* Store the size of a store path in the database (to be precise, the

  size of the NAR serialisation of the path, i.e., `nix-store --dump
  PATH').  This is useful for Hydra.

16 files changed, 144 insertions, 91 deletions

diff --git a/externals/Makefile.am b/externals/Makefile.am
index 060ae2c52..bdc29fcb5 100644
--- a/externals/Makefile.am
+++ b/externals/Makefile.am
@@ -47,7 +47,7 @@ build-sqlite:
 else
 build-sqlite: $(SQLITE)
 	(cd $(SQLITE) && \
-	CC="$(CC)" ./configure --disable-static --prefix=$(pkglibdir)/dummy --libdir=${pkglibdir} $(SUB_CONFIGURE_FLAGS) && \
+	CC="$(CC)" CFLAGS="-DSQLITE_ENABLE_COLUMN_METADATA=1" ./configure --disable-static --prefix=$(pkglibdir)/dummy --libdir=${pkglibdir} $(SUB_CONFIGURE_FLAGS) && \
 	$(MAKE) )
 	touch build-sqlite
 
diff --git a/src/libstore/build.cc b/src/libstore/build.cc
index 210486fbc..8b8be3e80 100644
--- a/src/libstore/build.cc
+++ b/src/libstore/build.cc
@@ -1547,7 +1547,7 @@ void DerivationGoal::startBuilder()
 
         /* Write closure info to `fileName'. */
         writeFile(tmpDir + "/" + fileName,
-            makeValidityRegistration(paths, false, false));
+            worker.store.makeValidityRegistration(paths, false, false));
     }
 
     
@@ -1870,7 +1870,7 @@ PathSet parseReferenceSpecifiers(const Derivation & drv, string attr)
 void DerivationGoal::computeClosure()
 {
     map<Path, PathSet> allReferences;
-    map<Path, Hash> contentHashes;
+    map<Path, HashResult> contentHashes;
 
     /* When using a build hook, the build hook can register the output
        as valid (by doing `nix-store --import').  If so we don't have
@@ -1927,7 +1927,7 @@ void DerivationGoal::computeClosure()
             if (ht == htUnknown)
                 throw BuildError(format("unknown hash algorithm `%1%'") % algo);
             Hash h = parseHash(ht, i->second.hash);
-            Hash h2 = recursive ? hashPath(ht, path) : hashFile(ht, path);
+            Hash h2 = recursive ? hashPath(ht, path).first : hashFile(ht, path);
             if (h != h2)
                 throw BuildError(
                     format("output path `%1%' should have %2% hash `%3%', instead has `%4%'")
@@ -1941,7 +1941,7 @@ void DerivationGoal::computeClosure()
 	   contained in it.  Compute the SHA-256 NAR hash at the same
 	   time.  The hash is stored in the database so that we can
 	   verify later on whether nobody has messed with the store. */
-        Hash hash;
+        HashResult hash;
         PathSet references = scanForReferences(path, allPaths, hash);
         contentHashes[path] = hash;
 
@@ -1970,14 +1970,18 @@ void DerivationGoal::computeClosure()
     }
 
     /* Register each output path as valid, and register the sets of
-       paths referenced by each of them.  !!! this should be
-       atomic so that either all paths are registered as valid, or
-       none are. */
-    foreach (DerivationOutputs::iterator, i, drv.outputs)
-        worker.store.registerValidPath(i->second.path,
-            contentHashes[i->second.path],
-            allReferences[i->second.path],
-            drvPath);
+       paths referenced by each of them. */
+    ValidPathInfos infos;
+    foreach (DerivationOutputs::iterator, i, drv.outputs) {
+        ValidPathInfo info;
+        info.path = i->second.path;
+        info.hash = contentHashes[i->second.path].first;
+        info.narSize = contentHashes[i->second.path].second;
+        info.references = allReferences[i->second.path];
+        info.deriver = drvPath;
+        infos.push_back(info);
+    }
+    worker.store.registerValidPaths(infos);
 
     /* It is now safe to delete the lock files, since all future
        lockers will see that the output paths are valid; they will not
@@ -2385,10 +2389,15 @@ void SubstitutionGoal::finished()
 
     canonicalisePathMetaData(storePath);
 
