home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
SGI Freeware 1998 November
/
Freeware November 1998.img
/
dist
/
fw_msql.idb
/
usr
/
freeware
/
msql
/
doc
/
HISTORY.msql.z
/
HISTORY.msql
Wrap
Text File
|
1998-10-27
|
17KB
|
460 lines
7 Oct 1997 2.0.3
-------------------------
mSQL
====
o Fixed cpp problem on build (compilation failed on some systems)
o Fixed infinite loop problem when updating an index value using
the index as the condition
o Fixed core dump when using a sub condition in conjunction with
and index for row selection
o Replaced hard coded make with $(MAKE) in makegen macros
Lite / W3-mSQL
==============
o Fixed several queries in msql-auth that did not include the
namespace to which they applied.
22 Sep 1997 2.0.2
-------------------------
mSQL
====
o Rolled in Bob Sum's patches
o Fixed translation of missing numerci values to NULL in msqlimport
o Fixed floating point rounding problems with money values
o Fixed denial of service attack problem with malformed protocols
o Fixed core dump if overflow buffers of a TEXT field extend
beyond the mapped region during an update
o Fixed core dump when query string is longer than the debug buffer
o Fixed SNI security concerns regarding buffer overflows
o Added forward name lookup testing of hostnames during ACL checks
o Fixed handling of reals, money and time in DISTINCT queries
o Fixed server memory leak in certain mmap errors
o Fixed memory leak in API library for empty selects
o Fixed problems with updating and index where the index is in
use by the update query (i.e. in the where clause)
o Changed the memory syncher to force a remap rather than using
msync as Linux appears to not sync the mapped regions.
o Fixed an mmap region leak
o Fixed problem where connection could be reached due to errors
causing session to abort
o Fixed problem with TCP Port number in bcmp during host_lookup
(when host_lookup = false). Thanks for the note Dirk.
# Code branch for experimental development
Jun 1997 2.0.1
-------------------------
mSQL
====
o Modified select loop range handling to avoid potential
infinite loop searching for active sockets
o Forced a mode of 0777 on the UNIX socket just in case
o Rolled in completed Win32 portability
o Fixed comparison of negative time/date values
o Fixed matching of null values to 0
o Fixed handling of null values in conditions other than
equals and not-equals
o Fixed value overwriting when updating the value of a TEXT
field that was originally set to NULL
o Fixed integrity of the overflow buffer on deletion of table
Rows containing NULL text fields.
o Fixed non-unique index error when updating a row using the
index for row selection
o Fixed wierd "field cannot be null" errors when inserting valid
data into not-null fields.
o Fixed CHAR and TEXT comparison problem where the in-table
value fills the entire field width and the literal value
is longer than the field width (sub string match problem).
o Fixed generation of values for _sysdate and _systime
o Fixed handling of system variable like _systime for queries
on empty tables
o Fixed problem in msqlDateOffset()
o Rolled in Bob Sum's *LIKE fixes. Some of the NULL handling
code replaces the above mentioned NULL fixes.
Lite / W3-mSQL
==============
o Fixed definitions of HAVE_DYNAMIC
o Added performance optimisations
o Fixed strseg() handling of strings shorter than the specified
substring.
Misc
====
o Updated doco for 2.0.1
o RTF version of documentation replaced by HTML. The
translation is still very rough and the PostScript version
should be used where possible.
o Added testing for sys/timeb.h to the configure script
8 Jun 1997 2.0 Release
-------------------------
mSQL
====
o Fixed minor read overflow in AVL tree initialisation
o Further optimised index selection during candidate setup
o Fixed memory leak in sub-condition handling
o Fixed bug in handling of reversed ident conds in inner loop
condition when using an index for the inner loop of a join
o Fixed unique index handling problem
o Fixed bug in comparison with some system variables
o Fixed pre/post increment problem of sequences during query
execution
o Fixed handling of TIME and DATE fields in msqldump
o Fixed core dump where and index fields is compared to an IDENT
values who's value is NULL during a join.
o Fixed core dump in certain situations if the value passed to
an index based condition is longer than the field length.
o Fixed bug in the sorting of real values
o Fixed obscure bug in resetting an index value to the same value
if that index was used for the lookup and the index node
contained duplicate values
o Optimised index updates so that indices are only modified if
the value has actually changed
o Made cache invalidation consistent
o Added "copy" and "move" to msqladmin
o Added section based name spaces to the config routines
o Added further configuration output to "msqladmin version"
o Rewrote msqlimport from scratch
o Added date / time format routines to the API library
Lite / W3-mSQL
==============
o Added .shtml as a valid HTML suffix for W3-mSQL processed files.
o Added @variable for accessing global symbols from a local
namespace (e.g. using globals from within a function)
o Fixed core dump when adding an empty function to a lite library
o Fixed double contraction of \\ escapes
8 May 1997 2.0 Beta 7
----------------------
mSQL
====
o Added test for "make target" before performing a build
o Fixed unique index handling if updating the index value with
itself
o Fixed access control for UPDATE queries
o Fixed core dump if _rowid > numRows given in a select
o Fixed problem with incorrect precision in update of real fields
o Forced debug initialisation messages to stderr
Lite / W3-mSQL
==============
o Fixed handling of multi-value web form data (i.e. multi-select
lists) during the casting of the values to an array.
o Upgraded limits in W3-mSQL passed variable lengths
o Fixed initialisation of symbol table prior to checking W3-Auth
based access control
11 April 1997 2.0 Beta 6
-------------------------
mSQL
====
o fixed rand() srand() portability macros
o added strptime check to autoconf script
o added strptime function to lib mSQL if needed
o completed Win32 porting of client API and client apps
o completed Win32 porting of server although indexing still
needs some debugging.
