hamsterdb 2.1.8                                     Mo 7. Jul 11:02:27 CEST 2014
(C) Christoph Rupp, chris@crupp.de; http://www.hamsterdb.com

This is the README file of hamsterdb.

Contents:

1. About

hamsterdb is a database engine written in C/C++. It is fast, production-proven
and easy to use.

This release has 4 big changes:

The remote protocol no longer uses Google's protocol buffer but a faster
homemade serialization protocol. Remote performance was improved.

A new set of APIs were added which demonstrate the analytical powers of
hamsterdb: "hola" (hamsterdb on-line analytics) can perform COUNT,
COUNT DISTINCT, AVERAGE, SUM and other calculations. These routines are directly
executed in the Btree context and very fast.

Under the hood, the two Btree layouts were rewritten. The result now allows a
much more efficient storage of fixed length keys with duplicates, or variable
length keys without duplicates. Prior to 2.1.8, a less efficient representation
was used in both cases. This change reduced the file size a lot.

A new Python API was added! A sample is included (see python/samples).

Other changes are:

A new API "ham_cursor_get_duplicate_position" was added to retrieve
the current position of a cursor in the duplicate list.

The APIs ham_db_get_error was deprecated, ham_get_license was removed.

A regression in 2.1.7 was fixed which created large files under certain
circumstances. A few other changes were made to decrease the file size.

On Windows, winsock.h is no longer included in order to avoid conflicts with
winsock2.h.

The database format is no longer endian agnostic but simply uses the host
endianness. Reason is that i do not have the hardware to test the agnostic
file format, and therefore the implementation was likely to have bugs anyway.

libuv was updated to version 0.11.22.

The file format of 2.1.8 is not compatible to 2.1.7.

2. Changes

New Features
* Adding the new "hola" API - hamsterdb analytical functions for COUNT,
	SUM, AVERAGE etc. See ham/hamsterdb_ola.h for the declarations
* Added new API ham_cursor_get_duplicate_position
* A new Python API was added

Bugfixes
* issue #33: upgraded to libuv 0.11.22
* Fixing a performance regression in 2.1.7 - large fixed-length keys
	created too many page splits, even if they were stored as extended
	keys

Other Changes
* The database format no longer tries to be endian agnostic; the database
	is now stored in host endian format. The endian agnostic code was broken
	anyway, and I had no hardware to test it.
* ham_db_get_error is now deprecated
* header files no longer include winsock.h to avoid conflicts with winsock2.h
	on Windows platforms
* Both btree layouts have been completely rewritten; PAX KeyLists
	can now be used in combination with duplicate RecordLists, and variable
	length KeyLists can now be used in combination with PAX RecordLists
* Avoiding Btree splits if keys are appended (HAM_HINT_APPEND)
* The internal communication with the remote server now uses a different
	protocol which is faster than google's protobuffer
* PAX layout now uses linear search for small ranges; this improves search
	performance by 5-10%
* Removed the ham_get_license API (and serial.h)

To see a list of all changes, look in the file ChangeLog.

3. Roadmap
- See https://github.com/cruppstahl/hamsterdb/wiki/Roadmap

4. Features

- PRO: SIMD instructions for lookups (numeric keys w/o duplicates)
- PRO: transparent AES encryption
- PRO: transparent compression for journal, keys and records using 
    zlib, snappy, lzf or lzo 
- Very fast sorted B+Tree with variable length keys
- Basic schema support for POD types (i.e. uint32, uint64, real32 etc)
- Can run as an in-memory database
- Supports multiple databases in one file
- Supports record number databases ("auto-increment")
- Supports duplicate keys
- Support for logging and recovery
- Support for unlimited number of parallel Transactions
- Support for partial reading/writing of records
- Support for network access (remote databases) via TCP/Protocol Buffers
- Very fast database cursors
- Configurable page size, cache size, key size etc
- Runs on Linux, Unices, Microsoft Windows and other architectures
- Uses memory mapped I/O for fast disk access (but falls back to read/write if
  mmap is not available)
- Uses 64bit file pointers and supports huge files (>2 GB)
- Easy to use and well-documented
- Open source and released under APL 2.0 license
- Wrappers for C++, Java, .NET, Erlang, Python, Ada and others

5. Known Issues/Bugs

None.

6. Compiling

6.1 Linux, MacOS and other Unix systems

To compile hamsterdb, run ./configure, make, make install.

Run `./configure --help' for more options (i.e. static/dynamic library,
build with debugging symbols etc).

6.2 Microsoft Visual Studio 8

A Solution file is provided for Microsoft Visual C++ 2005:
'win32/hamsterdb.sln'. It compiles a static and a dynamic library.
Please read the README.TXT file in the win32 directory.
All libraries can be downloaded precompiled from the hamsterdb webpage.