-    Hash contentHash = hashPath(htSHA256, storePath);
-
-    worker.store.registerValidPath(storePath, contentHash,
-        info.references, info.deriver);
+    HashResult hash = hashPath(htSHA256, storePath);
+    
+    ValidPathInfo info2;
+    info2.path = storePath;
+    info2.hash = hash.first;
+    info2.narSize = hash.second;
+    info2.references = info.references;
+    info2.deriver = info.deriver;
+    worker.store.registerValidPath(info2);
 
     outputLock->setDeletion(true);
     
diff --git a/src/libstore/local-store.cc b/src/libstore/local-store.cc
index c0c75e34d..9595561bc 100644
--- a/src/libstore/local-store.cc
+++ b/src/libstore/local-store.cc
@@ -327,9 +327,16 @@ void LocalStore::openDB(bool create)
             throw SQLiteError(db, "initialising database schema");
     }
 
+    /* Backwards compatibility with old (pre-release) databases.  Can
+       remove this eventually. */
+    if (sqlite3_table_column_metadata(db, 0, "ValidPaths", "narSize", 0, 0, 0, 0, 0) != SQLITE_OK) {
+        if (sqlite3_exec(db, "alter table ValidPaths add column narSize integer" , 0, 0, 0) != SQLITE_OK)
+            throw SQLiteError(db, "adding column narSize");
+    }
+
     /* Prepare SQL statements. */
     stmtRegisterValidPath.create(db,
-        "insert into ValidPaths (path, hash, registrationTime, deriver) values (?, ?, ?, ?);");
+        "insert into ValidPaths (path, hash, registrationTime, deriver, narSize) values (?, ?, ?, ?, ?);");
     stmtAddReference.create(db,
         "insert or replace into Refs (referrer, reference) values (?, ?);");
     stmtQueryPathInfo.create(db,
@@ -431,19 +438,6 @@ void canonicalisePathMetaData(const Path & path)
 }
 
 
-void LocalStore::registerValidPath(const Path & path,
-    const Hash & hash, const PathSet & references,
-    const Path & deriver)
-{
-    ValidPathInfo info;
-    info.path = path;
-    info.hash = hash;
-    info.references = references;
-    info.deriver = deriver;
-    registerValidPath(info);
-}
-
-
 unsigned long long LocalStore::addValidPath(const ValidPathInfo & info)
 {
     SQLiteStmtUse use(stmtRegisterValidPath);
@@ -454,6 +448,10 @@ unsigned long long LocalStore::addValidPath(const ValidPathInfo & info)
         stmtRegisterValidPath.bind(info.deriver);
     else
         stmtRegisterValidPath.bind(); // null
+    if (info.narSize != 0)
+        stmtRegisterValidPath.bind(info.narSize);
+    else
+        stmtRegisterValidPath.bind(); // null
     if (sqlite3_step(stmtRegisterValidPath) != SQLITE_DONE)
         throw SQLiteError(db, format("registering valid path `%1%' in database") % info.path);
     unsigned long long id = sqlite3_last_insert_rowid(db);
@@ -920,10 +918,18 @@ Path LocalStore::addToStoreFromDump(const string & dump, const string & name,
                the path in the database.  We may just have computed it
                above (if called with recursive == true and hashAlgo ==
                sha256); otherwise, compute it here. */
-            registerValidPath(dstPath,
-                (recursive && hashAlgo == htSHA256) ? h :
-                (recursive ? hashString(htSHA256, dump) : hashPath(htSHA256, dstPath)),
-                PathSet(), "");
+            HashResult hash;
+            if (recursive) {
+                hash.first = hashAlgo == htSHA256 ? h : hashString(htSHA256, dump);
+                hash.second = dump.size();
+            } else
+                hash = hashPath(htSHA256, dstPath);
+            
+            ValidPathInfo info;
+            info.path = dstPath;
+            info.hash = hash.first;
+            info.narSize = hash.second;
+            registerValidPath(info);
         }
 
         outputLock.setDeletion(true);
@@ -970,9 +976,15 @@ Path LocalStore::addTextToStore(const string & name, const string & s,
             writeFile(dstPath, s);
 
             canonicalisePathMetaData(dstPath);
+
+            HashResult hash = hashPath(htSHA256, dstPath);
             
