-	 decompProg = "copy";
-      else
-	 DestFile.erase(DestFile.length() - (CurrentCompressionExtension.length() + 1));
+	 return StageDecompressDone();
+      DestFile = "/dev/null";
    }
 
+   if (EraseFileName.empty())
+      EraseFileName = Filename;
+
    // queue uri for the next stage
-   Stage = STAGE_DECOMPRESS_AND_VERIFY;
-   Desc.URI = decompProg + ":" + FileName;
    QueueURI(Desc);
-   SetActiveSubprocess(decompProg);
+   SetActiveSubprocess(::URI(Desc.URI).Access);
 }
 									/*}}}*/
 // AcqIndex::StageDecompressDone - Final verification			/*{{{*/
-void pkgAcqIndex::StageDecompressDone(string const &,
-                                      HashStringList const &,
-                                      pkgAcquire::MethodConfig const * const)
+void pkgAcqIndex::StageDecompressDone()
 {
-   if (Target.KeepCompressed == true && DestFile == "/dev/null")
-      DestFile = GetPartialFileNameFromURI(Target.URI + '.' + CurrentCompressionExtension);
+   if (DestFile == "/dev/null")
+      DestFile = GetKeepCompressedFileName(GetPartialFileNameFromURI(Target.URI), Target);
 
    // Done, queue for rename on transaction finished
    TransactionManager->TransactionStageCopy(this, DestFile, GetFinalFilename());
-   return;
 }
 									/*}}}*/
 pkgAcqIndex::~pkgAcqIndex() {}
diff --git a/apt-pkg/acquire-item.h b/apt-pkg/acquire-item.h
index 93ff0a9c3..306b1772a 100644
--- a/apt-pkg/acquire-item.h
+++ b/apt-pkg/acquire-item.h
@@ -914,16 +914,12 @@ class APT_HIDDEN pkgAcqIndex : public pkgAcqBaseIndex
    AllStages Stage;
 
    /** \brief Handle what needs to be done when the download is done */
-   void StageDownloadDone(std::string const &Message,
-                          HashStringList const &Hashes,
-                          pkgAcquire::MethodConfig const * const Cfg);
+   void StageDownloadDone(std::string const &Message);
 
    /** \brief Handle what needs to be done when the decompression/copy is
     *         done 
     */
-   void StageDecompressDone(std::string const &Message,
-                            HashStringList const &Hashes,
-                            pkgAcquire::MethodConfig const * const Cfg);
+   void StageDecompressDone();
 
    /** \brief If \b set, this partially downloaded file will be
     *  removed when the download completes.
@@ -941,9 +937,6 @@ class APT_HIDDEN pkgAcqIndex : public pkgAcqBaseIndex
    /** \brief Do the changes needed to fetch via AptByHash (if needed) */
    void InitByHashIfNeeded();
 
-   /** \brief Schedule file for verification after a IMS hit */
-   void ReverifyAfterIMS();
-
    /** \brief Get the full pathname of the final file for the current URI */
    virtual std::string GetFinalFilename() const APT_OVERRIDE;
 
diff --git a/apt-pkg/aptconfiguration.cc b/apt-pkg/aptconfiguration.cc
index 53f29df79..01ae5a642 100644
--- a/apt-pkg/aptconfiguration.cc
+++ b/apt-pkg/aptconfiguration.cc
@@ -37,6 +37,7 @@ static void setDefaultConfigurationForCompressors() {
 	// Set default application paths to check for optional compression types
 	_config->CndSet("Dir::Bin::bzip2", "/bin/bzip2");
 	_config->CndSet("Dir::Bin::xz", "/usr/bin/xz");
+	_config->CndSet("Dir::Bin::lz4", "/usr/bin/lz4");
 	if (FileExists(_config->FindFile("Dir::Bin::xz")) == true) {
 		_config->Set("Dir::Bin::lzma", _config->FindFile("Dir::Bin::xz"));
 		_config->Set("APT::Compressor::lzma::Binary", "xz");
@@ -79,6 +80,7 @@ const Configuration::getCompressionTypes(bool const &Cached) {
 	_config->CndSet("Acquire::CompressionTypes::bz2","bzip2");
 	_config->CndSet("Acquire::CompressionTypes::lzma","lzma");
 	_config->CndSet("Acquire::CompressionTypes::gz","gzip");
+	_config->CndSet("Acquire::CompressionTypes::lz4","lz4");
 
