MySQL++

3.3.0, 2021.04.28

• MySQL++ now detects the availability of C++11 capable compilers on Autoconf based systems and adjusts the compiler calling command to require that if such options are required.

  The general policy for MySQL++ is to support the oldest version of C++ that our oldest supported system provides. Thus the update to C++11: the last compilers still supporting only C++03 or C++98 dropped out of support since the last time we checked. It may be that this release still supports these older versions of C++, but we will not longer go very far out of our way to ensure it. Patches to restore backwards compatibility will be accepted if they don’t break forward compatibility.

  Beware: This change means builds are only ABI-compatible within the 3.x series when built with the same C++ version! You may need to override CXX, CXXFLAGS and similar to get an ABI-compatible build, since older releases would use the generic compiler options, which might default to older versions of C++.

  You can adjust the build system to try for C++14 or C++17 by changing the AX_CXX_COMPILE_STDCXX call in configure.ac. Although this release contains features that depend on C++17, they’re optional, so unless you do this or adjust the build flags manually to ask your compiler to use a newer C++ version, these features will not be built into MySQL++.

• Several conditional C++11 thru C++17 improvements by GitHub user “BratSinot:”

  • Added move ctors and no-copy ctors to ScopedConnection.
  • The copy ctor and assignment operators for class ScopedLock change so the compiler can elide them where it finds that safe.
  • Added no-copy and no-assign ctors to class NoExceptions.
  • Added a no-assign operator to class mysql_ti_sql_type_info.
  • Using (std::string::operator==) in the inner loop of FieldNames::operator [] to speed up comparisons.

• Changed the “file slurp” idiom used by the load_jpeg example to a form that works with C++11, squishing an “address to rvalue” complaint with some compilers.

• Throwing exceptions (if enabled) from a few more places in the code now when the underlying C API library reports an error.

• Fixed a SQL type mapping problem with TIMESTAMP NULL. Patch by Fossil repository user nofree.

• Replaced a use of size() in bool context on an STL data structure where that isn’t a constant-time operation with faster !empty().

• Updated the stock paths in the MySQL C API library autodiscovery Autoconf macro so it'll work on MariaDB based systems where the headers and/or library are installed in a .../mariadb directory instead of .../mysql for backwards compatibility.

• The RPM spec file now declares a build dependency on mariadb-devel rather than mysql-devel.

• Reordered the -L. flag in the build lines to ensure the examples and test programs link against the in-tree version of MySQL++ if you have it installed at the system level as well.

• Squished pedantic warnings from newer compilers.

• Assorted documentation improvements.

• Assorted build system improvements.

3.2.5, 2019.07.21

• Supports ulonglong in MySQL++-to-SQL data type conversions so you can pass the return from Result.insert_id() and such into another SQL expression. This previously required an explicit cast to one of the sql_* types when using a C++ compiler that can’t figure out that sql_bigint is equivalent on its own.

• Updates to support newer versions of MySQL and MariaDB.

• Assorted documentation improvements.

• Assorted build system improvements.

3.2.4, 2018.07.26

• Added the MYSQLPP_UTF8_CS and MYSQLPP_UTF8_COL macros to allow a given bit of code use either the legacy "utf8" character set or the new full-Unicode "utf8mb4" version depending on whether MySQL++ is built against MySQL 5.5 or newer, the version where utf8mb4 was added. Using this in the resetdb example to demo the feature.

• Updated the Unicode chapter of the user manual to cover this and other changes in the Unicode world over the past several years.

• Recent versions of MySQL and MariaDB have removed some obsolete features, causing MySQL++ build errors in places where the library wraps one of these features:

  • MySQL 8.0 removed the embedded database feature, which caused build errors in four Option::set() overrides. For MySQL 8.0 and newer, these methods are now no-ops.
  • Removal of the CLIENT_LONG_PASSWORD define in MariaDB 10.2 caused a build error which we now fix with a conditional test for this before trying to use it. Thanks for this fix go to Augusto Caringi of Red Hat.

• Updated the library to work with newer toolchains, particularly Clang on current macOS and the C++17 aware versions of GCC:

  • Methods with throw specifications still have them for ABI compatibility on systems with older compilers, but if we detect that the compiler is expecting C++14 or newer, we elide them.
  • If the compiler is detected as expecting C++11 or newer, we now use std::unique_ptr everywere we used to use auto_ptr.
  • Removed all register variable qualifiers. It's officially obsolete in C++17 and it's been anywhere from unnecessary to unhelpful in C++ compilers for quite a long time now.

• It is now possible to build a release tarball (make dist) on systems like Debian where /bin/sh is not Bash.

• All prior MySQL++ 3.2.x release tarballs contain a symlink from ltmain.sh to the copy provided by Libtool on the system the tarball was made on. If your system doesn't have Libtool installed or it's installed somewhere other than where Red Hat Enterprise Linux puts it, that symlink will be broken, preventing those tarballs from building until you fix the symlink somehow. We're now dereferencing all symlinks when building the tarball to avoid this sort of problem.

There are also several changes to MySQL++ due purely to converting the MySQL++ source code repository from Subversion to Fossil:

• Updated the hackers' guide to reflect all of the differences this repository conversion entails.

• Converted the old HACKERS.txt and this change log file to Markdown format. Fossil will display plain text just fine, but it has a web renderer for Markdown, which gives nicer output.

3.2.3, 2016.12.29

• Now using snprintf() instead of _snprintf() on Visual C++ 2015.

• Cygwin changed its naming for the MySQL C API library since the last time we successfully tried building MySQL++ there. Reworked the autoconf detection macro to make it less dependent on such individual platform differences.

  The macro is in fact now entirely side-effect-free, just setting MYSQL_C_* variables that the top-level Bakefile uses when and as it needs them, rather than expecting that major variables like CPPLFAGS and LIBS were overridden by the autoconf macro. This makes the autoconf case more like that of the non-autoconf Bakefile outputs.

• Squished a Clang warning

3.2.2, 2015.03.18

• The DBDriver::operator= would leak a live connection if you assigned a disconnected DBDriver to it. Patch by Quentin Armitage.

• Plugged a potential DBDriver instance leak in Connection ctor. Patch by Yury Alekseev.

• Explicitly #including algorithm.h in examples/multiquery.cpp for std::max().

• Added missing "break;" statements in Transaction ctor taking an isolation level parameter, which caused only the last one to do the right thing. Problem noticed by Tony Whyman, fix by me.

• The bootstrap script now checks for the existence of needed build tools before trying to use them. This lets us emit a clearer error message than the shell will. Original patch by Adrian Cornish.

• Fixed a build problem related to std::min/max with Visual C++ 2013 due to a change in the definitions provided by its header files.

• Tracked several changes made to Connector/C made since 3.2.1.

• Assorted documentation updates.

• Generating HTML with UTF-8 encoding, not 8859-1

3.2.1, 2013.12.10

• Connection copy ctor wasn't copying the "throw exceptions" flag. The copy always had exceptions enabled.

• The resetdb example would print a bogus message about logging in as '1'@'1' instead of showing the actual values passed on the command line. This did not affect how it actually logged in.

• Previous release tarball was missing a few text files present in the source repository.

• Updated build paths for VC++ and MinGW to track a file layout change in Connector/C 6.1.2.

3.2.0, 2013.06.20

• References to MySQL 5.1 have been replaced with 5.6, that being the current GA release.

• All Windows builds now assume you're using Connector/C 6.1 instead of having installed the MySQL server on your machine just to get development files. If you are running a MySQL server on your development machine and don't want to install Connector/C just to make MySQL++ happy, see the README-*.txt file for your platform to get instructions on fixing the paths so the build system finds the development files under the MySQL server directory.

• The generated VC++ 2008 project files now assume you want 64-bit executables. (It is 2013 now, after all.)

  Since the VC++ 2005 project files continue to be configured for 32-bit builds, the easiest way to get a 32-bit project in VC++ 2008+ is to import the VC++ 2005 project files, rather than adjust the build target settings of all 44 vc2008/*.vcproj files within the IDE.

  See README-Visual-C++.txt for more on this change.

• Added Query::replacefrom(): like insertfrom() but it uses REPLACE statements instead of INSERT. Patch by Adrian Cornish.

• Added support for SQL "unsigned float" columns. Patch by Leonardo Lang.

• Added "abicheck" top-level build target, which runs the ISPRAS ABI checker tool (http://goo.gl/e19lD) against a predecided "ABI-stable" version of MySQL++, comparing it against the current version. This should help avoid future unplanned ABI breakages, as happened between 3.0.9 and 3.1.0.

  We have chosen not to revert those changes that broke the ABI, but instead have chosen to accept the 3.1.0 ABI as the new stable ABI for the rest of the 3.x series. Running the ABI checker is now part of the release process, so we should not change the ABI again until 4.0!

• The Query::insert() and replace() overloads that take a pair of iterators now work with containers that only provide forward iterators. Idea by Adrian Cornish.

• Using libtool again on Autotools systems. It gives us relocatable libraries, which is needed in some linking situations. Patch by Adrian Cornish.

• VC++ release builds now have optimization enabled. This might also affect other toolchains that create separate Debug and Release builds, if Bakefile obeys the directive for those systems. (Such as, perhaps, Xcode.) Initial patch by Matthew Bradbury.

• Fixed a crash in CommandLine mechanism when the usage message is called up for programs that have no "extra" parameter and corresponding usage appendage. Initial patch by "Crazy Pete".

• Query::storein() now tolerates empty result sets, due either to DBMS failure or passing a query that never returns results, such as INSERT.

• Exposed DBDriver::result_empty() as Query::result_empty(). We did this primarily because it was needed as a helper for the previous fix, but it could also be useful more broadly.

• Added -Wextra to the pedantic build flags.

• Fixed the "escape" manipulator so it will work with Null<> wrapped types. Patch by "Kemet".

• ssqls2parse now depends on libmysqlpp. (It wasn't getting relinked when you changed the library.)

• The configure script's test for the existence of mysql_ssl_set() got broken by a change in the mysql_loc.m4 test, causing it to always return "no" even if the platform's MySQL C API library did have the function. Therefore, the SslOption could never be set. Fixes this bug: stackoverflow.com/questions/7794408

• Fixed a missing a "buried headers" ifdef wrapper for type_info.cpp. Patch provided by Xavier Richez.

• We now export the DBDriver class from the MySQL++ DLL when building on Windows. Client programs can now access it directly, instead of being able to access only the fields and members exposed through Connection and Query. Fix suggested by Xavier Richez.

• MinGW builds no longer require that you manually create an import library for the MySQL C API DLL from a .def file we provide, which would inevitably get outdated. We can now link directly to the DLL, and rely on the MinGW linker to figure out the imports.

• Split the -l flags out of LDFLAGS in the userman's example Makefiles. They're now in LDLIBS, which causes them to be put after -o $(EXE) by the default GNU make rules, where they should be.

• Assorted documentation and build system improvements.

• Assorted fixes for newer GCCs.

3.1.0, 2010.06.03

• Default MySQL version now 5.1, having entered GA since 3.0.9.

• Lots of platform updates tracked, lots of warnings from newer compilers squished. Most by me, some by Adrian Cornish.

• Added Query::insertfrom(), plus associated classes SQLStream, NoTransactions, and the InsertPolicy hierarchy. Also adds examples/ssqls6.cpp, which shows how to use this new feature. Thanks for this feature go to Rick Gutleber, except for RowCountInsertPolicy, by me.

• Added comparison operators to tiny_int<T>. Thanks for this patch go to Edward Diener.

• Added SQLTypeAdapter::is_null(). This lets you retrieve a nullable column from the DB, then re-insert a value from that column back into the DB via a template query without explicit checks for null-ness; it will do the right thing now.

• Added -f flags to lib/*.pl header file generating scripts to allow overriding the default limit of 25 fields for tqueries and SSQLSes without having to edit the file, as in previous versions of MySQL++. Also added --with-field-limit option to the configure script so you can give both -f options at once during the build process. This latter is especially useful for automated binary package builders.

• The file lib/querydef.h (generated by lib/querydef.pl) now defines the MYSQLPP_TQUERY_MAX_PARAMETERS macro, giving the maximum number of parameters the functions in the generated file allow for template queries. Similarly, lib/ssqls.h (generated by lib/ssqls.pl) defines MYSQLPP_SSQLS_MAX_MEMBERS.

• Added ConnectionPool::safe_grab(), which pings the grabbed connection before returning it. If the ping fails, we remove the dead connection from the pool and try again. This doesn't eliminate the possibility of returning a dead connection; there is no protection against race conditions. It is merely a best-effort attempt.

• Added ConnectionPool::remove(Connection*), which removes a connection from the pool.

• Added ConnectionPool::exchange(), which wraps remove() and grab(). It works like a customer support counter in a store: "I'm sorry, but this Connection is defective. Can I have another, please?"

• Added a feature to Transaction, where it can send a SET TRANSACTION ISOLATION LEVEL query before it starts the transaction proper, affecting the isolation level and the scope of that change.

• Added ScopedConnection class. Initial version by Joel Fielder, lots of changes by me. Using it instead of explicit safe_grab() and release() calls on the pool in examples/cpool.cpp.

• Added FieldNames::operator[](size_type) overloads. This is necessary to correctly overload the base class versions inherited from std::vector.

• FieldNames wasn't doing field-by-name matches case insentitively, most likely since 3.0.0 when we changed it to derive from std::vector. This caused SSQLS (at least) to be case sensitive. Fixed this and renamed the stock.sdate field in the SSQLS examples to stock.sDate to test this.

