streams: cache file position within AutoFile

Author: sipa

src/bench/streams_findbyte.cpp

@@ -19,7 +19,7 @@ static void FindByte(benchmark::Bench& bench)
     uint8_t data[file_size] = {0};
     data[file_size-1] = 1;
     file << data;
-    std::rewind(file.Get());
+    file.seek(0, SEEK_SET);
     BufferedFile bf{file, /*nBufSize=*/file_size + 1, /*nRewindIn=*/file_size};
 
     bench.run([&] {

src/index/txindex.cpp

@@ -87,10 +87,7 @@ bool TxIndex::FindTx(const uint256& tx_hash, uint256& block_hash, CTransactionRe
     CBlockHeader header;
     try {
         file >> header;
-        if (fseek(file.Get(), postx.nTxOffset, SEEK_CUR)) {
-            LogError("%s: fseek(...) failed\n", __func__);
-            return false;
-        }
+        file.seek(postx.nTxOffset, SEEK_CUR);
         file >> TX_WITH_WITNESS(tx);
     } catch (const std::exception& e) {
         LogError("%s: Deserialize or I/O error - %s\n", __func__, e.what());

src/node/blockstorage.cpp

@@ -683,7 +683,7 @@ bool BlockManager::UndoWriteToDisk(const CBlockUndo& blockundo, FlatFilePos& pos
     fileout << GetParams().MessageStart() << nSize;
 
     // Write undo data
-    long fileOutPos = ftell(fileout.Get());
+    long fileOutPos = fileout.tell();
     if (fileOutPos < 0) {
         LogError("%s: ftell failed\n", __func__);
         return false;
@@ -981,7 +981,7 @@ bool BlockManager::WriteBlockToDisk(const CBlock& block, FlatFilePos& pos) const
     fileout << GetParams().MessageStart() << nSize;
 
     // Write block
-    long fileOutPos = ftell(fileout.Get());
+    long fileOutPos = fileout.tell();
     if (fileOutPos < 0) {
         LogError("%s: ftell failed\n", __func__);
         return false;

src/node/utxo_snapshot.cpp

@@ -73,7 +73,9 @@ std::optional<uint256> ReadSnapshotBaseBlockhash(fs::path chaindir)
     }
     afile >> base_blockhash;
 
-    if (std::fgetc(afile.Get()) != EOF) {
+    int64_t position = afile.tell();
+    afile.seek(0, SEEK_END);
+    if (position != afile.tell()) {
         LogPrintf("[snapshot] warning: unexpected trailing data in %s\n", read_from_str);
     } else if (std::ferror(afile.Get())) {
         LogPrintf("[snapshot] warning: i/o error reading %s\n", read_from_str);

src/streams.cpp

@@ -7,18 +7,25 @@
 
 #include <array>
 
+AutoFile::AutoFile(std::FILE* file, std::vector<std::byte> data_xor)
+    : m_file{file}, m_xor{std::move(data_xor)}
+{
+    if (!IsNull()) {
+        auto pos{std::ftell(m_file)};
+        if (pos >= 0) m_position = pos;
+    }
+}
+
 std::size_t AutoFile::detail_fread(Span<std::byte> dst)
 {
     if (!m_file) throw std::ios_base::failure("AutoFile::read: file handle is nullptr");
-    if (m_xor.empty()) {
-        return std::fread(dst.data(), 1, dst.size(), m_file);
-    } else {
-        const auto init_pos{std::ftell(m_file)};
-        if (init_pos < 0) throw std::ios_base::failure("AutoFile::read: ftell failed");
-        std::size_t ret{std::fread(dst.data(), 1, dst.size(), m_file)};
-        util::Xor(dst.subspan(0, ret), m_xor, init_pos);
-        return ret;
+    size_t ret = std::fread(dst.data(), 1, dst.size(), m_file);
+    if (!m_xor.empty()) {
+        if (!m_position.has_value()) throw std::ios_base::failure("AutoFile::read: position unknown");
+        util::Xor(dst.subspan(0, ret), m_xor, *m_position);
     }
+    if (m_position.has_value()) *m_position += ret;
+    return ret;
 }
 
 void AutoFile::seek(int64_t offset, int origin)
@@ -29,18 +36,23 @@ void AutoFile::seek(int64_t offset, int origin)
     if (std::fseek(m_file, offset, origin) != 0) {
         throw std::ios_base::failure(feof() ? "AutoFile::seek: end of file" : "AutoFile::seek: fseek failed");
     }
+    if (origin == SEEK_SET) {
+        m_position = offset;
+    } else if (origin == SEEK_CUR && m_position.has_value()) {
+        *m_position += offset;
+    } else {
+        int64_t r{std::ftell(m_file)};
+        if (r < 0) {
+            throw std::ios_base::failure("AutoFile::seek: ftell failed");
+        }
+        m_position = r;
+    }
 }
 