-            registerValidPath(dstPath,
-                hashPath(htSHA256, dstPath), references, "");
+            ValidPathInfo info;
+            info.path = dstPath;
+            info.hash = hash.first;
+            info.narSize = hash.second;
+            info.references = references;
+            registerValidPath(info);
         }
 
         outputLock.setDeletion(true);
@@ -998,7 +1010,7 @@ struct HashAndWriteSink : Sink
     Hash currentHash()
     {
         HashSink hashSinkClone(hashSink);
-        return hashSinkClone.finish();
+        return hashSinkClone.finish().first;
     }
 };
 
@@ -1136,7 +1148,7 @@ Path LocalStore::importPath(bool requireSignature, Source & source)
     Path deriver = readString(hashAndReadSource);
     if (deriver != "") assertStorePath(deriver);
 
-    Hash hash = hashAndReadSource.hashSink.finish();
+    Hash hash = hashAndReadSource.hashSink.finish().first;
     hashAndReadSource.hashing = false;
 
     bool haveSignature = readInt(hashAndReadSource) == 1;
@@ -1200,9 +1212,15 @@ Path LocalStore::importPath(bool requireSignature, Source & source)
             
             /* !!! if we were clever, we could prevent the hashPath()
                here. */
-            if (deriver != "" && !isValidPath(deriver)) deriver = "";
-            registerValidPath(dstPath,
-                hashPath(htSHA256, dstPath), references, deriver);
+            HashResult hash = hashPath(htSHA256, dstPath);
+            
+            ValidPathInfo info;
+            info.path = dstPath;
+            info.hash = hash.first;
+            info.narSize = hash.second;
+            info.references = references;
+            info.deriver = deriver != "" && isValidPath(deriver) ? deriver : "";
+            registerValidPath(info);
         }
         
         outputLock.setDeletion(true);
@@ -1263,12 +1281,14 @@ void LocalStore::verifyStore(bool checkContents)
 
             /* Check the content hash (optionally - slow). */
             printMsg(lvlTalkative, format("checking contents of `%1%'") % *i);
-            Hash current = hashPath(info.hash.type, *i);
+            Hash current = hashPath(info.hash.type, *i).first;
             if (current != info.hash) {
                 printMsg(lvlError, format("path `%1%' was modified! "
                         "expected hash `%2%', got `%3%'")
                     % *i % printHash(info.hash) % printHash(current));
             }
+            
+            /* !!! Check info.narSize */
         }
     }
 }
diff --git a/src/libstore/local-store.hh b/src/libstore/local-store.hh
index 0d7ec1f49..2fd31c26d 100644
--- a/src/libstore/local-store.hh
+++ b/src/libstore/local-store.hh
@@ -97,6 +97,8 @@ public:
 
     PathSet queryValidPaths();
     
+    ValidPathInfo queryPathInfo(const Path & path);
+
     Hash queryPathHash(const Path & path);
 
     void queryReferences(const Path & path, PathSet & references);
@@ -173,8 +175,7 @@ public:
        execution of the derivation (or something equivalent).  Also
        register the hash of the file system contents of the path.  The
        hash must be a SHA-256 hash. */
-    void registerValidPath(const Path & path,
-        const Hash & hash, const PathSet & references, const Path & deriver);
+    void registerValidPath(const ValidPathInfo & info);
 
     void registerValidPaths(const ValidPathInfos & infos);
 
@@ -224,10 +225,6 @@ private:
         
     void addReference(unsigned long long referrer, unsigned long long reference);
     