• Added Query::replace(iter, iter), on the model of existing insert(iter, iter). Thanks for this patch go to David Walthour.

• Added Query::operator!(), overriding basic_ios version so that if (!query) does the right thing. Fixes problem report in thread http://lists.mysql.com/plusplus/8594

• Query copy ctor and assignment operator now perform deeper copies, so you can pass Query objects around in certain ways that used to cause crashes. Thanks for this patch go to Steven Van Ingelgem.

• Defined *_null variants for every sql_* typedef, wrapping all those earlier types in Null<>. All example programs now use these types instead of explicit Null<>-wrapped types.

• Created a C++ command line parsing mechanism for the examples today, and for programs coming in the future. It uses the platform version of getopt() if available, substituting a public-domain implementation that gets built into MySQL++ itself otherwise. It's documented in the refman -- see the CommandLineBase class -- but it's in the newly created mysqlpp::internal namespace, which means we don't promise any sort of ABI or API stability for it. You're welcome to use it in your own programs, but expect it to change without warning.

• Added a few missing MYSQLPP_EXPORT qualifiers to operator<<() overloads for stream manipulators to allow their use when MySQL++ is built as a DLL.

• Added backticks around table and column identifiers in all code paths exercised by dtest. There are some others remaining, particularly things like Row::field_list(). Thanks for this patch go to Adrian Cornish.

• Added mysqlpp::NOW() which returns a value that, when inserted into a SQL query, results in a call to SQL's NOW() function. DateTime::now() -- added in 3.0.0 -- does the same thing, but this is shorter and matches the SQL syntax.

• Calling DBDriver::disconnect() twice no longer calls mysql_close() twice.

• Worked around a weakness on Windows that caused problems in code with a connect(), disconnect(), connect() call pattern, when also using a named pipe to connect to the DB. (There never has been a problem with this on other platforms or with other IPC mechanisms.)

• Initializing a SQLTypeAdapter with a floating-point value equal to infinity or NaN results in SQL null if the type is also nullable, or 0 otherwise. Previously, we got an implementation-defined string, which could give invalid SQL, or could give 0, a different value from NaN, infinity, or SQL null. This does collapse the notions of NaN and infinity, but the MySQL reference manual doesn't say how you're supposed to preserve these differences.

• Removed the dependency of stadapter.h on sql_types.h just to get sql_tinyint*. We now #include "tinyint.h" instead, and use equivalent types to what sql_types.h defines. This is necessary because sql_types.h should be one of the last things included from mysql++.h, so that it can define sql_* equivalents for everything else in MySQL++, which means all those other things need to be included by that point.

• The "find MySQL++" autoconf macro (config/mysql++.m4) now searches for the library in lib64 subdirectories everywhere it also looks in lib. We did this for the C API finder macro way back in MySQL++ 2.2.0, and only neglected to do it for this macro because it's not used by MySQL++ itself, just provided as a courtesy to MySQL++ users that also use autoconf.

• Also improved the "find C API" macro (config/mysql_loc.m4) so it, too, is more useful in third-party projects.

• Moved private FieldNames::str_to_lwr() method out into its own module, in the new mysqlpp::internal namespace, so other modules can use it. Also added an overload for this new global method taking std::string instead of char*.

• userman build process can now find FOP when installed from the standard Ubuntu/Debian package repository.

• No longer generating refman.pdf. It's difficult to generate properly on some platforms, and isn't all that useful anyway compared to the HTML version, which builds properly everywhere.

• Dropped custom.h backwards-compatibility header. (Deprecated over a year ago in MySQL++ 3.0.0.)

• Fixed a bad pointer access crash in Connection::set_option() triggered when the option set call fails. Thanks for this patch go to Kevin Regan.

• ReconnectOption behavior now knows about a fix in MySQL 5.1.6 and newer that lets you set it either before the connection is up, or after. If you try to set this option when MySQL++ is built against earlier versions of the MySQL C API, it will refuse to set the option because it knows the option will be reset when the connection comes up.

• No longer trying to make the C API library read from a "my" options file before connect. It does that by default, and the file is supposed to be .my.cnf anyway.

• Reworked the load_jpeg example code for clarity, extracting load_jpeg_file() routine as an implementation detail and making that use the C++ stringstream "slurp into string" idiom instead of allocating a C char buffer and copying that into a std::string.

• Restored support for repeating query parameters. Apparently you could say things like "stuff %0 morestuff %0" in v2.1, but this was broken by initial 1-parameter fixes in v2.2. This patch reportedly lets us have both things at once, and improves handling of 1-parameter template queries in general. Thanks for this patch go to Martin Gallwey.

• Added examples/tquery4.cpp based on test code and feedback from Martin.

• Put Comparable class in namespace mysqlpp, where it should always have been. Aside from being just plain wrong, it caused a conflict with Boost if you used its Comparable class and hoist them both into the global namespace. Thanks for this patch go to Michael Radzewitz.

• Fixed all the known problems with the Xcode project files.

• Skeletal support for SSQLS v2. Doesn't do anything useful yet, it just got caught up in the need to release 3.1 as-is.

• Various enhancements to the test suite.

• Many, many documentation enhancements.

3.0.9, 2009.02.04

• It is now possible to use Null objects in const context.

• Query was emitting thousands separators when used in some locales. Thanks for the fix go to Tomaž Šolc.

• Restored support for using Query::storein() with template queries. Was mistakenly removed in the 3.0.0 development process.

• GCC 4.4 build fixes. Thanks for this go to Remi Collet.

3.0.8, 2008.11.27 (r2432) The "Thanksgiving" release

• Fixed a VC++ build error introduced in 3.0.7 in the Sun CC fix.

• Fixed the Makefile we generate from Makefile.simple for the devel RPM. The recent Bakefile upgrade changed the output so our munger stopped munging.

• Changed all instances of "CREDITS" in header comments to "CREDITS.txt".

• Brought ebuild file up to date with what Gentoo has been using for 2.3.2.

3.0.7, 2008.11.21

• Fixed bug where initting a Query with a query string and then appending more to it with operator <<() would start overwriting the beginning of the buffer. Thanks for the fix go to Chris Frey.

• Added String::empty(), to parallel std::string better.

• Added SQLTypeAdapter ctor for Null, to allow SSQLS with NULL BLOB columns. Thanks for this patch go to Russell Robinson.

• Passing on "throw exceptions" flag in Query::storein() to the temporary Row objects it creates. (Should have done this all along.)

• All MySQL++ classes with operator and at() methods now throw the new BadIndex exception on out of range indices now, unconditionally. Before, it was a mishmash of MySQL++-specific mechanisms, std::range_error, and unchecked access.

• Fixed some bugs in load_jpeg and cgi_jpeg examples that prevented them from working correctly on Windows.

• Squished some warnings and incompatibilities that caused difficulty with Solaris CC.

• Several minor fixes to allow building on Cygwin again.

• Dedicated VC++ 2008 support. (Not just a clone of VC++ 2005 project files in a different directory.)

• Lots of documentation improvements.

3.0.6, 2008.08.16

• Several fixes to allow thread-aware build on Solaris

• More improvements to search for Fink version of C API library.

• Various improvements to Windows installer (install.hta)

3.0.5, 2008.08.06

• Replaced install.bat with new install.hta, which has a GUI and a lot of embeded logic for doing The Right Thing, which we couldn't do in a dumb batch file.

• Several fixes to allow it to build on Solaris 10.

• Fixed a bug in comparison of Null to T: wouldn't always give the right result for one particular combination of argument values and order of parameters to operator <(). It wouldn't fail all the time, either; it depended on the way the system's memory allocator dealt with newly allocated RAM. The bug was discovered during the Solaris 10 porting effort, but it is not a Solaris-specific bug.

• Split Linux-specific material out of README-Unix.txt into README-Linux.txt, and created README-Solaris.txt.

• Shipping a vc2008 directory. Populated by bootstrap script with copies of vc2005 files when those are newer, with the idea being to update them by hand by running them through VC++2008 before release.

• Several fixes to VS-only examples. They hadn't been updated to track several of the changes in MySQL++ v3, so they wouldn't build OOTB at all, crashed when you did get them building, and emitted a lot of warnings during the build. Cleaned it all up.

• Autoconf now explicitly checks whether we need to link to zlib to link to MySQL C API library. It used to be required, but lately, MySQL has been shipping the library with zlib stuff embedded, causing multiply defined symbol errors on some systems.

• configure script now looks in more locations for the MySQL C API library, adding default locations for Fink and Solaris.

3.0.4, 2008.07.02

• Fixed a few bugs in SslOption that prevented it from actually working. If you've been having SSL problems since upgrading to MySQL++ v3, this is why! Thanks for these patches go to Petteri Kiiskinen.

• Changed type of String's internal "size" typedefs so it'll build without errors on 64-bit systems where these ints are different sizes.

• Many user manual improvements.

3.0.3, 2008.05.11

• Fixed query well-formedness handling in Query::store() and use(). It was quietly eating these errors, confusing them with the no-results case.

• Fixed examples/cpool.cpp to build without thread support.

• No longer hiding all stdint.h typedefs inside namespace mysqlpp.

• Fixed mysqlpp::String comparisons of empty strings against nonempty strings; if the empty string was on the left hand side, it would succeed because it was only comparing against characters in the empty side, which matches everything. (Fixes bug 11588.)

3.0.2, 2008.04.13

• Increased float to string conversion precision from 7 to 9 and double from 16 to 17 digits. Previous values weren't enough near the maximum values that each can hold.

• Replaced Query's safe bool mechanism with an override of basic_ios::operator void*() to avoid a conflict between the two mechanisms. As the base class version isn't virtual, this is arguably evil, but it should be harmless in typical use. Besides, testing Query in bool context usually isn't the right thing anyway: test the result set instead.

• Made ConnectionPool::grab() virtual.

• Overriding ConnectionPool::grab() and release() in examples/cpool.cpp to show how to do connection-in-use count limiting. Also, added a few more output indicator states to allow one to better understand program flow.

3.0.1, 2008.03.23

• String objects can now be compared to mysqlpp::null directly.

• Added a template operator== and operator!= to String, syntactic sugar for the existing String::compare() methods.

• String::compare() now returns 0 ("equal") when one of the strings is an uninitialized String() (no refcounted buffer) and the other is empty. It used to consider any initialized string greater than an uninitted one. An uninitialized String appears empty, though, so this was incorrect.

• Made Connection::thread_aware(), thread_start() and thread_end() static methods, so they can be called before you create your first connection. Ditto for DBDriver versions of these methods.

• Calling Connection::thread_start() and thread_end() in examples/cpool.cpp, as appropriate. Above changes were necessary to make this work sensibly.

• Made ConnectionPool::release() virtual, so your pool subclass can override it.

• Added ConnectionPool::size(), so a subclass can know the current number of extant connections.

• No longer single-quoting NOW() call generated for default init of DateTime type when building queries from SSQLS objects in Query::insert(), replace() and update(). The template query and stream interfaces of Query treated NOW() correctly already.

• Fixed a bug that left SSQLS::table_override_ uninitted if you used certain of the generated ctors or set() member functions instead of others used by the examples. This could cause a crash any time you caused SSQLS.table() to be called, such as when passing the SSQLS to Query::insert().

• Minor memset bug fixed in test/uds.cpp. Patch by Dave Jones.

3.0.0, 2008.02.29 (r2236) The "Giant Leap Forward" release

WARNING: THIS IS NOT A DROP-IN REPLACEMENT FOR MySQL++ v2.x!

You will have to recompile your program against this version of the library, and you will almost certainly have to make code changes as well. Please see thee API and ABI breakages sections in the user manual for information on migrating your code to this new version.

• Added ConnectionPool class, primarily to let multithreaded programs share a set of Connection objects safely in situations where it isn't acceptable to have a Connection per thread.

• Created examples/cpool.cpp to demonstrate this new class.

• Added RefCountedPointer template, which provides automatic memory management and data sharing. It's not intended for use outside of MySQL++ itself, but it's the mechanism behind everything below where reference counting is mentioned. I created the initial version of it, but Jonathan Wakely almost completely rebuilt it, and Joseph Artsimovich provided helpful commentary and advice as well.

• Many improvements to Specialized SQL Structures (SSQLS):

  • Renamed custom* to ssqls*. There's still a custom.h which #includes ssqls.h for you, but it's only intended to ease the transition to the new name. It will go away in a future release, probably as soon as v3.1.
  • SSQLSes are finally compatible with Null<>-wrapped types. This feature is based loosely on the "Waba" patch posted to the mailing list back in the v2.0 days, but extended to allow Null types for key fields. (The Waba patch only allowed these types in non-key fields.)
  • It's no longer necessary to define a different SSQLS for each different field set you use in queries. That is to say, you can define an SSQLS for an entire table and store just a subset of the table in it now, with the other fields keeping default values. Removed examples/custom6.cpp, as custom1.cpp can now demonstrate the same thing, implicitly.
  • An SSQLS's field order no longer has to match the order of fields in the result set it is populated from.
  • As a result of previous, removed sql_create_c_order_* macros; they have no purpose now.
  • Removed order parameters from sql_create_complete_*, which now gives it the same functionality as sql_create_c_names_* so removed the latter, too.
  • Removed "basic" variants of SSQLS creation macros. They've been unofficially deprecated by dint of being all but undocumented and unexemplified for a very long time now.
  • It's now possible to use mysqlpp::String, Date, DateTime, and Time types in the key field positions in an SSQLS as they now support the necessary comparison interfaces.
  • If you use a floating-point data type in one of the key field positions, it no longer uses exact comparison logic. Instead, it now does [in]equality comparisons by testing whether the difference between two floating-point values is less than a configurable threshold defaulting to 0.00001.
  • You can now use 'bool' type in an SSQLS.
  • Renamed _table static member variable in each SSQLS to table_ and made it private. There are now public setter and getter methods, table().
  • Added per-instance table name overriding via instance_table() setter. table() getter returns static version if this is not set, so it's still a global setting by default.

