From 5576e7f76da73f3f5217f90d816cc19b6c0a5a77 Mon Sep 17 00:00:00 2001
From: Julian Andres Klode <julian.klode@canonical.com>
Date: Tue, 1 Aug 2023 13:59:09 +0200
Subject: Compare SHA256 to check if versions are really the same

If we know both SHA256, and they're different, the packages are. This
approach stores the SHA256 only at runtime, avoiding the overhead of
storing it on-disk, because when we update repositories we update all
of them anyhow.

Note that pkgCacheGenerator is hidden, so we can just modify its
ABI, hooray.

Closes: #931175
LP: #2029268
---
 apt-pkg/pkgcachegen.cc | 27 +++++++++++++++++++++++++--
 1 file changed, 25 insertions(+), 2 deletions(-)

(limited to 'apt-pkg/pkgcachegen.cc')

diff --git a/apt-pkg/pkgcachegen.cc b/apt-pkg/pkgcachegen.cc
index 807f3bf6c..9e47ef369 100644
--- a/apt-pkg/pkgcachegen.cc
+++ b/apt-pkg/pkgcachegen.cc
@@ -12,6 +12,7 @@
 #include <config.h>
 
 #include <apt-pkg/configuration.h>
+#include <apt-pkg/deblistparser.h>
 #include <apt-pkg/error.h>
 #include <apt-pkg/fileutl.h>
 #include <apt-pkg/indexfile.h>
@@ -376,6 +377,12 @@ bool pkgCacheGenerator::MergeListVersion(ListParser &List, pkgCache::PkgIterator
    void const * oldMap = Map.Data();
 
    auto Hash = List.VersionHash();
+   APT::StringView ListSHA256;
+
+   bool const Debug = _config->FindB("Debug::pkgCacheGen", false);
+   auto DebList = dynamic_cast<debListParser *>(&List);
+   if (DebList != nullptr)
+      ListSHA256 = DebList->SHA256();
    if (Ver.end() == false)
    {
       /* We know the list is sorted so we use that fact in the search.
@@ -392,8 +399,18 @@ bool pkgCacheGenerator::MergeListVersion(ListParser &List, pkgCache::PkgIterator
 	 // Versionstrings are equal - is hash also equal?
 	 if (Res == 0)
 	 {
-	    if (List.SameVersion(Hash, Ver) == true)
-	       break;
+	    if (List.SameVersion(Hash, Ver))
+	    {
+               // We do not have SHA256 for both, so we cannot compare them, trust the call from SameVersion()
+               if (ListSHA256.empty() || VersionExtra[Ver->ID].SHA256[0] == 0)
+                  break;
+               // We have SHA256 for both, so they must match.
+               if (ListSHA256 == APT::StringView(VersionExtra[Ver->ID].SHA256, 64))
+                  break;
+	       if (Debug)
+		  std::cerr << "Found differing SHA256 for " << Pkg.Name() << "=" << Version.to_string() << std::endl;
+	    }
+
 	    // sort (volatile) sources above not-sources like the status file
 	    if (CurrentFile == nullptr || (CurrentFile->Flags & pkgCache::Flag::NotSource) == 0)
 	    {
@@ -440,6 +457,8 @@ bool pkgCacheGenerator::MergeListVersion(ListParser &List, pkgCache::PkgIterator
    if (oldMap != Map.Data())
 	 LastVer = static_cast<map_pointer<pkgCache::Version> *>(Map.Data()) + (LastVer - static_cast<map_pointer<pkgCache::Version> const *>(oldMap));
    *LastVer = verindex;
+   if (ListSHA256.size() == 64)
+	 memcpy(VersionExtra[Ver->ID].SHA256, ListSHA256.data(), 64);
 
    if (unlikely(List.NewVersion(Ver) == false))
       return _error->Error(_("Error occurred while processing %s (%s%d)"),
@@ -869,6 +888,10 @@ map_pointer<pkgCache::Version> pkgCacheGenerator::NewVersion(pkgCache::VerIterat
    Ver->Hash = Hash;
    Ver->ID = Cache.HeaderP->VersionCount++;
 
+   // Allocate size for extra store
+   if (VersionExtra.size() <= Ver->ID)
+      VersionExtra.resize(Ver->ID + 1);
+
    // try to find the version string in the group for reuse
    pkgCache::PkgIterator Pkg = Ver.ParentPkg();
    pkgCache::GrpIterator Grp = Pkg.Group();
-- 
cgit v1.2.3-70-g09d2