-    void registerValidPath(const ValidPathInfo & info);
-
-    ValidPathInfo queryPathInfo(const Path & path);
-
     void appendReferrer(const Path & from, const Path & to, bool lock);
     
     void rewriteReferrers(const Path & path, bool purge, PathSet referrers);
diff --git a/src/libstore/optimise-store.cc b/src/libstore/optimise-store.cc
index 3ed54e24d..89be6ac65 100644
--- a/src/libstore/optimise-store.cc
+++ b/src/libstore/optimise-store.cc
@@ -68,7 +68,7 @@ static void hashAndLink(bool dryRun, HashToPath & hashToPath,
            the contents of the symlink (i.e. the result of
            readlink()), not the contents of the target (which may not
            even exist). */
-        Hash hash = hashPath(htSHA256, path);
+        Hash hash = hashPath(htSHA256, path).first;
         stats.totalFiles++;
         printMsg(lvlDebug, format("`%1%' has hash `%2%'") % path % printHash(hash));
 
diff --git a/src/libstore/references.cc b/src/libstore/references.cc
index a6f6e85fc..ade9c9aa2 100644
--- a/src/libstore/references.cc
+++ b/src/libstore/references.cc
@@ -81,7 +81,7 @@ void RefScanSink::operator () (const unsigned char * data, unsigned int len)
 
 
 PathSet scanForReferences(const string & path,
-    const PathSet & refs, Hash & hash)
+    const PathSet & refs, HashResult & hash)
 {
     RefScanSink sink;
     std::map<string, Path> backMap;
diff --git a/src/libstore/references.hh b/src/libstore/references.hh
index 7d068eb51..158c08a77 100644
--- a/src/libstore/references.hh
+++ b/src/libstore/references.hh
@@ -7,7 +7,7 @@
 namespace nix {
 
 PathSet scanForReferences(const Path & path, const PathSet & refs,
-    Hash & hash);
+    HashResult & hash);
     
 }
 
diff --git a/src/libstore/remote-store.cc b/src/libstore/remote-store.cc
index 92d517bbb..517c886b4 100644
--- a/src/libstore/remote-store.cc
+++ b/src/libstore/remote-store.cc
@@ -247,6 +247,12 @@ bool RemoteStore::querySubstitutablePathInfo(const Path & path,
 }
 
 
+ValidPathInfo RemoteStore::queryPathInfo(const Path & path)
+{
+    throw Error("not implemented");
+}
+
+
 Hash RemoteStore::queryPathHash(const Path & path)
 {
     openConnection();
diff --git a/src/libstore/remote-store.hh b/src/libstore/remote-store.hh
index 02a1c4752..519f46fd1 100644
--- a/src/libstore/remote-store.hh
+++ b/src/libstore/remote-store.hh
@@ -29,6 +29,8 @@ public:
 
     PathSet queryValidPaths();
     
+    ValidPathInfo queryPathInfo(const Path & path);
+
     Hash queryPathHash(const Path & path);
 
     void queryReferences(const Path & path, PathSet & references);
diff --git a/src/libstore/schema.sql b/src/libstore/schema.sql
index 7438632ed..c1b4a689a 100644
--- a/src/libstore/schema.sql
+++ b/src/libstore/schema.sql
@@ -3,7 +3,8 @@ create table if not exists ValidPaths (
     path             text unique not null,
     hash             text not null,
     registrationTime integer not null,
-    deriver          text
+    deriver          text,
+    narSize          integer
 );
 
 create table if not exists Refs (
diff --git a/src/libstore/store-api.cc b/src/libstore/store-api.cc
index 01dd51621..4b04f5751 100644
--- a/src/libstore/store-api.cc
+++ b/src/libstore/store-api.cc
@@ -190,7 +190,7 @@ std::pair<Path, Hash> computeStorePathForPath(const Path & srcPath,
     bool recursive, HashType hashAlgo, PathFilter & filter)
 {
     HashType ht(hashAlgo);
-    Hash h = recursive ? hashPath(ht, srcPath, filter) : hashFile(ht, srcPath);
+    Hash h = recursive ? hashPath(ht, srcPath, filter).first : hashFile(ht, srcPath);
     string name = baseNameOf(srcPath);
     Path dstPath = makeFixedOutputPath(recursive, hashAlgo, h, name);
     return std::pair<Path, Hash>(dstPath, h);
@@ -216,7 +216,7 @@ Path computeStorePathForText(const string & name, const string & s,
 /* Return a string accepted by decodeValidPathInfo() that
    registers the specified paths as valid.  Note: it's the
    responsibility of the caller to provide a closure. */
-string makeValidityRegistration(const PathSet & paths,
+string StoreAPI::makeValidityRegistration(const PathSet & paths,
     bool showDerivers, bool showHash)
 {
     string s = "";
@@ -224,18 +224,19 @@ string makeValidityRegistration(const PathSet & paths,
     foreach (PathSet::iterator, i, paths) {
         s += *i + "\n";
 