• You can now use mysqlpp::null as a template query parameter to get a SQL null.

• Replaced template ColData_Tmpl<T>:

  • Used to have typedef ColData_Tmpl<std::string> MutableColData. It was used only once within MySQL++ itself, and was never documented as a class for end users. This one use within the library was a crock, so we just replaced this use with std::string and removed the typedef.
  • This left just one use of ColData_Tmpl<T>, instantiating it with the MySQL++ utility class const_string, basically a clone of std::string with all the mutating features removed. Folded the functionality of const_string into the template, renamed the result to String, and deleted the const_string class. It'd be a complete std::string replacement -- with SQL-related enhancements -- if it were modifiable, but MySQL++ doesn't need it to be modifiable. Yet, it's still the closest thing MySQL++ has to its own string type; thus the name.
  • Replaced its internal buffer management with a much more clever reference counted scheme. This shows its greatest advantage in the return from Row::operator, which for technical reasons must return by value, not by reference as is more common. This lets you pass around Strings by value while having the efficiency of reference semantics. This can be important with large return values, like BLOBs.
  • Converting String to numeric types (ints, floats...) uses a new, cleaner system by Jonathan Wakely. Unless you were abusing weaknesses in the old system, you won't see a difference. It's just more robust and flexible.

• Redesigned SQLString:

  • It used to derive from std::string, and while MySQL++'s internals did use it in place of std::string, these places didn't take advantage of the additional features offered by SQLString. So, replaced all those uses with std::string.
  • All the remaining uses are MySQL++ public interfaces that need to be able to accept any of many different data types, and we want that data to be automatically converted to a SQL-compatible string form. Because it no longer has the parentage to be a general-purpose string type and MySQL++ has a new contender for that role (String), renamed SQLString to SQLTypeAdapter to reflect its new, limited purpose. ("STA" for short.)
  • Since we don't have the std::string base class to manage the string buffer any more, using the same reference counted buffer mechanism as String. In addition to saving code by not creating yet another buffer management mechanism, it means objects of the two classes can share a buffer when you assign one to the other or pass one to the other's copy ctor.
  • Added many more conversion ctors.
  • STA interfaces using the 'char' data type now treat them as single-character strings instead of one-byte integers, as does the Standard C++ Library.
  • Added mysqlpp::tiny_int interfaces to STA to replace the former char interfaces for those needing one-byte integers.

• As a result of the ColData -> String redesign, removed Row::raw_*(). Before String copies were efficient, this was helpful in accessing BLOB data efficiently. It was also required back when ColData didn't deal correctly with embedded null characters, but that reason is gone now, too.

• Row::operator[](const char*) no longer unconditionally throws the BadFieldName exception when you ask for a field that doesn't exist. It will still throw it if exceptions are enabled, but if not, it'll just return an empty String. This was necessary to make the SSQLS subset and field order independence features work.

• Similarly, Result::field_num() returns -1 when exceptions are disabled and you ask for a field that doesn't exist.

• You can now use the OptionalExceptions mechanism to disable exceptions on const MySQL++ objects.

• Redesigned query result classes:

  • Instead of Result deriving from ResUse, the two derive from a common base class -- ResultBase -- containing the bits that are truly the same between them. Before, Result inherited several methods that didn't really make sense for "store" query result sets.
  • Renamed Result to StoreQueryResult and ResUse to UseQueryResult so it's clearer what each is for.
  • Renamed ResNSel to SimpleResult.
  • Made all SimpleResult data members private and hid them behind const accessor functions of the same name.
  • The result set classes all used to be friends of Connection for various lame reasons. Since they are created by Query, and Query has a good reason for a strong relationship with Connection, moved Connection access out of each result set class into the code in Query that creates that type of result set object.
  • StoreQueryResult now derives from vector in addition to ResultBase; it used to merely emulate a vector of Rows, poorly. It can now dispose of the MYSQL_RESULT at the end of object construction, because it creates all the Row objects up front instead of on the fly. And as a result of that, operator[] returns by reference instead of by value, operator -> works correctly on iterators, all STL algorithms work, etc., etc.
  • IMPORTANT COMPATIBILITY BREAK: because we used fetch_row() stuff in Result previously, it was okay to index past the end of the result set: you'd just get a falsy Row when you did this, just as happens when doing the same thing in a "use" query. The simple1 and simple2 examples did this, so it's likely that code exists that takes advantage of this misfeature. New versions of these examples show how to index through a StoreQueryResult without running past its end.
  • ResUse used to delay creation of its FieldNames and FieldTypes objects until the point of need. This had several implications for thread and exception safety that we fix by just creating them in the ctor. If your code is multi-threaded and was avoiding certain usage patterns due to crashes, it's worth trying your preferred way again.
  • Result sets create a few data structures to hold information common to all rows in that set. The row objects need access to these shared data structures, so on creation each gets a pointer back to the result set object that creates it. This was efficient, but required that a result set object outlive any row objects it creates. Now these shared data structures are reference-counted, decoupling the lifetime of the child row objects from their result set parent.
  • Copy operations for result sets used to actually be "moves" before, for efficiency. (MySQL++ itelf only copied result sets in returning them by value from the query execution methods of Query, so this was acceptable if you didn't do anything uncommon with these objects.) Reference counted data structures allow us to have copy semantics now without sacrificing efficiency.
  • You can now use Query::storein() with an STL container of Row objects now, instead of having to use SSQLSes. The lifetime issue guaranteed a crash if you tried this before.
  • Removed a bunch of unnecessary alias methods:
    • columns() -> num_fields()
    • names() -> field_names()
    • rows() -> num_rows()
    • types() -> field_types()
  • Renamed several methods for grammar reasons:
    • fields(unsigned int) -> field(unsigned int)
    • names(const std::string&) -> field_num(const std::string&)
    • names(int) -> field_name(int)
    • types(int) -> field_type(int)
  • Removed several "smelly" methods:
    • purge()
    • raw_result()
    • reset_names()
    • reset_field_names()
    • reset_types()
    • reset_field_types()

• Field class used to just be a typedef for the corresponding C API class. Now it's a real C++ class providing a more MySQL++ sort of interface, plus good OO things like information hiding and implementation detail abstraction. This changes several things about the interface.

• Fields class was basically a specialized std::vector work-alike for dealing with the C API to get access to MYSQL_FIELD objects and present them as contained Field objects. New Field type let us replace it with "typedef std::vector Fields"

• Major improvements to the quoting and escaping mechanisms:

  • Replaced almost all of the type-specific interfaces in manip.h with a single version taking STA. The compiler can convert almost anything to STA without losing any information we need for correct quoting and escaping. This has the side benefit that we can now do correct quoting and escaping for more data types now, including plain C and C++ string types.
  • Fixed a bug in quote_double_only manipulator for String: was using single quotes by mistake.
  • Escaping and quoting only works in instances where MySQL++ can tell you're building a SQL query and are using a data type that requires it. This affects many things, but the one most likely to cause trouble is that inserting MySQL++'s quoting and escaping manipulators in non-Query ostreams is now a no-op.
  • Added escape_string() member functions to Query and SQLQueryParms::escape_string(), and removed the global function of the same name. Because these are tied indirectly to a Connection object, this also has the effect that escaping is now aware of the current default character set used by the database server. There's only one case where this isn't done now, and that's when we're disconnected from the server.
  • Previous two items form a trade-off: if your code was depending on MySQL++ to get SQL escaping and it no longer happens for what we consider a good reason, you can build a replacement mechanism using these new functions. Quoting needs no special support in MySQL++.
  • Removed 'r' and 'R' template query parameter modifiers, which meant "always quote" and "always quote and escape" regardless of the data type of the parameter. There are no corresponding manipulators (for good reason), so the removal restores symmetry.

• Created DBDriver class from code previously in Connection and Query to almost completely wrap the low-level MySQL C API:

  • Connection creates a DBDriver object upon connection and passes a pointer to it down to Query objects it creates. In turn, they pass the pointer on to any of their children that need access to the C API.
  • Nothing outside DBDriver calls the C API directly now, though DBDriver leaks C API data structures quite a lot, so this feature doesn't constitute "database independence." See the Wishlist for what must be done to get to that point.

• Completely redesigned the connection option setting mechanism:

  • There's now just a single Connection::set_option() method that takes a pointer to the abstract Option base class, and there is an Option subclass for every connection option we understand. Thus, type errors are now caught at compile time instead of at run time.
  • Replaced Connection::enable_ssl() with SslOption class.
  • Enabling data compression and setting the connection timeout are no longer set via parameters to Connection interfaces. These are now set with CompressOption and ConnectTimeoutOption.
  • Similarly, removed client_flag parameters from Connection's ctor and connect() method and added corresponding Option subclasses. There's about a dozen, so rather than list them here, look for similarly-named classes in lib/options.h.

• Added Connection::count_rows() to execute SELECT COUNT(*) FROM tablename queries for you.

• Moved Connection::affected_rows(), info() and insert_id() methods to class Query, as they relate to the most recently-executed query, not to the connection.

• Several method name changes in Connection:

  • client_info() -> client_version()
  • host_info() -> ipc_info()
  • proto_info() -> protocol_version()
  • server_info() -> server_version()
  • stat() -> status()

• Removed Connection::api_version(). It does the same thing as client_version().

• Lots of changes to Date, DateTime, and Time classes:

  • If you use the default constructor for DateTime and don't subsequently set its year, month, day, hour, minute or second data members to nonzero values, it becomes the SQL function "NOW()" in a query string. You can also use DateTime::now() as syntactic sugar for this.
  • As a result of above, had to hide all of DateTime's data members behind accessor functions, to keep the state of the object consistent. (If it's initialized as "now" and you give it an explicit year value, say, it is no longer "now", so the setter has to clear the now-flag.) There are getters and setters for year, month, day, hour, minute and second, all named after the member.
  • Did the same for Date and Time for consistency, even though it isn't technically required.
  • The sql_timestamp typedef now aliases DateTime instead of Time.
  • Renamed template class DTbase to Comparable. The fact that it's the common base class of all date and time classes is irrelevant; making subclasses comparable is what it does, so that's what it should be named after.
  • Added a DateTime ctor taking discrete year, month, day, hour, minute, and second values.
  • Implicit conversion from stringish types to the date and time types is no longer allowed. This is part of the "Waba" Null patch mentioned above; allowing implicit conversions would break this new feature.
  • Added operator std::string and str() methods to all of these classes. Adding this to the existing operator << support, you now have several ways to convert these objects to string form.
  • Added time_t conversion to Date and Time classes. DateTime already had it, since it's more legitimate to convert time_t to DateTime, but you could already "slice" it with something like Time(DateTime(time(0))) so there's no point pretending you can't get from time_t to Date or Time. Might as well legitimize it.

• Improved tiny_int class:

  • Turned it into a template parameterized on the value type so you can get both signed and unsigned TINYINTs
  • Changed the sql_tinyint and sql_tinyint_unsigned typedefs to use mysqlpp::tiny_int<VT> instead of raw chars
  • Added a bool conversion ctor and operator, and typedef'd it to sql_bool and sql_boolean to match MySQL server behavior

• Added many more sql_* typedefs. We now have a typedef for every type the MySQL server knows about, including those it supports just for compatibility with other database servers.

• Changed the sql_*int* typedefs to use integer types of the same size as the MySQL server. (Run test/inttypes.cpp to test it.)

• Added copy ctor and assignment operator to Row.

• Row::operator takes int now instead of unsigned int. This finally (!) makes it possible to say row[0] without the compiler giving an ambiguous overload error.

• Changed all uses of row.at(0) in the examples to row[0]

• Added operator[] to all classes that only had at().

• Query now automatically resets itself unless the query fails or you're using template queries. In either case, the contents of the query's internal buffers are considered precious, either for debugging, or future queries. Except when using template queries, this means you may be able to avoid calling Query::reset() entirely. It's still safe to call reset() as before, just unnecessary most of the time.

• Removed reset_query parameter from all Query methods. It was almost completely broken before, and above change does what was really wanted to begin with.

• Query::store_next() and Result::fetch_row() no longer throw the EndOfResults and EndOfResultSets exceptions; these are not exceptional conditions! These methods simply return false now.

• Removed examples/usequery.cpp: there's no essential difference between what it did and what examples/simple3.cpp does now as a result of the previous change.

• Added Query::exec(void), paralleling Query::execute(void).

• Removed Query::preview(). The most direct replacement is str(), which has always done the same thing.

• You can now insert a Query object into an ostream to get a copy of the built query. This means Query::str() is only necessary when using template queries.

• Removed overloads of Query::execute(), store(), and use() that take const char*. It was redundant because const char* converts implicitly to STA, for which overloads already exist.

• Renamed Query::def to Query::template_defaults to make its purpose clearer.

• Query::error() now returns const char*, not a std::string by value. There's no point in making a copy of the error string. The method is now const as well, as it doesn't change the Query object.