 	setDefaultConfigurationForCompressors();
 	std::vector<APT::Configuration::Compressor> const compressors = getCompressors();
diff --git a/apt-pkg/deb/debmetaindex.cc b/apt-pkg/deb/debmetaindex.cc
index 46a9f9c97..2521fccd6 100644
--- a/apt-pkg/deb/debmetaindex.cc
+++ b/apt-pkg/deb/debmetaindex.cc
@@ -138,6 +138,21 @@ static void GetIndexTargetsFor(char const * const Type, std::string const &URI,
 	 DefCompressionTypes = os.str();
       }
    }
+   std::string DefKeepCompressedAs;
+   {
+      std::vector<APT::Configuration::Compressor> comps = APT::Configuration::getCompressors();
+      if (comps.empty() == false)
+      {
+	 std::sort(comps.begin(), comps.end(),
+	       [](APT::Configuration::Compressor const &a, APT::Configuration::Compressor const &b) { return a.Cost < b.Cost; });
+	 std::ostringstream os;
+	 for (auto const &c : comps)
+	    if (c.Cost != 0)
+	       os << c.Extension.substr(1) << ' ';
+	 DefKeepCompressedAs = os.str();
+      }
+      DefKeepCompressedAs += "uncompressed";
+   }
    std::string const NativeArch = _config->Find("APT::Architecture");
    bool const GzipIndex = _config->FindB("Acquire::GzipIndexes", false);
    for (std::vector<debReleaseIndexPrivate::debSectionEntry>::const_iterator E = entries.begin(); E != entries.end(); ++E)
@@ -155,11 +170,35 @@ static void GetIndexTargetsFor(char const * const Type, std::string const &URI,
 	 bool const UsePDiffs = APT_T_CONFIG_BOOL("PDiffs", E->UsePDiffs);
 	 std::string const UseByHash = APT_T_CONFIG_STR("By-Hash", E->UseByHash);
 	 std::string const CompressionTypes = APT_T_CONFIG_STR("CompressionTypes", DefCompressionTypes);
+	 std::string KeepCompressedAs = APT_T_CONFIG_STR("KeepCompressedAs", "");
 #undef APT_T_CONFIG_BOOL
 #undef APT_T_CONFIG_STR
 	 if (tplMetaKey.empty())
 	    continue;
 
+	 if (KeepCompressedAs.empty())
+	    KeepCompressedAs = DefKeepCompressedAs;
+	 else
+	 {
+	    std::vector<std::string> const defKeep = VectorizeString(DefKeepCompressedAs, ' ');
+	    std::vector<std::string> const valKeep = VectorizeString(KeepCompressedAs, ' ');
+	    std::vector<std::string> keep;
+	    for (auto const &val : valKeep)
+	    {
+	       if (val.empty())
+		  continue;
+	       if (std::find(defKeep.begin(), defKeep.end(), val) == defKeep.end())
+		  continue;
+	       keep.push_back(val);
+	    }
+	    if (std::find(keep.begin(), keep.end(), "uncompressed") == keep.end())
+	       keep.push_back("uncompressed");
+	    std::ostringstream os;
+	    std::copy(keep.begin(), keep.end()-1, std::ostream_iterator<std::string>(os, " "));
+	    os << *keep.rbegin();
+	    KeepCompressedAs = os.str();
+	 }
+
 	 for (std::vector<std::string>::const_iterator L = E->Languages.begin(); L != E->Languages.end(); ++L)
 	 {
 	    if (*L == "none" && tplMetaKey.find("$(LANGUAGE)") != std::string::npos)
@@ -253,6 +292,7 @@ static void GetIndexTargetsFor(char const * const Type, std::string const &URI,
 	       Options.insert(std::make_pair("BY_HASH", UseByHash));
 	       Options.insert(std::make_pair("DEFAULTENABLED", DefaultEnabled ? "yes" : "no"));
 	       Options.insert(std::make_pair("COMPRESSIONTYPES", CompressionTypes));
+	       Options.insert(std::make_pair("KEEPCOMPRESSEDAS", KeepCompressedAs));
 	       Options.insert(std::make_pair("SOURCESENTRY", E->sourcesEntry));
 
 	       bool IsOpt = IsOptional;
diff --git a/apt-pkg/indexfile.cc b/apt-pkg/indexfile.cc
index f0fae907f..dfd482c04 100644
--- a/apt-pkg/indexfile.cc
+++ b/apt-pkg/indexfile.cc
@@ -144,6 +144,7 @@ std::string IndexTarget::Option(OptionKeys const EnumKey) const		/*{{{*/
       APT_CASE(COMPRESSIONTYPES);
       APT_CASE(SOURCESENTRY);
       APT_CASE(BY_HASH);
+      APT_CASE(KEEPCOMPRESSEDAS);
 #undef APT_CASE
       case FILENAME: return _config->FindDir("Dir::State::lists") + URItoFileName(URI);
       case EXISTING_FILENAME:
diff --git a/apt-pkg/indexfile.h b/apt-pkg/indexfile.h
index f87693a8f..5c666671c 100644
--- a/apt-pkg/indexfile.h
+++ b/apt-pkg/indexfile.h
@@ -90,6 +90,7 @@ class IndexTarget							/*{{{*/
       DEFAULTENABLED,
       SOURCESENTRY,
       BY_HASH,
+      KEEPCOMPRESSEDAS,
    };
    std::string Option(OptionKeys const Key) const;
    bool OptionBool(OptionKeys const Key) const;