-        if (showHash)
-            s += printHash(store->queryPathHash(*i)) + "\n";
+        ValidPathInfo info = queryPathInfo(*i);
 
-        Path deriver = showDerivers ? store->queryDeriver(*i) : "";
+        if (showHash) {
+            s += printHash(info.hash) + "\n";
+            s += (format("%1%\n") % info.narSize).str();
+        }
+
+        Path deriver = showDerivers ? info.deriver : "";
         s += deriver + "\n";
 
-        PathSet references;
-        store->queryReferences(*i, references);
+        s += (format("%1%\n") % info.references.size()).str();
 
-        s += (format("%1%\n") % references.size()).str();
-            
-        foreach (PathSet::iterator, j, references)
+        foreach (PathSet::iterator, j, info.references)
             s += *j + "\n";
     }
 
@@ -252,6 +253,8 @@ ValidPathInfo decodeValidPathInfo(std::istream & str, bool hashGiven)
         string s;
         getline(str, s);
         info.hash = parseHash(htSHA256, s);
+        getline(str, s);
+        if (!string2Int(s, info.narSize)) throw Error("number expected");
     }
     getline(str, info.deriver);
     string s; int n;
diff --git a/src/libstore/store-api.hh b/src/libstore/store-api.hh
index 082a9edc4..b0071da83 100644
--- a/src/libstore/store-api.hh
+++ b/src/libstore/store-api.hh
@@ -90,6 +90,21 @@ struct SubstitutablePathInfo
 };
 
 
+struct ValidPathInfo 
+{
+    Path path;
+    Path deriver;
+    Hash hash;
+    PathSet references;
+    time_t registrationTime;
+    unsigned long long narSize; // 0 = unknown
+    unsigned long long id; // internal use only
+    ValidPathInfo() : registrationTime(0), narSize(0) { }
+};
+
+typedef list<ValidPathInfo> ValidPathInfos;
+
+
 class StoreAPI 
 {
 public:
@@ -102,6 +117,9 @@ public:
     /* Query the set of valid paths. */
     virtual PathSet queryValidPaths() = 0;
 
+    /* Query information about a valid path. */
+    virtual ValidPathInfo queryPathInfo(const Path & path) = 0;
+
     /* Queries the hash of a valid path. */ 
     virtual Hash queryPathHash(const Path & path) = 0;
 
@@ -214,6 +232,12 @@ public:
     /* Clear the "failed" status of the given paths.  The special
        value `*' causes all failed paths to be cleared. */
     virtual void clearFailedPaths(const PathSet & paths) = 0;
+
+    /* Return a string representing information about the path that
+       can be loaded into the database using `nix-store --load-db' or
+       `nix-store --register-validity'. */
+    string makeValidityRegistration(const PathSet & paths,
+        bool showDerivers, bool showHash);
 };
 
 
@@ -307,22 +331,6 @@ boost::shared_ptr<StoreAPI> openStore();
 string showPaths(const PathSet & paths);
 
 
-string makeValidityRegistration(const PathSet & paths,
-    bool showDerivers, bool showHash);
-    
-struct ValidPathInfo 
-{
-    Path path;
-    Path deriver;
-    Hash hash;
-    PathSet references;
-    time_t registrationTime;
-    unsigned long long id; // internal use only
-    ValidPathInfo() : registrationTime(0) { }
-};
-
-typedef list<ValidPathInfo> ValidPathInfos;
-
 ValidPathInfo decodeValidPathInfo(std::istream & str,
     bool hashGiven = false);
 
diff --git a/src/libutil/hash.cc b/src/libutil/hash.cc
index bd7e33a48..b9e784699 100644
--- a/src/libutil/hash.cc
+++ b/src/libutil/hash.cc
@@ -286,12 +286,14 @@ Hash hashFile(HashType ht, const Path & path)
 HashSink::HashSink(HashType ht) : ht(ht)
 {
     ctx = new Ctx;
+    bytes = 0;
     start(ht, *ctx);
 }
     