• Added Query::errnum(), which just wraps Connection::errnum().

• Added error number parameters and accessor functions to BadQuery, ConnectionFailed and DBSelectionFailed exceptions, to preserve the state of Connection::errnum() at the point of the exception, so you don't have to rely on this value remaining unchanged during the exception throw process. All places that use these exceptions now include this value where possible. Thanks for the initial patch go to Jim Wallace.

• Removed Lockable mechanism from Connection and Query; it was conceptually flawed. See the new user manual chapter on threading for advice on using MySQL++ safely without locking. There is mutex locking now in ConnectionPool, but that's it.

• Connection::query() now takes an optional query string, allowing the returned Query object to start off with a value. Especially useful when the query string is static, either because it's a simple query or because it's a template. You can now build chains like "if (conn.query("CREATE INDEX ...").exec()) { ..."

• Added Connection::thread_aware(), thread_end(), thread_id() and thread_safe(). See user manual's threading chapter for explanations.

• Renamed "success" data members in Connection, Query and SimpleResult (neé ResNSel) to "copacetic_", making them private if they weren't before. This better reflects their actual use, which isn't to say that there has necessarily been actual success, but rather that everything's okay with the object.

• Removed success() member functions from above classes. All can be tested in bool context to get the same information.

• Replaced all operator bool()s in MySQL++ classes with safer alternatives. See http://www.artima.com/cppsource/safebool.html Thanks to Jonathan Wakely for much helpful commentary, advice, and code used in these mechanisms.

• Decoupled Connection::copacetic_ from Connection::is_connected_. It is now possible for the object to be copacetic without being connected. However, if it tries to connect and fails, then it is not copacetic. If it is copacetic and not connected, it means we haven't even tried to connect yet, a useful distinction.

• Collapsed Connection's host, port, and socket name down into a new combined server parameter which is parsed to determine what kind of connection you mean. These interfaces are still compatible with v2.3 and earlier up through the port parameter. There are differences beyond this.

• Added TCPConnection, UnixDomainSocketConnection and WindowsNamedPipeConnection subclasses for Connection giving simpler construction and connect() method interfaces for instances where you know what kind of connection you want at compile time.

• Changed Connection::ping() return value from int to bool.

• Renamed NullisNull to NullIsNull -- capital I -- and similar for NullisZero and NullisBlank.

• It's now a compile-time error to try to convert a MySQL++ representation of a SQL null to any other data type, rather than a run-time error as in previous versions. Removed BadNullConversion exception as a result.

• Way back in v1.7.x we used the BadQuery exception for all kinds of exceptional conditions, not just bad queries. Replaced most of these in v2.0.0 with new dedicated exceptions, but a few remained:

  • Errors that occur during the processing of a "use" query after the query itself executes correctly now throw UseQueryError. It's not a "bad query", because the query executed successfully. It just choked during subsequent processing, so it's a different exception. Thanks for this patch go to Jim Wallace.
  • Replaced BadQuery exceptions thrown in Row constructor due to bad ctor parameters with new ObjectNotInitialized exception This is also Jim Wallace's work.

• The examples now all use getopt() type command line options instead of positional options. This makes it possible to pass options in any order, leave at their default options that used to be in the middle of the sequence, and offer different subsets of options for different programs. Also allows for special internal-use options, like -D passed by dtest to let examples change their behavior when run under dtest to produce only predictable output.

• Split old libutil functionality into two modules, one holding all the "print data" functions, and another holding all the command line parsing stuff. This makes it easier for newbies to ignore the command line stuff, treating it like a black box. The wish to understand the "print data" routines is much more common, so the two needed to be disentangled.

• Renamed examples' libutil to libexcommon.

• Removed connect_to_db() libutil function. It combined command line parsing, which users don't care about, with database connection establishment, which they do care about. Now the examples just call out to libexcommon to parse the command line, and use the values they get back to explicitly make the connection, so it isn't hidden.

• Removed cchar and uint typedefs.

• Redesigned dbinfo example's output to be easier to read.

• Fixed an output formatting bug created in 2.3.0 that caused the tabular output from examples to not line up.

• Renamed examples/tquery.cpp to tquery1.cpp. Created tquery2.cpp to demonstrate passing parameters via a SQLQueryParametrs object instead of discretely. Created tquery3.cpp for testing unquoted template parameters, such as integers.

• Renamed fieldinf1.cpp example to fieldinf.cpp, and simplified its output so it can be part of the dtest sequence.

• Renamed examples/xaction.cpp to transaction.cpp. It created too much cognotive dissonance whenever thinking about both it and lib/transaction.cpp.

• Added examples/deadlock.cpp, to test handling of exceptions due to server-side transaction deadlock detection. Also added code to resetdb to create a table needed to test this. Initial version created by Jim Wallace to test the value of all his BadQuery exception work, with reworking by me.

• Greatly expanded dtest suite. Primary change is that we now have a handful of unit tests, where in v2.3.2 we only tested a subset of the examples. Still very low coverage ratio, but it's a big improvement.

• Optimized #includes, especially in lib/*.h to reduce dependencies and thus compile time when one of these changes.

• Fixed a typo in RPM filename generation that prevented -devel RPM from recognizing that the corresponding MySQL++ library RPM was installed.

• RPM spec file improvements by Remi Collet.

• Renamed NO_LONG_LONGS to MYSQLPP_NO_LONG_LONGS to avoid a risk of collision in the global macro namespace.

• First cut at Xcode2 project support. Testing needed!

• Debug build of library on VC++ and Xcode have a _d suffix now so you can have both versions of the library installed without conflict.

• Moved the VC++ 2003 project files into a new vs2003 subdirectory because there are so many of them. Also created vs2005 subdirectory for VC++ 2005 and 2008 compatible project files. 2005 makes an even bigger mess of the directory containing the .sln file, so the incentive is bigger. Plus, we have to disable several things to get VC++ 2003 to build MySQL++ now, so we need a special 2005+ version of the project files for a complete build, if the user has one of the newer compilers.

• ...plus dozens of small bug fixes and internal enhancements, many documentation improvements, and expansion of support for newer operating systems and compilers.

2.3.2, 2007.07.11

NOTE: Because this is the last release in the 2.x line, and because I have no plans to make another formal 2.x release, there is now a branch with minimal changes to allow this version of the software to build on modern systems.

• Previous release's const_string change caused more problems than it fixed. This release contains the real fix. :)

• New Connection::set_option() handling deals with the multi statements option correctly again. examples/multiquery now runs again as a result.

• Added new unit testing script, called dtest. See the HACKERS file for details. (This tool caught the previous two problems!)

• Squished a GCC pedantic warning. Thanks for the patch go to Andrew Sayers.

2.3.1, 2007.07.10 (r1659) The "After the Fireworks" release

• const_string objects now keep a copy of their data, not just a pointer to it. This is less efficient, but necessary to allow SSQLS to work with BLOBs. Without this, we were seeing segfaults due to accessing freed memory pointed to by the const_string, because the underlying object went out of scope.

• Fixed many more potential embedded null handling problems in manip.h.

• MySQL++ can now optionally reference MySQL C API headers as being in a mysql subdirectory, a common thing on *ix systems, by defining MYSQLPP_MYSQL_HEADERS_BURIED before #including mysql++.h.

• Restored ColData_Tmpl<T>::get_string(), removed in v2.3.0, along with warnings in the docs saying why you don't want to use it, and what your alternatives are.

• VC++ and MinGW builds now define the HAVE_MYSQL_SSL_SET macro, which lets you use the C API's SSL features. This assumes your C API library does actually have these features enabled, which is the case with the official binary releases on Windows. (Builds on *ix systems continue to test for these features at configure time.)

• Fixed simple examples-only Makefile generation, for RPMs.

2.3.0, 2007.07.02

• Added Query::for_each() and Query::store_if() methods proposed by Joel Fielder, and added examples for each.

• It's now possible to store BLOB data in an SSQLS. It's not foolproof, so added a section to the user manual (5.9) to document the method. Also, changed examples/cgi_jpeg to use this new mechanism, instead of the ugly "raw row data" method it used to use.

• Revamped Connection::set_option() handling. These options used to be queued up, and applied only just before actually establishing the connection. This made error reporting less helpful because the diagnostic was separated from the cause. Plus, the error messages were misleading to begin with. Now, set_option() takes effect immediately if the connection is not yet up (excepting one special option that can actually be set after the connection is up) and issues better diagnostics when it detects errors.

• Connection::connect() used to set a few options in such a way that the user couldn't override them. Now it's smart enough to set them with the desired default values only when we see that the user hasn't given them other values.

• SQLString can now be initialized from a mysqlpp::null, giving a "NULL" string. This is useful for template queries. Patch by Michael Hanselmann.

• resetdb error message about mixing library and header version numbers is now more explicit.

• Changed BadConversion exception's "what" message text to be more like the other exceptions. The inconsistency lead one to incorrectly copy-paste code from another exception handler, expecting it to behave the same way. Now it does.

• Added Row::raw_size(), as a shortcut for Row::at().size().

• ssqls-pretty now detects when it's being run from within the MySQL++ distribution tree and gives a different -I flag to the compiler, so that it picks up the distribution headers instead of those that may be on the system already.

• The quote manipulator now works for char[] correctly. Thanks for this patch go to Andrew Sayers. (It's always worked for char*, but C++ doesn't consider that to be the same type, so it used the generic quote handling path, which doesn't do anything for char[].)

• Fixed a build bug on older Solaris versions where the test for the C API library was erroneously failing, stopping the configuration process.

• Simplified mysql_shutdown() level argument detection. Already had to do a version number ifdef check for the Windows case, so there's really no point to doing it with autoconf on Unixy platforms. Moved version number check into lib/connection.cpp, and nuked the separate autoconf and Windows tests.

• Removed dependency of sql_types.h on myset.h and (indirectly) datetime.h. Now we only define sql_* typedef aliases for those MySQL++ types if the headers are included before sql_types.h.

• Fixed a typo in one of the storein_sequence() template overloads, which is apparently rarely (or never?) used, because no one reported the compiler error you'd get if you tried.

• Fixed a few more embedded null handling problems.

• ColData used to keep two copies of all data it held. Now it keeps just one.

• Fixed install.bat script to track the unified Bakefile change and the lack of separate debug and release builds under MinGW.

• Yet another STLport + Query memory leak fix.

• Squished a warning in newer GCCs having to do with identifier shadowing. Patch by Jonathan Wakely.

• Fixed a null-termination bug in Query::parse(). If you somehow constructed a query string without a terminating null character, then tried to parse it as a template query, it could walk off the end of the string. Patch by Worster Chen.

• Removed MYSQLPP_EXPORT tag from FieldNames and FieldTypes class declarations, as this can cause problems in programs that use vector in VC++. It has to do with multiply defined templates, since these classes derive from that template, and VC++ can't resolve the conflict without help. Since these classes aren't actually used outside the library, this shouldn't cause a problem. Patch by Nils Woetzel.

• Partial fix to Doxygen PDF build on RHEL4 and 5. Needs hand-coaxing to complete successfully on RHEL4, and doesn't yet work at all on RHEL5.

• Shortened the no* options to the bootstrap script, so that the usage message fits on a single line.

• Added "nodoc" bootstrap script option, for disabling the documentation build during the dist target build. Allows for building binary RPMs on CentOS 5.0, where doc building is currently broken.

• Removed the updel example program. It was kind of silly, and if you were to rewrite it today, you'd use for_each() anyway.

• Lots of documentation improvements.

2.2.3, 2007.04.17 (r1538) The "Tax Day" release

• Previous version left examples/vstudio/* out of the tarball by accident.

• Improved generation of RPM temporary build directory path name generation. Was using a hacked variant of the Fedora Packaging Guidelines' second best choice. Now we're using the choice they recommend most highly, without changes.

• Removed unnecessary resources from vstudio/wforms example.

• Minor URL fix in refman

2.2.2, 2007.04.13 (r1526) The "Nervousmaking Friday the 13th" release

• More small fixes to embedded null handling in Query.

• Fixed a bug in single-parameter template query handling.

• Added tquery example, to demonstrate proper use of template queries. Previously, resetdb was the only exemplar, and it wasn't really suited for that. This example also tests the previous item.

• Added examples/vstudio/mfc, allowing us to improve the way we demonstrate Unicode handling. Old way wasn't realistic. On *ix, people will depend on the terminal code to handle UTF-8. On Windows, users are almost certain to be writing a GUI program, which requires different Unicode handling than the old examples showed.

• Removed explicit Unicode conversion stuff from command line examples, and reworked the Unicode chapter in the user manual.

• Added examples/vstudio/wforms to show integration with C++/CLI and Windows Forms. Documented this in README.vc.

• Rewrote load_file and cgi_image examples to be more useful, renaming them to load_jpeg and cgi_jpeg along the way. Also, resetdb now creates a second table in the sample database for these two examples' use. Also, added examples/logo.jpg to the distribution as sample data for these examples.

• Limited the ostream base class casting stuff in Query to VC++ 2003, which is the only platform that really needed it. VC++ 2005 emits a warning with that hack in place, and on other platforms it's just replicating work that the compiler does already.

• Added library version information to main library target so that systems that version shared libraries work as expected. Thanks for this patch go to Jack Eidsness.

• Merged much of the diffs between Remi Collet's RPM spec file into the official one.

• Reorganized the doc subdir a bit. Generated HTML is now all under doc/html instead of scattered under other subdirs, and renamed doc/README.mysql++ to doc/README.manuals.