 int64_t AutoFile::tell()
 {
-    if (IsNull()) {
-        throw std::ios_base::failure("AutoFile::tell: file handle is nullptr");
-    }
-    int64_t r{std::ftell(m_file)};
-    if (r < 0) {
-        throw std::ios_base::failure("AutoFile::tell: ftell failed");
-    }
-    return r;
+    if (!m_position.has_value()) throw std::ios_base::failure("AutoFile::tell: position unknown");
+    return *m_position;
 }
 
 void AutoFile::read(Span<std::byte> dst)
@@ -60,6 +72,7 @@ void AutoFile::ignore(size_t nSize)
             throw std::ios_base::failure(feof() ? "AutoFile::ignore: end of file" : "AutoFile::ignore: fread failed");
         }
         nSize -= nNow;
+        if (m_position.has_value()) *m_position += nNow;
     }
 }
 
@@ -70,19 +83,19 @@ void AutoFile::write(Span<const std::byte> src)
         if (std::fwrite(src.data(), 1, src.size(), m_file) != src.size()) {
             throw std::ios_base::failure("AutoFile::write: write failed");
         }
+        if (m_position.has_value()) *m_position += src.size();
     } else {
-        auto current_pos{std::ftell(m_file)};
-        if (current_pos < 0) throw std::ios_base::failure("AutoFile::write: ftell failed");
+        if (!m_position.has_value()) throw std::ios_base::failure("AutoFile::write: position unknown");
         std::array<std::byte, 4096> buf;
         while (src.size() > 0) {
             auto buf_now{Span{buf}.first(std::min<size_t>(src.size(), buf.size()))};
             std::copy(src.begin(), src.begin() + buf_now.size(), buf_now.begin());
-            util::Xor(buf_now, m_xor, current_pos);
+            util::Xor(buf_now, m_xor, *m_position);
             if (std::fwrite(buf_now.data(), 1, buf_now.size(), m_file) != buf_now.size()) {
                 throw std::ios_base::failure{"XorFile::write: failed"};
             }
             src = src.subspan(buf_now.size());
-            current_pos += buf_now.size();
+            *m_position += buf_now.size();
         }
     }
 }

src/streams.h

@@ -390,9 +390,10 @@ class AutoFile
 protected:
     std::FILE* m_file;
     std::vector<std::byte> m_xor;
+    std::optional<int64_t> m_position;
 
 public:
-    explicit AutoFile(std::FILE* file, std::vector<std::byte> data_xor={}) : m_file{file}, m_xor{std::move(data_xor)} {}
+    explicit AutoFile(std::FILE* file, std::vector<std::byte> data_xor={});
 
     ~AutoFile() { fclose(); }
 

src/test/streams_tests.cpp

@@ -261,7 +261,7 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file)
     for (uint8_t j = 0; j < 40; ++j) {
         file << j;
     }
-    std::rewind(file.Get());
+    file.seek(0, SEEK_SET);
 
     // The buffer size (second arg) must be greater than the rewind
     // amount (third arg).
@@ -391,7 +391,7 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file_skip)
     for (uint8_t j = 0; j < 40; ++j) {
         file << j;
     }
-    std::rewind(file.Get());
+    file.seek(0, SEEK_SET);
 
     // The buffer is 25 bytes, allow rewinding 10 bytes.
     BufferedFile bf{file, 25, 10};
@@ -444,7 +444,7 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file_rand)
         for (uint8_t i = 0; i < fileSize; ++i) {
             file << i;
         }
-        std::rewind(file.Get());
+        file.seek(0, SEEK_SET);
 
         size_t bufSize = m_rng.randrange(300) + 1;
         size_t rewindSize = m_rng.randrange(bufSize);

src/util/asmap.cpp

@@ -203,10 +203,10 @@ std::vector<bool> DecodeAsmap(fs::path path)
         LogPrintf("Failed to open asmap file from disk\n");
         return bits;
     }
-    fseek(filestr, 0, SEEK_END);
-    int length = ftell(filestr);
+    file.seek(0, SEEK_END);
+    int length = file.tell();
     LogPrintf("Opened asmap file %s (%d bytes) from disk\n", fs::quoted(fs::PathToString(path)), length);
-    fseek(filestr, 0, SEEK_SET);
+    file.seek(0, SEEK_SET);
     uint8_t cur_byte;
     for (int i = 0; i < length; ++i) {
         file >> cur_byte;