 HashSink::HashSink(const HashSink & h)
 {
     ht = h.ht;
+    bytes = h.bytes;
     ctx = new Ctx;
     *ctx = *h.ctx;
 }
@@ -304,18 +306,20 @@ HashSink::~HashSink()
 void HashSink::operator ()
     (const unsigned char * data, unsigned int len)
 {
+    bytes += len;
     update(ht, *ctx, data, len);
 }
 
-Hash HashSink::finish()
+HashResult HashSink::finish()
 {
     Hash hash(ht);
     nix::finish(ht, *ctx, hash.hash);
-    return hash;
+    return HashResult(hash, bytes);
 }
 
 
-Hash hashPath(HashType ht, const Path & path, PathFilter & filter)
+HashResult hashPath(
+    HashType ht, const Path & path, PathFilter & filter)
 {
     HashSink sink(ht);
     dumpPath(path, sink, filter);
diff --git a/src/libutil/hash.hh b/src/libutil/hash.hh
index 81425b234..13740954d 100644
--- a/src/libutil/hash.hh
+++ b/src/libutil/hash.hh
@@ -40,7 +40,6 @@ struct Hash
 
     /* For sorting. */
     bool operator < (const Hash & h) const;
-
 };
 
 
@@ -72,7 +71,8 @@ Hash hashFile(HashType ht, const Path & path);
    (essentially) hashString(ht, dumpPath(path)). */
 struct PathFilter;
 extern PathFilter defaultPathFilter;
-Hash hashPath(HashType ht, const Path & path,
+typedef std::pair<Hash, unsigned long long> HashResult;
+HashResult hashPath(HashType ht, const Path & path,
     PathFilter & filter = defaultPathFilter);
 
 /* Compress a hash to the specified number of bytes by cyclically
@@ -93,13 +93,14 @@ class HashSink : public Sink
 private:
     HashType ht;
     Ctx * ctx;
+    unsigned long long bytes;
 
 public:
     HashSink(HashType ht);
     HashSink(const HashSink & h);
     ~HashSink();
     virtual void operator () (const unsigned char * data, unsigned int len);
-    Hash finish();
+    HashResult finish();
 };
 
 
diff --git a/src/nix-hash/nix-hash.cc b/src/nix-hash/nix-hash.cc
index 8a6655403..4867234bf 100644
--- a/src/nix-hash/nix-hash.cc
+++ b/src/nix-hash/nix-hash.cc
@@ -44,7 +44,7 @@ void run(Strings args)
 
     if (op == opHash) {
         for (Strings::iterator i = ss.begin(); i != ss.end(); ++i) {
-            Hash h = flat ? hashFile(ht, *i) : hashPath(ht, *i);
+            Hash h = flat ? hashFile(ht, *i) : hashPath(ht, *i).first;
             if (truncate && h.hashSize > 20) h = compressHash(h, 20);
             std::cout << format("%1%\n") %
                 (base32 ? printHash32(h) : printHash(h));
diff --git a/src/nix-store/nix-store.cc b/src/nix-store/nix-store.cc
index 695eb10dc..120f6ce72 100644
--- a/src/nix-store/nix-store.cc
+++ b/src/nix-store/nix-store.cc
@@ -393,9 +393,8 @@ static void opDumpDB(Strings opFlags, Strings opArgs)
     if (!opArgs.empty())
         throw UsageError("no arguments expected");
     PathSet validPaths = store->queryValidPaths();
-    foreach (PathSet::iterator, i, validPaths) {
-        cout << makeValidityRegistration(singleton<PathSet>(*i), true, true);
-    }
+    foreach (PathSet::iterator, i, validPaths)
+        cout << store->makeValidityRegistration(singleton<PathSet>(*i), true, true);
 }
 
 
@@ -410,8 +409,11 @@ static void registerValidity(bool reregister, bool hashGiven, bool canonicalise)
             /* !!! races */
             if (canonicalise)
                 canonicalisePathMetaData(info.path);
-            if (!hashGiven)
-                info.hash = hashPath(htSHA256, info.path);
+            if (!hashGiven) {
+                HashResult hash = hashPath(htSHA256, info.path);
+                info.hash = hash.first;
+                info.narSize = hash.second;
+            }
             infos.push_back(info);
         }
     }