• Improvements to top-level manual building make targets: manuals now only rebuild at need, it's easier to request a rebuild of all manuals, and we force a rebuild attempt before building the distribution tarball so we don't ship outdated manuals.

• Added ability to run examples under gdb using exrun, using same mechanism as we currently have for valgrind. Thanks for this patch go to Michael Hanselmann.

• Added "Important Underlying C API Limitations" chapter to the user manual, to cover problems we keep seeing on the mailing list that are the result of ignorance of the way libmysqlclient behaves, not bugs MySQL++ is really in a position to fix.

2.2.1, 2007.02.28

• Fixed the new localtime() alternative selection code for VS2003 and various uses of STLport.

• No longer inserting a null character into the query stream on calling one of the preview() functions. This was harmless in v2.1, which used C strings more extensively, but began causing problems in v2.2 due to its wider use of C++ strings.

• Fixed a bug in the Connection copy ctor where it didn't completely initialize the object.

• Optimized Query::preview_char() a bit. Patch by Jonathan Wakely.

• Reordered directory list used by autconf when locating the MySQL C API library. The list is now ordered with the most likely locations for the library first, so we're less distracted by incorrect libraries. This fixes a specific build error under RHEL4 with recent versions of MySQL 5.0.

2.2.0, 2007.01.23

• ColData, const_string, and SQLString can now be constructed with an explicit length parameter. Furthermore, Query class's execute(), store() and use() call chains terminate in a version taking an explicit length parameter, instead of one taking a simple C string. Together, this means that it's now easier to handle data from the SQL server containing nulls. The library is almost certainly not yet capable of handling embedded nulls in all cases, but this is a big first step towards that.

• Can now construct a DateTime object from a time_t, and convert a DateTime back to a time_t. Patch by Korolyov Ilya.

• Changed the way we're handling exported functions in the Windows DLL case so that it works more reliably under MinGW.

• Added proper copy semantics to Connection, so that you get a new connection with the same parameters, not just a bitwise copy of the object.

• Using an explicitly thread-safe variant of localtime() for time conversions where one is available.

• Removed ListInsert template from myset.h. This wasn't used within the library, and was never documented, so I'm betting that no one actually uses it.

• Result::copy() was not copying the exception flag in all cases. Fix by Steven Van Ingelgem.

• Added exrun shell script and exrun.bat files to distribution, to avoid linkage errors when running the examples while you still have an older version of MySQL++ installed.

• Renamed MYSQLPP_LIB_VERSION to MYSQLPP_HEADER_VERSION, as what it really encodes is the version number in the mysql++.h file you're using, not the actual library version number.

• Added mysqlpp::get_library_version(), which returns the library version number at build time. Between this and the header version constant, you can check that you're not mixing MySQL++ header and library versions.

• resetdb example uses these new version number affordances to double-check that you're not mixing libraries and headers from different versions. This happens easily unless you take care of it (such as by using exrun) when you have one version of MySQL++ installed and you're trying to build and test a new version without blowing away the old one first or overwriting it.

• No longer using recursive Makefiles on Unixy platforms or split lib + examples project files on VC++. Everything is handled by a single top-level Makefile or project file, which is simpler for the end user, and makes better dependency management possible.

• When looking for the MySQL C library on systems using autoconf, looking in .../lib64 wherever we are also looking in .../lib.

• RPM build process no longer depends on Bakefile. It means you have to build the examples when building an RPM even though they're never used within the RPM, but it's a better tradeoff in my opinion.

• Updated include and library paths on Windows to reflect changes in the most recent MySQL installers.

• Merged lib/defs.h and lib/platform.h into new file, lib/common.h. Just cleans up the library internals.

• Fixed build errors on Windows due to recent changes in MySQL.

• Fixed a few memory leaks and double-deletes in Query class.

• Fixed compatibility with STLPort's string implementation. Patch by dengxy at cse.buaa.edu.cn.

• Fixed a compatibility problem between Set<> template and SSQLS. Patch by Korolyov Ilya.

• Fixed build bug in SQLQueryParms due to a character signedness issue on PowerPC with GCC. Patch by Michael Hanselmann.

• ~Transaction() can no longer throw exceptions. It'll just quietly eat them, to avoid program termination. Fix suggested by Alex Burton.

• Fixed thread safety testing in autoconf case, accidentally broken during v2.1.0 development cycle.

• Using Doxygen 1.5.1 to generate documentation.

2.1.1, 2006.04.04

• MinGW and Cygwin will now build and link to mysqlpp DLLs.

• Fixed bug in Query, causing it to initialize the "throw exceptions" flag incorrectly. Thanks for this patch go to Joel Fielder.

• Added -v flag for custom.pl script, which turns off the multiply-defined static variable fix. Needed for VS 2003, which doesn't support variadic macros. Also, added a diagnostic to detect the need for the -v flag, and suppressed the test for this feature in examples/util.cpp.

2.1.0, 2006.03.24

• Converted automake and makemake files to their equivalents in Bakefile format.

• Added the Transaction class, which makes it easy to use transaction sets in MySQL++.

• Added xaction example to test new Transaction class.

• Resetdb example now creates its example table using the InnoDB storage engine, in order to test the new transaction support. Resetdb also declares the table as using UTF-8 text; this doesn't change anything, but it does correctly document what we're doing.

• Added sql_types.h header, containing C++ typedefs corresponding to each MySQL column type. Using those new types in the type_info module, and in the SSQLS examples.

• Replaced the way we were handling the template query version of Query member functions, to allow an arbitrary number of template query parameters. By default, we now support 25 parameters, up from the old limit of 12. It's now possible to change just one number, run a script, and have a new limit.

• Connection class does a better job of returning error messages if you call certain member functions that depend on a connection to the server before the connection is established.

• Updated libmysqlclient.def for newer versions of MySQL. (Fixes build errors having to do with mysql_more_results() and mysql_next_result().

• Replaced final use of strcpy() with strncpy().

• custom.pl now runs without complaint in strict mode, with warnings turned on. Thanks for this patch go to "Waba".

• Fixed a bug in custom.pl where incorrect code would be generated for some SSQLS set() methods. Thanks for this patch go to "Waba".

• SSQLS structures now support long and unsigned long fields. Thanks for this patch go to "Waba".

• It's now possible to put SSQLS definitions in a header file used by multiple modules in a program without getting multiple static member definition errors. See the documentation for details. Thanks for this patch go to Viktor Stark.

• Moved the definition of the 'stock' SSQLS out of the custom*.cpp example files and into a new stock.h file. Also, #including that file in the util module to test out the new SSQLS multiple static definition fix.

• Using all of the digits of precision guaranteed by the IEEE 754 spec when stringizing floating point numbers to build queries. Previously, we would use the platform default, which can be as few as 6 digits.

• Removed lib/compare.h. Not used within the library, never documented, and nobody seems to want to defend it.

2.0.7, 2005.11.23

• Added explicit mysqlpp namespace qualifiers to generated code in custom*.h so you can use SSQLS in places where it doesn't make sense to say "using namespace mysqlpp" before the declaration. Also updated some of the examples to not have this "using" declaration to make it clear to users that it isn't needed, if you want to use explicit namespace qualifiers as well. Thanks for this patch to Chris Frey.

• Removed an apparently useless unlock() call from ResUse; there is no nearby lock() call, so if this unlock() is in fact necessary, it shouldn't be here anyway, because the two calls should be nearby each other. Thanks for this patch to Chris Frey.

• Fixed Query ostream initialization bug affecting SunPro CC (at least). While this bug violates the Standard, it doesn't affect many real compilers because they don't enforce this rule. Fixed by Chris Frey.

• Previously, we only used the C99 style "long long" support when building under GNU CC. This is now the default. This should allow the code to work under SunPro CC.

• Added another dynamic cast needed for proper Query ostream subclass overloading under VC++. (7.1 at least...)

• Detecting whether MySQL is built with SSL support on platforms using autotools. Needed on some old Sun systems, for instance. Thanks for this patch to Ovidiu Bivolaru.

• Fixed a potential memory bug in ColData's conversion to SQL null.

• Many minor packaging tweaks. (README clarifications, file permission fixes, better adherence to GNU packaging standards, etc.)

2.0.6, 2005.09.28

• Fixed makemake.bat so it works on cmd.exe, not just 4NT.

• Documentation fixes.

2.0.5, 2005.09.13

• Visual C++ build now requires GNU make. It is tested to work with either the Cygwin or the MinGW versions. The previous version of MySQL++ used nmake. This change enabled the following features:

  o Debug and Release versions are both built into separate subdirectories.

  o Dependency tracking for release version works correctly now. (Previously dependencies worked only for debug version.)

  o 'make clean' removes release version binaries in addition to debug versions.

• MinGW makemake support updated to support new release/debug subdirectory system. This is probationary support, since this code currently can't be built as a DLL. As a result, it is no more useful than the Cygwin version, for licensing reasons.

• Several fixes to allow building on Solaris 8. These fixes may also help on other SVR4-derived systems.

• Removed Borland C++ makemake support, because this version of the library does not work completely, and there seems to be almost no user interest in fixing it.

• Clarified "Handling SQL Nulls" section of user manual's Tutorial chapter.

2.0.4, 2005.08.29

• Made mysql_shutdown() second parameter autoconf check less sensitive to compiler pedantry.

• VC++ library Makefile is now smart enough to re-create the import library, if it is deleted while leaving the DLL alone.

• Added libmysqlclient.def to tarball.

• Reworked most of the top-level README* files.

• Renamed LGPL file to LICENSE.

2.0.3, 2005.08.25

• Visual C++ makemake system updated to build both debug and release versions of library DLL.

• Fixed bug in simple1 example that caused crashes on Windows.

• Doing UTF-8 to ANSI text translation in simple examples now.

• Previous two releases built libmysqlpp with wrong soname on autotools-based systems. Fixed.

2.0.2, 2005.08.18

• Fixes to makemake system for cmd.exe.

• Fixed the case where the system's C++ library includes an slist implementation in namespace std.

2.0.1, 2005.08.17

• Added new simple1 example, showing how to retrieve just one column from a table. Old simple1 is now called simple2, and simple2 is likewise shifted to simple3.

• Added custom6 example, showing how to do the same thing with SSQLS.

• Updated user manual to cover new examples.

• Was accidentally shipping Subversion crap with tarball. Fixed.

2.0.0, 2005.08.16 (r1031) The "Excess Hair Removal" release

WARNING: THIS IS NOT A DROP-IN REPLACEMENT FOR MySQL++ v1.7!

At minimum, you will have to recompile your program against this library. You may also have to make code changes. Please see the "Incompatible Library Changes" chapter of the user manual for a guide to migrating your code to this new version.

• The library's shared object file name (soname) scheme has changed. (This mainly affects POSIX systems.)

  The soname for the last 1.7.x releases of MySQL++ was libmysqlpp.so.4, meaning the fourth version of the library's application binary interface (ABI). (The first ABI version in this scheme was that provided by 1.7.9.) MySQL++ 2.0.0's soname is libmysqlpp.so.2.0.0. Since the dynamic linker setup on some systems will create a symlink to that file called libmysqlpp.so.2, it's possible that this library could be confused with that for MySQL++ 1.7.19 through .21, which also used this number. Do not install this library on a system which still has binaries linked against that version of the library!

  The new scheme is {ABI}.{feature}.{bug fix}. That is, the first number changes whenever we break the library's binary interface; the second changes when adding features that do not break the ABI; and the last changes when the release contains only internal bug fixes. This means that we will probably end up with MySQL++ 3.0 and 4.0 at some point, so there will be further soname conflicts. Hopefully we can put these ABI changes off long enough to avoid any real problems.

• autoconf now installs headers into $prefix/include/mysql++, instead of $prefix/include. If you were using the --includedir configure script option to get this behavior before, you no longer need it.

• Linux binary RPMs will henceforth include only the libmysqlpp.so.X.Y.Z file, and create any short names required, to allow multiple versions to be installed at once. Currently, you cannot install two MySQL++ library RPMs at once, because they both have /usr/lib/libmysqlpp.so.X, for instance.

• Replaced the Visual C++ and Borland C++ project files with a new "makemake" system, which creates Makefiles specific to a particular toolchain. This new mechanism also supports MinGW and generic GCC-on-*ix. This was done partly to reduce the number of places we have to change when changing the file names in MySQL++ or adding new ones, and partly so we're not tied to one particular version of each of these tools.

• VC++ Makefiles create a DLL version of the library only now, so there's no excuse for LGPL violations now. This same mechanism should make DLL builds under other Windows compilers easy.

• Added Connection::enable_ssl(), which enables encrypted connections to the database server using SSL.

• Connection::create_db() and drop_db() now return true on success, not false.

• Connection::create_db() and drop_db() use Query::exec() now, for efficiency, rather than Query::execute().

• Removed Connection::infoo(). Apparently just there to save you from a typo when calling the info() method, since it was a mere alias.

• Renamed Connection::real_connect() to connect(), gave several more of its parameters defaults, and removed old connect() function. Then changed user manual and examples to use new APIs.

• Replaced Connection::read_option() with new set_option() mechanism. The name change matches the method's purpose better. Functional changes are that it returns true on success instead of 0, it supports a broader set of options than read_option() did, and it enforces the correct option argument type.

• You can now call Connection::set_option() before the connection is established, which will simply queue the option request up until the connection comes up. If you use this feature, you should use exceptions, because that's the only way an option setting failure can be signalled in this case.