To download Microsoft Visual C++ 2005 Express Edition for free, go to
http://msdn.microsoft.com/vstudio/express/visualc/default.aspx.

6.3 Dependencies

On Ubuntu, the following packages are required:
  - libdb-dev (optional)
  - protobuf-compiler
  - libprotobuf-dev
  - libgoogle-perftools-dev
  - libboost-system-dev
  - libboost-thread-dev
  - libboost-dev
  - (libuv needs to be installed from sources - see
      https://github.com/joyent/libuv)

For Windows, precompiled dependencies are available here:
https://github.com/cruppstahl/hamsterdb-alien

7. Testing and Example Code

Make automatically compiles several example programs in the directory
'samples'. To see hamsterdb in action, just run 'samples/db1'
or any other sample. (or 'win32/out/samples/db1/db1.exe' on Windows platforms).

8. API Documentation

The header files in 'include/ham' have extensive comments. Also, a doxygen
script is available; run 'make doc' to start doxygen. The generated
documentation is also available on the hamsterdb web page.

9. Other Ways to Compile hamsterdb

If you want to compile hamsterdb without using the provided ./configure
environment, you have to set some preprocessor macros:

DEBUG                   enable debugging output and diagnostic checks (slow!)
HAM_32BIT               compile for 32bit (alias: WIN32)
HAM_64BIT               compile for 64bit (alias: WIN64, also needs WIN32)

Also, if you compile for windows, you have to compile the file
'src/os_win32.cc' and ignore the file 'src/os_posix.cc'. Vice versa on
non-Windows platforms.

10. Porting hamsterdb

Porting hamsterdb shouldn't be too difficult. All operating
system dependend functions are declared in 'src/os.h' and defined
in 'src/os_win32.cc' or 'src/os_posix.cc'.
Other compiler- and OS-specific macros are in 'include/ham/types.h'.
Most likely, these are the only files which have to be touched. Also see item
9) for important macros.

11. Migrating files from older versions

Usually, hamsterdb releases are backwards compatible. There are some exceptions,
though. In this case tools are provided to migrate the database. First, export
your existing database with ham_export linked against the old version.
(ham_export links statically and will NOT be confused if your system has a
newer version of hamsterdb installed). Then use the newest version of
ham_import to import the data into a new database. You can find ham_export
and ham_import in the "tools" subdirectory.

    Example (ham_export of 2.1.2 was renamed to ham_export-2.1.2 to document
        that it's an older version):

    ham_export-2.1.2 input.db | ham_import --stdin output.db

12. Licensing

hamsterdb is released under the APL 2.0 license, which allows
unrestricted use for commercial and non-commercial applications. See the
file COPYING for more information.

A commercial, closed source version hamsterdb pro with additional functionality
is available on request. See http://hamsterdb.com for more information.

13. Contact

Author of hamsterdb is
    Christoph Rupp
    Paul-Preuss-Str. 63
    80995 Muenchen/Germany
    email: chris@crupp.de
    web: http://www.hamsterdb.com

14. Other Copyrights

The Google Protocol Buffers ("protobuf") library is Copyright 2008, Google Inc.
It has the following license:

    Copyright 2008, Google Inc.
    All rights reserved.

    Redistribution and use in source and binary forms, with or without
    modification, are permitted provided that the following conditions are
    met:

    * Redistributions of source code must retain the above copyright
      notice, this list of conditions and the following disclaimer.
    * Redistributions in binary form must reproduce the above
      copyright notice, this list of conditions and the following disclaimer
      in the documentation and/or other materials provided with the
      distribution.
    * Neither the name of Google Inc. nor the names of its
      contributors may be used to endorse or promote products derived from
      this software without specific prior written permission.

    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
    "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
    A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
    OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
    SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
    LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
    DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
    THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
    (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
    OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

    Code generated by the Protocol Buffer compiler is owned by the owner
    of the input file used when generating it.  This code is not
    standalone and requires a support library to be linked with it.  This
    support library is itself covered by the above license.

The libuv library is part of the Node project: http://nodejs.org/
libuv may be distributed alone under Node's license:

    Copyright Joyent, Inc. and other Node contributors. All rights reserved.
    Permission is hereby granted, free of charge, to any person obtaining a copy
    of this software and associated documentation files (the "Software"), to
    deal in the Software without restriction, including without limitation the
    rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
    sell copies of the Software, and to permit persons to whom the Software is
    furnished to do so, subject to the following conditions:

    The above copyright notice and this permission notice shall be included in
    all copies or substantial portions of the Software.

    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
    FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
    IN THE SOFTWARE.