• Removed query-building functions (exec*(), store*(), use()) from class Connection, and moved all the implementation code to class Query. Query no longer delegates the final step of sending the query to the database server to Connection().

• Added Connection::enable_ssl(), for turning on SSL support on a connection.

• Extracted exception disabling mechanism out of the many classes that had the feature into a new OptionalExceptions base class, which all classes having this feature now derive from. Also, removed all per-method exception handling flags. Finally, added NoExceptions class. With all of these changes, there is now a common way to disable exceptions with fine granularity on all objects that support the feature.

• All custom MySQL++ exceptions now derive from the new Exceptions class. This regularizes the exception interface and allows you to use a single catch() block if you want.

• The "throw exceptions" flag is passed from parent to child in all situations now. (Or if not, please report it as a bug.) This fulfills a promise made in the v1.7.9 user manual, with the cost being that some programs will see new exceptions thrown that they're not expecting.

• Added a bunch of new exception types: BadOption, ConnectionFailed, DBSelectionFailed, EndOfResults, EndOfResultSets, LockFailed, and ObjectNotInitialized. Some of these replace the use of BadQuery, which in v1.7.x was a kind of generic exception, thrown when something more specific wasn't available. Beware, this means that programs may start crashing after recompiling them under v2.0 due to uncaught exceptions, if they were only trying to catch BadQuery.

  There are additional instances where the library will throw new exceptions. One is when calling a method that forces the internals to use an out-of-bounds index on a vector; previously, this would just make the program likely to crash. Another is that the library uses the BadFieldName exception -- created in v1.7.30 -- in more apropos situations.

• Renamed SQLQueryNEParms to BadParamCount, to match naming style of other concrete exception types.

• Extracted lock()/unlock() functions from Connection and Query classes into a new Lockable interface class. Locking is implemented in terms of a different class hierarchy, Lock, which allows multiple locking strategies with a single ABI.

• Removed ResUse::eof(). It's based on a deprecated MySQL C API feature, and it isn't needed anyway.

• Removed arrow operator (->) for iterator returned by Fields, Result and Row containers. It was inherently buggy, because a correct arrow operator must return the address of an object, but the underlying element access functions in these classes (e.g. at()) return objects by value, of necessity. Therefore, this operator could only return the address of a temporary, which cannot be safely dereferenced.

• Returned Row subscripting to something more like the v1.7.9 scheme: there are two operator[] overloads, one for an integer (field by index) and another for const char* (field by name). lookup_by_name() has been removed. Because row[0] is ambiguous again, added Row::at() (by analogy with STL sequence containers), which always works.

• Collapsed two of the Row::value_list*() overloads into two other similar functions using default parameters. This changes the API, but the removed functions aren't used within the library, and I doubt they are used outside, either.

• Merged RowTemplate into Row.

• Merged SQLQuery class into Query class.

• Query is now derived from std::ostream instead of std::stringstream, and we manage our own internal string buffer.

• Moved SQLParseElement and SQLQueryParms into their own module, qparms.

• Added multiple result set handling to Query. MySQL 4.1 and higher allow you to give multiple SQL statements in a single "store" call, which requires extensions to MySQL++ so you can iterate through the multiple result sets. Also, stored procedures in MySQL 5.0 reportedly return multiple result sets. Thanks for the initial patch go to Arnon Jalon; I reworked it quite a bit.

• Query::storein*() now supports more varieties of the nonstandard slist comtainer. (Singly-linked version of STL std::list.)

• Template query mechanism and user manual had several mismatches. Made manual match actual behavior, or made library match documented behavior, as apropriate. Initial patch by Jürgen MF Gleiss, with corrections and enhancements by Warren Young.

• Collapsed mysql_* date and time base classes' methods and data into the subclasses. Also, DateTime no longer derives from Date and Time; you could get away with that in the old hierarchy, but now it creates an inheritance diamond, and allows unsupported concepts like comparing a Time to a DateTime.

• Removed "field name" form of Row::field_list(). It was pretty much redundant -- if you have the field names, why do you need a list of field names?

• ColData can convert itself to bool now. Thanks for this patch go to Byrial Jensen.

• Removed simp_list_b type; wasn't being used, and doesn't look to be useful for end-user code.

• Several methods that used to take objects by value now do so by const reference, for efficiency.

• Several variable and function renamings so that MySQL++ isn't needlessly tied to MySQL. Even if we never make the library work with other database servers, there's little point in tying this library to MySQL blindly.

• Renamed all private data members of MySQL++ classes to have trailing underscores.

• 'private' section follows 'public' section in all classes now.

• Removed mysql++.hh and sqlplus.hh backwards-compatibility headers.

• Added copy ctors to Date/Time classes so that they will work in SSQLS under GCC 4.0.0. Without these, the compiler couldn't make the conversion from raw MySQL row data.

• Fixed a bunch of GCC 4.0 pedantic warnings: added virtual dtors to all base classes, calling base class ctors from leaf classes, etc.

• All warnings fixed under VC++ at warning level 3. (Mostly harmless signedness and integer conversion stuff.)

• Updated LGPL license/copyright comments at the top of several files to use FSF's new physical address.

• Relicensed user manual under a close variant of the Linux Documentation Project License, as it's designed for documentation, which the LGPL is not. Permission for this received from Kevin Atkinson and MySQL AB.

• Added ABI and API breakages chapter to user manual. It is basically a subset of this change log, with only the information an end-user must know when migrating between versions.

• Reworked user manual's DocBook code quite a bit after reading Bob Stayton's book "DocBook XSL" 3/e. Better handling of stylesheets, taking advantage of some superior DocBook features, prettier output (especially the HTML version), etc.

• Rewrote doc/userman/README to make it clearer how to get started contributing to the user manual. It's essentially a "getting started with DocBook" guide now!

• Lots of small text improvements to user and reference manuals. Aside from the obvious tracking of library changes, made a bunch of minor style and clarity improvements.

• Added CSS stylesheets for userman and refman to make the HTML versions of each a) not ugly; and b) match tangentsoft.com. (Yes, some may say that these are incompatible goals....)

• Standardized exception handling code in the examples that use it.

• Fixed a potential memory leak due to exceptions thrown from ResUse. Thanks for this patch go to Chris Frey.

• Using new "no exceptions" feature of library in simple1 example, so it is now truly simple.

• simple1 example no longer depends as much on util module, so that all of the important code is in one place. Makes learning MySQL++ a little less intimidating.

• Added new simple2 and usequery examples, to demonstrate the proper way to handle a "use" query, with exceptions disabled, and not, respectively. Added them to the user manual, in the appropriate place.

• Refactored the "print stock table" example functions again, to make code using them clearer.

• UTF-8 to UCS-2 handling in examples is now automatic on Windows.

• Removed debug code from Windows Unicode output examples that slipped into previous release.

• resetdb example is now clearer, and more robust in the face of database errors.

• Simplified connect_to_db() in examples' util module.

• Added sample autoconf macro for finding MySQL++ libraries, for people to use in their own autotools-based projects.

• Lots and lots of minor cleanups not worth mentioning individually...

1.7.40, 2005.05.26

• Multiple item form of insert() now works if you're using the SQLQuery class, or its derivative, Query. Thanks to Mark Meredino for this patch.

• Fixed a bug in const_string::compare(), in which MySQL++ would walk off the end of the shorter of the two strings. All was well if the two were the same length.

• ResUse::operator=() now fully updates the object, so it's more like the behavior of the full ctor.

• All source files now contain a license and copyright statement somewhere within them.

• Optimized mysql++.h a bit: it now #includes only the minimum set of files required, and there is now an idempotency guard. This improves compile times a smidge, but mainly it was done to clean up the generated #include file graph in the reference manual. Before, it was a frightful tangle because we #included everything except custom*.h.

• Constness fix in MySQL++ date/time classes to avoid compiler warnings with SSQLS. Thanks to Wolfram Arnold for this patch.

• Fixed some compiler warnings in custom*.h. Thanks to Chris Frey for this patch.

• Added "Submitting Patches" and "Maintaining a Private CVS Repository" sections to the HACKERS file. Thanks to Chris Frey for the source material for these sections. The HACKERS file was improved in several other ways at the same time.

• PDF version of user manual no longer has links to the reference manual. They were ugly, and they were broken anyway due to the way we move the PDFs after generating them. If you want interlinked manuals, use the HTML version.

• PDF version of user manual now has hard page breaks between chapters.

• Removed complic1 example. Wasn't pulling its own weight. Everything it is supposed to demonstrate is shown in other examples already.

• Refactored print_stock_table() in examples/util module to be four functions, and made all the examples use various of these functions where appropriate. Before, several of the examples had one-off stock table printing code because print_stock_table() wasn't exactly the right thing, for one reason or another. One practical problem with this is that some of the examples missed out on the recent Unicode updates; now such a change affects all examples the same way.

• Since so many of the examples rely on the util module, the user manual now covers it. The simple1 example in the user manual didn't make much sense before, in particular, because it's really just a driver for the util module.

• Added custom5 example. It shows how to use the equal_list() functionality of SSQLS. Thanks to Chris Frey for the original version of this program. (I simplified it quite a bit after accepting it.)

• New user manual now covers the value_list(), equal_list() and field_list() stuff that the old manual covered but which was left out in previous versions of the new manaul. Most of the examples are the same, but the prose is almost completely new. This new section includes the custom5 example.

• Every declaration in MySQL++ is now documented in the reference manual, or explicitly treated as "internal only".

• Improved docs for MySQL++'s mechanism to map between MySQL server types and C++ types. Initial doc patch by Chris Frey, which I greatly reworked.

• Improved a lot of existing reference manual documentation while adding the new stuff.

• Expanded greatly on the exception handling discussion in the user manual.

• Added all-new "Quoting and Escaping" section to the user manual's Tutorial chapter. Moved some existing comments on quoting and escaping around and added some new ones to other sections as a result.

• Added all-new "Handling SQL Nulls" section to the user manual's Tutorial chapter.

• Many improvements to the Overview section of the user manual.

• Row::operator[] reference now explains the right and wrong way to use the values it returns. This is in response to a mailing list post where someone was incorrectly using this feature and getting a bunch of dangling pointers.

• Updated Doxyfile so 1.3.19.1 parses it without warnings. Still works with versions back to 1.2.18, at least. (These are the versions shipped with Fedora Core 3 and Red Hat Linux 9, respectively.)

• Using a superior method to make Doxygen ignore certain sections of the source code. Between this change and the fact that everything not so ignored is documented, Doxygen no longer generates any warnings.

• Lots of code style updates. Everything should now be consistently formatted.

1.7.35, 2005.05.05 (r601) The "Cinco de Mayo" release

• Added a "how to use Unicode with MySQL++" chapter to the user manual. (Too bad "Cinco de Mayo" doesn't have any accented characters. That would be just too precious.)

• VC++ examples now use the Unicode Win32 APIs, so they can display Unicode data from MySQL++.

• Added an optional conversion function to examples/util.cpp to handle the conversion from UTF-8 to UCS-2 on Win32.

• Moved "brief history of MySQL++" from intro section of refman to intro section of userman.

• Lots of small bits of documentation polishing.

• Made some minor constness fixes. Thanks to Erwin van Eijk for this patch.

• Made some warning fixes for GCC 4.0. Not all warnings are fixed, because some of the needed changes would break the ABI. Thanks to Chris Frey for this patch.

• Added lib/Doxyfile to distribution.

1.7.34, 2005.04.30

• Added a multiple-insert method for Query, which lets you insert a range of records from an STL container (or the whole thing, if you like) in a single SQL query. This is faster, and it reduces coding errors due to less repetition. Thanks to Mark Meredino for the patch.

• Reference and user manual now get rebuilt automatically when required. (E.g. on 'make dist', or explicitly now through 'make docs'.)

• Made it easier to change the maximum number of SSQLS data members in generated custom-macros.h file. It used to be hard-coded in several places in lib/custom.pl; now it's a variable at the top of the file.

• Changed default SSQLS data member limit to 25, which is what it has been documented as for a long time now. It was actually 26 within custom.pl.

• Fixed a regression in previous version.

• Trimmed some fat from the distribution packages.

• Some more small doucmentation improvements.

1.7.33, 2005.04.29

• Worked around an overloaded operator lookup bug in VC++ 7.1 that caused SSQLS insert, replace and update queries to get mangled. (Symptom was that custom2 and custom3 examples didn't work right.) Thanks to Mark Meredino for digging up the problem's characterization and its solution.

• Some VC++ warning fixes.

• Major documentation improvements:

  o Using DocBook for user manual and Doxygen for reference manual. The former now references the latter where useful.

  o Split out HACKERS and CREDITS files from main README, and improved remaining bits of README.

  o Moved the text from the old v1.7.9 LaTeX-based documentation over into the new systems, and reworked it to more closely resemble English.

  o Added a lot of new material to documentation, and simplified a lot of what already existed.

  o Documentation is now being built in HTML and PDF forms.

• ebuild file updated to take advantage of recent configure script features. Thanks to Chris Frey for this patch.

1.7.32, 2005.03.10

• Example building may now be skipped with --disable-examples configure script flag.

• Changed stock items added in resetdb. One is now UTF-8 encoded, to show that basic use of Unicode with MySQL++ is easy, yet not foolproof. (See formatting of table on systems where cout isn't UTF-8 aware!) Other stock items now follow a theme, for your amusement. :)

• custom3 example now changes UTF-8 item's name to the 7-bit ASCII equivalent. Previously, this example would fix a spelling error in the table.

• resetdb example now says 'why' when it is unable to create the sample database.

• Small formatting change to print_stock_table(), used by several examples.

• Was issuing a VC++-specific warning-disable pragma when built by any Windows compiler. Fixed.

1.7.31, 2005.03.05 (r462) The "Inevitable Point-one Followup" release

• Check for threads support must now be explicitly requested via configure script's new --enable-thread-check flag.

• Fix for contacting MySQL server on a nonstandard port number. Thanks to Chris Frey for this patch.

• Example programs using standard command line format now accept a fourth optional parameter, a port number for the server. Thanks to Chris Frey for this patch.

• One more g++ 3.4 pedantic warning fix by Chris Frey.

• Exception handling in resetdb is no longer nested, because you'd get a segfault on some systems when an exception was thrown from one of the inner try blocks.

• Improvements to Connection class's handling of locking mechanism. Concept based on patches by Rongjun Mu.

• Implemented the declared-but-never-defined Query::lock(). Thanks to Rongjun Mu for this patch.

• Cleaned up some unclear if/else blocks in connection.cpp by adding explicit braces, correct indenting and putting normal code path in the if side instead of the else.

1.7.30, 2005.02.28 (r443) The "Power of Round Numbers" release

• bootstrap script now accepts a 'pedantic' argument, which sets a bunch of CFLAGS that make g++ very picky about the code it accepts without warnings.

• Fixed a bunch of things that generated warnings with g++ in pedantic mode. Only two warnings remain, having to do with floating point comparisons. (See Wishlist for plans on how to deal with these.) Thanks to Chris Frey for this patch.

• Split long tests out of configure.in into M4 files in new config subdir. This makes configure.in easier to read.

• Added preliminary thread support. Currently, this just means that we detect the required compiler and linker thread flags, and link against the proper thread-safe libraries. THERE MAY BE UN-THREAD-SAFE CODE IN MYSQL++ STILL!

• Standard C++ exceptions are the default now. Old pre-Standard exception stuff removed.

• Row::lookup_by_name() will throw the new BadFieldName exception if you pass a bad field name. Thanks for this patch to Chris Frey.

• Row::operator[] will throw a Standard C++ out of bounds exception by way of std::vector::at() if you pass it a bad index. Thanks for this patch to Chris Frey.

• Setting Connection::is_connected flag to false on close(). Previously, is_connected() would continue to return true after close() was called.

• All number-to-string conversion ctors in SQLString class now use ostringstream to do the conversion. Previously, we used snprintf(), which isn't available on all systems. Also, we used a C99 format specifier for the "long long" conversion, which is also not available on all systems. This new ostringstream code should be platform-independent, finally.

1.7.28, 2005.02.04

• --with-mysql* flags to configure script now try the given directory explicitly, and only if that fails do they try variations, like tacking '/lib' and such onto it to try and find the MySQL includes and libraries. Thanks to Matthew Walton for the patch.

• Finally removed sql_quote.h's dependence on custom.h by moving the one definition it needed from custom.h to deps.h. This will help portability to compilers that can't handle the SSQLS macros by making that part of the library truly optional.

1.7.27, 2005.01.12

• configure check for libmysqlclient now halts configuration if the library isn't found. Previously, it would just be flagged as missing, and MySQL++ would fail to build.

• Added sql_string.cpp to VC++ and BCBuilder project files.

• Removed Totte Karlsson's 'populate' example, which never made it into the distribution anyway.

• Removed last vestiges of 'dummy.cpp'.

• Renamed *.cc to *.cpp in BCBuilder project files.

• Worked around a BCBuilder C++ syntax processing bug in row.h.

1.7.26, 2004.12.17

• Moved all of the SQLString definitions out of the header and into a new .cpp file, reformatted it all, and made the integer conversion functions use snprintf() or _snprintf() instead of sprintf(). Also, widened some of the buffers for 64-bit systems.

• Using quoted #include form for internal library headers, to avoid some problems with file name clashes. (The headers should still be installed in their own separate directory for best results, however.) Thanks to Chris Frey and Evan Wies for the patch and the discussion that lead to it.

• Removed unnecessary semicolons on namespace block closures. Thanks to Evan Wies for this patch.

• Fixed namespace handling in the legacy headers mysql++.hh and sqlplus.hh. Thanks to Chris Frey for this patch.

• #including iostream instead of ostream in lib/null.h for broader C++ compatibility. (This may allow MySQL++ to work on GCC 2.95.2 again, but this is unconfirmed.)

• Detecting proper mysql_shutdown() argument handling automatically in platform.h for the Windows compiler case instead of making the user edit the file. Thanks to Evan Wies for this patch.

• Fixed examples/Makefile.simple to use new *.cpp file naming.

• Fix to Gentoo ebuild file's exception configure switch handling. Thanks to Chris Frey for this patch.

• Rebuilding lib/custom*.h intelligently now, to avoid unnecessary recompiles after running bootstrap script.

1.7.25, 2004.12.09

• Yet more fixes to the --with-mysql-lib and --with-mysql-include flags.

• Added DLLEXPORT stuff to platform.h, hopefully so that someone can figure out how to make VC++ make a DLL version of MySQL++.

• Renamed *.cc to *.cpp.

• Made 'set -> myset' change in VC++ project files.

• Some style changes (mostly whitespace) in header files.

1.7.24, 2004.12.08

• Fixed the --with-mysql-lib and --with-mysql-include flags' behavior, and extended their search ability to handle one other common case. (Fixed by Steve Roberts)

• Fixes to put freestanding functions in namespace mysqlpp. (They weren't in the namespace, while all the class member functions were.) This required bumping the ABI version number to 4.

• Renamed set module to myset, to avoid conflicts with Standard C++ Library's set.h when MySQL++ headers were installed into one of the standard system include directories.

• Renamed all the idempotency guards to make them consistent in style and unique to MySQL++.

• Reformatted all of lib/*.cc.

1.7.23, 2004.11.20

• Query::reset() now empties the stored query string. If you subsequently stored a longer query in the object, you'd overwrite the previous query, but otherwise the longer part of the previous one would stick out past the new query.

• We now look to the NO_LONG_LONGS macro only to decide whether to fake 64-bit integer support using 32-bit integers.

• 64-bit integer support under Visual C++ may be working now, using that platform's __int64_t type. This has not been tested.

• Removed 64-bit integer support for Codewarrior on Mac OS 9 and earlier. OS X uses GCC, so it requires no special support.

• Added MinGW detection in platform.h.

• If you pass a flag (-X) to the examples that take the standard parameters (resetdb, simple1, etc.), it prints a usage message.

• Better error handling in resetdb example, where errors are the most critical. (If that one runs without errors, the others probably will, too, and you have to run that one first.)

• resetdb now reports success, rather than succeeding silently.

• Removed the code in sample1 example that duplicated util module's print_stock_table(), and called that function instead.

• Moved the preview() calls in the example programs to before the query execution calls, because execution modifies the query.

• All examples that take the standard command line parameters now exit when connect_to_db() fails in one of the ways that don't throw an exception, rather than bulling onward until the next MySQL database call fails because the connection isn't up.

• dbinfo example now takes the standard command line parameters.

• Much better output formatting in dbinfo example.

• Calling reset() where appropriate in the various example programs. Before, the programs may have worked, but not for the right reason. This lead some people to believe that calling reset() was not necessary.

• Fixed an incorrect use of row["string"] in complic1 example.

• Lots of code style improvements to the examples.

• Some VC++ type warnings squished. Some remain.

1.7.22, 2004.11.17

• Applied patches by Zahroof Mohammed to allow it to build under GCC 3.4.2. Tested on MinGW and Fedora Core 3 systems.

• Removed all the forward declarations in defs.h, and added forward declarations where necessary in individual header files.

  #including defs.h in fewer locations as a result.

• Legacy headers sqlplus.hh and mysql++.hh now declare they are using namespace mysqlpp, to allow old code to compile against the new library without changes.

• Removed query_reset parameter from several class Query member functions. In the implementation, these parameters were always overridden! No sense pretending that we pay attention to these parameters. This changes the ABI version to 3.

• #including custom.h in sql_query.h again...it's necessary on GCC 3.4.

• bootstrap script runs lib/config.pl after configure. This is just a nicety for those running in 'maintainer mode'.

1.7.21, 2004.11.05

• Generating a main mysql++ RPM containing just the library files and basic documentation, and the -devel package containing everything else.

• Devel package contains examples now, along with a new Makefile that uses the system include and library files, rather than the automake-based Makefile.am we currently have which uses the files in the mysql++ source directory.

• Renamed sqlplusint subdirectory in the package to lib.

• Removed the obsolete lib/README file.

• lib/sql_query.h no longer #includes custom.h, simplifying build-time dependencies and shortening compile times.

1.7.20, 2004.11.03

• Collapsed all numbered *.hh headers into a single *.h file. For example, the contents of row1.hh, row2.hh and row3.hh are now in row.h.

• While doing the previous change, broke several circular dependencies. (The numbered file scheme was probably partly done to avoid this problem.) The practical upshot of most of these changes is that some functions are no longer inline.

• Removed define_short.hh and everything associated with it. The library now uses the short names exclusively (e.g. Row instead of MysqlRow).

• Put all definitions into namespace mysqlpp. For most programs, simply adding a 'using namespace mysqlpp' near the top of the program will suffice to convert to this version.

• Once again, the main include file was renamed, this time to mysql++.h. Hopefully this is the last renaming!

• mysql++.hh still exists. It emits a compiler warning that the file is obsolete, then it #includes mysql++.h for you.

• sqlplus.hh is back, being a copy of the new mysql++.hh. Both of these files may go away at any time. They exist simply to help people transition to the new file naming scheme.

• Renamed mysql++-windows.hh to platform.h, and added code to it to handle #inclusion of config.h on autotools-based systems intelligently. This fixes the config.h error when building under Visual C++.

• There is now only one place where conditional inclusion of winsock.h happens: platform.h.

• Beautified the example programs.

1.7.19, 2004.10.25

• Fixed an infinite loop in the query mechanism resulting from the strstream change in the previous version. There is an overloaded set of str() member functions that weren't a problem when query objects were based on strstream.

• Query mechanism had a bunch of const-incorrectness: there were several function parameters and functions that were const for the convenience of other parts of the code, but within these functions the constness was const_cast away! This was evil and wrong; now there are fewer const promises, and only one is still quietly broken within the code. (It's in the SQLQuery copy ctor implementation; it should be harmless.)

• Removed operator=() in Query and SQLQuery classes. It cannot take a const argument for the same reason we have to cast away const in the SQLQuery copy ctor. It's tolerable to do this in the copy ctor, but intolerable in an operator. Since the copy ctor is good enough for all code within the library and within my own code, I'm removing the operator.

• Above changes required bumping the ABI to version 2.

• Visual C++ projects now look for MySQL build files in c:mysql, since that's the default install location. (Previously, it was c:program filesmysql.)

1.7.18, 2004.10.01

• Changed all the strstream (and friends) stuff to stringstream type classes. Let there be much rejoicing.

• Query object now lets you use store() even when the SQL query cannot return a result, such as a DROP TABLE command. This is useful for sending arbitrary SQL to the server. Thanks to Jose Mortensen for the patch.

• Quote fix in configure.in, thanks to David Sward.

• Renamed undef_short file to undef_short.hh.

• Gentoo ebuild file is actually being shipped with the tarball, instead of just sitting in my private CVS tree since 1.7.14 was current. Ooops....

1.7.17, 2004.09.16

• Reverted one of the VC++ warning fix changes from 1.7.16 that caused crashes on Linux.

• Added a configure test that conditionally adds the extra 'level' parameter to mysql_shutdown() that was added in MySQL 4.1.3 and 5.0.1.

1.7.16, 2004.09.13

• Building VC++ version with DLL version of C runtime libraries, and at warning level 3 with no warnings emitted.

• VC++ build no longer attempts to fake "long long" support. See the Wishlist for further thoughts on this.

1.7.15, 2004.09.02

• Renamed Configure file to common.am, to avoid file name conflict with configure script on case-sensitive file systems.

• Added ebuild file and ebuild target to top-level Makefile for Gentoo systems. Thanks to Chris Frey for this.

• Small efficiency improvements to BadQuery exception handling. Initial idea by Chris Frey, improvements by Warren Young.

1.7.14, 2004.08.26

• Builds with Visual C++ 7.1.

• Fixed a bug in custom macro generation that caused problems with GCC 3.4. (X_cus_value_list ctor definition was broken.)

1.7.13, 2004.08.23

• Removed USL CC support. (System V stock system compiler.) Use GCC on these platforms instead.

• Added examples/README, explaining how to use the examples, and what they all do.

• Most of the example programs now accept command line arguments for host name, user name and password, like resetdb does.

• Renamed sinisa_ex example to dbinfo.

• Several Standard C++ syntax fixes to quash errors emitted by GCC 3.4 and Borland C++ Builder 6. Thanks to Steffen Schumacher and Totte Karlsson for their testing and help with these.

• Added proper #includes for BCBuilder, plus project files for same. Thanks to Totte Karlsson for these.

1.7.12, 2004.08.19

• Many Standard C++ fixes, most from the GCC 3.4 patch by Rune Kleveland.

• Added Wishlist file to distribution.

• Fixed a problem in the bootstrap script that caused complaints from the autotools on some systems.

• RPM building is working properly now.

• Fixed the idempotency guard in datetime1.hh.

1.7.11, 2004.08.17

• Renamed mysql++, defs and define_short files, adding .hh to the end of each. (They're header files!) This shouldn't impact library users, since these are hopefully used internal to the library only.

• Removed sqlplus.hh file. Use mysql++.hh instead.

• Added mysql++.spec, extracted from contributed 1.7.9 source RPM, and updated it significantly. Also, added an 'rpm' target to Makefile.am to automate the process of building RPMs.

• Added bootstrap and LGPL files to distribution tarball.

• Added pre-1.7.10 history to this file.

• Removed .version file. Apparently it's something required by old versions of libtool.

1.7.10, 2004.08.16

Maintenance taken over by Warren Young. See the mailing list post for rationale.

• Applied many of the GCC 3.x patches submitted for 1.7.9 over the years. This allows it to build on everything from 3.0 to 3.3.3, at least. Because so many patches are rolled up in one big jump, it's difficult to describe all the changes and where they came from. Mostly they're Standard C++ fixes, as GCC has become more strict in the source code that it will accept.

• MysqlRow used to overload operator[] for string types as well as integers so you could look up a field by its name, rather than by its index. GCC 3.3 says this is illegal C++ due to ambiguities in resolving which overload should be used in various situations. operator[] is now overloaded only for one integer type, and a new member function lookup_by_name() was added to maintain the old by-field-name functionality.

• Fixed another operator overloading problem in SSQLS macro generation with GCC 3.3.

• The_table member of SSQLS-defined structures is now const char*, so you can assign to it from a const char* string.

• Got autoconf/automake build system working with current versions of those tools again. Removed the generated autotools files from CVS.

• Renamed library file from libsqlplus to libmysqlpp.

1.7.9, 2001.05.01

• Fixed a serious bug in Connection constructor when reading MySQL options

• Improved copy constructor and some other methods in Result / ResUse

• Many other minor improvements

• Produced a complete manual with chapter 5 included

• Updated documentation, including a Postscript format

1.7.8, 2000.11.14

• Introduced a new, standard way of dealing with C++ exceptions.

• MySQL++ now supports two different methods of tracing exceptions. One is by the fixed type (the old one) and one is standard C++ type by the usage of what() method. A choice of methods has to be done in building a library. If configure script is run with -enable-exception option , then new method will be used. If no option is provided, or -disable-exception is used, old MySQL++ exceptions will be enforced. This innovation is a contribution of Mr. Ben Johnson ben@blarg.net

• MySQL++ now automatically reads at connection all standard MySQL configuration files

• Fixed a bug in sql_query::parse to enable it to parse more then 99 char's

• Added an optional client flag in connect, which will enable usage of this option, e.g. for getting matched and not just affected rows. This change does not require any changes in existing programs.

• Fixed some smaller bugs

• Added better handling of NULL's. Programmers will get a NULL string in result set and should use is_null() method in ColData to check if value is NULL

• Further improved configuration

• Updated documentation, including a Postscript format

1.7.6, 2000.09.22

This release contains some C++ coherency improvements and scripts enhacements.

• result_id() is made available to programmers to fetch LAST_INSERT_ID() value

• Connection ctor ambiguity resolved, thanks to marc@mit.edu

• Improved cnnfigure for better finding out MySQL libraries and includes

• Updated documentation, including a Postscript format

1.7.5, 2000.07.30

This release has mainl bug fixes and code improvements:

• A bug in FieldNames::init has been fixed, enabling a bug free usage of this class with in what ever a mixture of cases that is required

• Changed behaviour of ResUse, Result and Row classes, so that they could be re-used as much as necessary, without any memory leaks, nor with any re-initializations necessary

• Fixed all potential leaks that could have been caused by usage of delete instead of delete[] after memory has been allocated with new[]

• Deleted all unused classes and macros. This led to a reduction of library size to one half of the original size. This has furthermore brought improvements in compilation speed

• Moved all string manipulation from system libraries to libmysqlclient, thus enabling uniformity of code and usage of 64 bit integers on all platforms, including Windows, without reverting to conditional compilation. This changes now requires usage of mysql 3.23 client libraries, as mandatory

• Changed examples to reflect above changes

• Configuration scripts have been largely changed and further changes shall appear in consecutive sub-releases. This changes have been done and shall be done by our MySQL developer Thimble Smith tim@mysql.com

• Changed README, TODO and text version of manual. Other versions of manual have not been updated

• Fixed .version ``bug''. This is only partially fixed and version remains 1.7.0 due to some problems in current versions of libtool. This shall be finally fixed in a near future.

• Several smaller fixes and improvements

• Added build.sh script to point to the correct procedure of building of this library. Edit it to add configure options of your choice.

1.7, 2000.05.17

This is mainly a release dealing with bug fixes, consistency improvements and easier configure on some platforms.

• A bug in fetch_row() method of ResUse class has been fixed. Beside changes that existed in a distributed patch, some additional error checking has been introduced.

• A bug in escape manipulator has been fixed that could cause an error if all characters had to be escaped.

• An inconsistency in column indexing has been fixed. Before this version, column names in row indexing with strings, i.e. row[] , has been case sensitive, which was inconsistent with MySQL server handling of column names.

• An inconsistency in conversion from strings to integers or floats has been fixed. In prior version a space found in data would cause a BadConversion exception. This has been fixed, but 100% consistency with MySQL server has not been targeted, so that other non-numeric characters in data will still cause BadConversion exception or error. As this API is used in applications, users should provide feedback if full compatibility with MySQL server is desired, in which case BadConversion exception or error would be abolished in some of future versions.

• A new method in ColData class has been introduced. is_null() method returns a boolean to denote if a column in a row is NULL.

• Finally, as of this release, testing for NULL values is possible.

• Those are columns with empty strings for which is_null() returns true.

• Some SPARC Solaris installations had C++ exception problems with g++ 2.95.2 This was a bug that was fixed in GNU gcc, as from release 2.95 19990728. This version was thoroughly tested and is fully functional on SPARC Solaris 2.6 with the above version of gcc.

• A 'virtual destructor ' warning for Result class has been fixed

• Several new functions for STL strings have been added. Those functions (see string_util.hh) add some of the functionality missing in existing STL libraries

• Conversion for 64 bit integers on FreeBSD systems has been added. On those systems FIX_FOR_BSD should be defined in CXXFLAGS prior to configuring. Complete conversion to the usage of functions for integer conversion found in mysqlclient library is planned for one of the next releases.

• A completely new, fully dynamic, dramatic and fully mutable result set has been designed and will be implemented in some of 2.x releases.

• Several smaller fixes and improvements, including defaulting exceptions to true, instead of false, as of this version.

• An up-to-date and complete Postscript version of documentation is included in this distribution.

• Large chunks of this manual are changed, as well as README and TODO files.

1.6, 2000.02.03

• This is a major release as it includes new features and major rewrites.

• Automatic quoting and escaping with streams. It works automatically , depending on the column type. It will work with << on all ostream derived types. it is paricularly handy with query objects and strstreams. Automatic quoting and escaping on cout, cerr and clog stream objects is intentionally left out, as quoting / escaping on those stream objects is not necessary. This feature can be turned of by setting global boolean dont_quote_auto to true.

• Made some major changes in code, so that now execute method should be used only with SSQL and template queries, while for all other query execution of UPDATE's, INSERT's, DELETE's, new method exec() should be used. It is also faster.

• New method get_string is inroduced for easier handling / casting ColData into C++ strings.

• Major rewrite of entire code, which led to it's reduction and speed improvement. This also led to removal of several source files.

• Handling of binary data is introduced. No application program changes are required. One of new example programs demonstrates handling of binary data.

• Three new example programs have been written and thoroughly tested. Their intention is to solve some problems addressed by MySQL users.

• Thorough changes is Makefile system has been made

• Better configuration scripts are written, thanks to D.Hawkins dhawkins@cdrgts.com

• Added several bug fixes

• Changed Manual and Changelog

1.5, 1999.12.01

• Fixed bug in template queries, introduced in 1.4 (!)

• Fixed connect bug

• Fixed several bug in type_info classes

• Added additional robustness in classes

• Added additional methods for SQL type info

• Changed Changelog and README

1.4, 1999.11.25

• Fixed bug in store and storein methods

• Fixed one serious memory leak

• Fixed a very serious bug generated by gcc 2.95.xx !!

• Added robustness in classes, so that e.g. same query and row objects can be re-used. Changed sinisa_ex example to reflect and demonstrate this ability.

• Changed Changelog and README

• Few other bug fixes and small improvements and speed-ups

1.3, 1999.11.10

• Fixed several erronous definitions

• Further changed source to be 2.95.2 compatible

• Expunged unused statements, especially dubious ones, like use of pointer_tracker

• Corrected bug in example file fieldinf1

• Finally fixed mysql_init in Connection constructor, which provided much greater stability!

• Added read and get options, so that clients, like mysqlgui can use it

• Changed Changelog and README

• Many other bug fixes.

1.2, 1999.10.15

First offical release by Sinisa Milivojevic after Kevin Atkinson made him the offical maintainer,

• Many manual fixes.

• Changed README and Changelog

• Changed source to be compilable by gcc 2.95.xx, tribute to Kevin Atkinson kevinatk@home.com

• Added methods in Connection class which are necessary for fullfilling administrative functions with MySQL

• Added many bug fixes in code pertaining to missing class initializers, as notified by Michael Rendell michael@cs.mun.ca

1.1, 1999.08.02

This release is the first one made by Sinisa Milivojevic.

• Added several bug fixes

• Fixed memory leak problems and variables overlapping problems.

• Added automake and autoconf support by loic@ceic.com

• Added Makefile for manual

• Added support for cygwin

• Added example sinisa_ex (let modesty prevail) which used to crash a lot when memory allocation, memory leak and overlap problems were present. Smooth running of this example proves that all those bugs are fixed

• Corrected bugs in sql_query.cc regarding delete versus delete[] and string length in manip.cc

• Changed manual

• Changed README

• Many other smaller things

1.0, 1999.06.09

All prior versions were released by Kevin Atkinson. This release was made by Monty Widenius

• Added patches from Orion Poplawski orion@bvt.com to support the UnixWare 7.0 compiler

.64.1.1a, 1998.09.27

• Fixed several bugs that caused my library to fail to compile with egcs 1.1. Hopefully it will still compile with egcs 1.0 however I have not been able to test it with egcs 1.0.

• Removed some problem causing debug output in sql++pretty.

.64.1a, 1998.08.01

• Added an (almost) full guide to using Template Queries.

• Fixed it so the SQLQuery will throw an exception when all the template parameters are not provided.

• Proofread and speedchecked the manual (it really needed it).

• Other minor document fixes.

.64.0.1a, 1998.07.31

• Reworked the Class Reference section a bit.

• Minor document fixes

• Added more examples for SSQLS.

• Changed the syntax of equal_list for SSQLS from equal_list(cchar *, Manip, cchar *) to (cchar *, cchar *, Manip).

• Added set methods to SSQLS. These new methods do the same thing as there corresponding constructors.

• Added methods for creating a mysql_type_info from a C++ type_info.

.64.a, 1998.07.24

• Changed the names of all the classes so they no longer have to have Mysql in the begging of it. However if this creates a problem you can define a macro to only use the old names instead.

• The Specialized SQL Structures (formally known as Custom Mysql Structures) changed from mysql_ to sql_.

• Added the option of using exceptions thoughout the API.

• ColData (formally known as MysqlStrings) will now throw an exception if there is a problem in the conversion.

• Added a null adapter.

• Added Mutable Result Sets

• Added a very basic runtime type identification for SQL types

• Changed the document format from POD to LYX .

• Am now using a modified version of Perceps to extract the class information directly from the code to make my life easier.

• Added an option of defining a macro to avoid using the automatic conversion with binary operators.

• Other small fixed I probully forgot to mentune.

.63.1.a

• Added Custom Mysql Structures.

• Fixed the Copy constructor of class Mysql

• Started adding code so that class Mysql lets it children now when it is leaving.

• Attempted to compile it into a library but still need help. As default it will compile as a regular program.

• Other small fixes.

.62.a, 1998.05.03

• Added Template Queries

• Created a separate SQLQuery object that is independent of an SQL connection.

• You no longer have to import the data for the test program as the program creates the database and tables it needs.

• Many small bug fixes.

.61.1.a 1998.04.28

• Cleaned up the example code in test.cc and included it in the manual.

• Added an interface layout plan to the manual.

• Added a reverse iterator.

• Fixed a bug with row.hh (It wasn't being included because of a typo).

.61.0.a

• Major interface changes. I warned you that the interface may change while it is in pre-alpha state and I wasn't kidding.

• Created a new and Separate Query Object. You can no longer execute queries from the Mysql object instead you have to create a query object with Mysql::query() and use it to execute queries.

• Added the comparison operators to MysqlDate, MysqlTime and MysqlDateTime. Fixed a few bugs in the MysqlDate... that effected the stream output and the conversion of them to strings.

• Reflected the MysqlDate... changes in the manual.

• Added a new MysqlSet object and a bunch of functions for working with mysql set strings.

.60.3a, 1998.04.24

• Changed strtoq and strtouq to strtoll and strtull for metter compatibility Minor Manual fix.

• Changed makefile to make it more compatible with Solaris (Thanks Chris H)

• Fixed bug in comparison functions so that they would compare in he right direction.

• Added some items to the to do list be sure to have a look.