o fixed per field cache pointer during tmp table creation (core
dump when ordering the result of a join with a TEXT field)
o fixed handling of 8 bit char sets in regexp matching
o fixed ident handling in joins (particularly in self joins)
when multiple idents used in the comparison clause
o fixed timestamp bug where timestamps were reset in a condition
o fixed NOT NULL testing for NULL fields
o fixed NOT NULL testing for NULL indices
o fixed resetting overflow table mapping length to 0 when
dropping a table cache entry in loadTableDef()
o fixed potential leak of mapped region in initTable()
o updated client library to use MSQL_DEBUG not MINERVA_DEBUG
o updated libmsql to use stderr for debugging
o added short-circuits for condition testing (i.e. bail as soon
as we can tell the result of and AND or an OR etc)
o fixed leak of overflow buffers when deleting rows containing
TEXT fields
o added soundex based matching using the SLIKE operator
o fixed handling on blank date fields (ie. '')
o fixed LIMIT handling
Lite / W3-mSQL
==============
o More performance upgrades
- Symbol names no-longer malloced
- Symbol values < 10 bytes long no longer malloced
- Results show 300% performance increase in virtual
machine operating speed
o Fixed string length mis-match between locally created and
passed variable values (e.g. html form data etc)
o Forced flushing of output after an echo() as a call to
system() following a non-flushed echo() could cause the
output order to skew
20 March 1997 2.0 Beta 5
--------------------------
mSQL
====
o Rewrote regression test suite to obtain better code coverage
o Added real value precision handling to solve real number
rounding errors. The desired precision is now stored with
the value in the database table.
o Added code to translate system variables (eg _rowid) into
"real" table fields during creation of a result table. This
allows the sysvar value from the original table to be carried
through a DISTINCT or an ORDER rather than it being incorrectly
generated from the result table.
o Added ability to order a result on the value of a system
variable
o Added more system diagnostics for low-level errors
o Added more file ownership testing (for those that chown the
data file and don't get it correct)
o Added /dev/tty test in makegen macros. I though expecting
/dev/tty to exists was fair but apparantly not so on IRIX
5.3.
o Added README.sco for build details on SCO
o Added msqlGetSequenceInfo() to API and server
o Added support for showing sequence structure to relshow
o Added support for dumping current sequence status to msqldump
o Added support for UINT data type
o Added support for DATE data type
o Added support for TIME data type
o Added support for MONEY data type
o Added mmap region syncing for data files, overflow files and
the mapped regions of the AVL index files. Frequency of sync's
is configured in msql.conf.
o Added support for sequences in relshow (sequence creation at
the current sequence value)
o Added msqlGetSequenceInfo() API function
o Added _systime system variable
o Added mSQL time and date functions to the API library. These
allow for conversion from the text representations used by
mSQL into a valid UNIX time_t value.
o _sysdate system variable now returns a date type value of the
current data on the server machine (not the UNIX time value
as previously returned)
o Rows are now padded to an 8 byte boundary for memory alignment
reasons (mmap() access to the row headers structures on Alpha)
o Row header timestamp now time_t rather than int for machines
that use long time values (e.g. the Alpha)
o Fixed memory leak in yyprev symbol handling
o Fixed wild free in lexer symbol handling (core dump)
o Fixed timestamp overwrite problem for the Alpha
o Fixed relshow core dump for invalid index name
o Fixed core dump in AVL code when accessing an unaligned real
(i.e. a double) value in the tree node
o Fixed bad handling of numeric index types by moving the field
value auto-casting earlier in the control flow
o Fixed AVL Lookup problem for indices comprised of real values
o Fixed compound index handling bug (key could be incorrectly
formatted on lookup resulting in no results).
o Fixed problem with ident comparisons on single table selects
o Fixed problem with ident to sysvar comparisons on single table
selects (e.g. select * from foo where f1 = _rowid)
o Fixed wild free problem for ident values in join code (core dump)
o Fixed memory over-read on some TEXT field operations
o Fixed data hole free list handling
o Fixed multiple increment of sequence value (i.e. select _seq
from foo would increment the seq for each row)
o Fixed multiple increment of seq in conditions (i.e.
"select _seq from foo where id > _seq" would increment
the sequence twice (once for each call to getSysVar() )
o Fixed handling of system variables in join results
o Fixed 8 byte alignment problems in the AVL tree code. The
entire server is now 64bit clean and passes the
new regression test on an Alpha under OSF/1
Lite / W3-mSQL
==============
o Rewrote all symbol table handling code to use a 64 way
hash table for symbol lookups. Also added a short-cut for
symbol searches. General performance increase in every aspect
of the language results from this modification.
o Added support for non-local authentication databases via the
msql.conf file.
o Added support for HTML comments and other non-Lite tags that
start with <!
o Adding hooks for the mSQL time and date routines to mod_std
o Added getTzOffset() to mod_std for hacking time values
o Added toksplit() to mod_std
o Added literal symbol re-use code
o Fixed auto-casting problems with the new uint type
o Fixed auto-base-conversion (i.e. octal & hex). Conversions
now only performed on literals, not on variables during a cast
or an auto-cast
o Fixed off-by-one error in count operator for text strings
o Fixed regression test results to reflect the current test
output
o Fixed space handling in urlEncode()
o Fixed problem with string truncation in sub()
o Fixed problem with handling some C styled comments
o Fixed core dump when trying to cast an array to a scalar type
o Fixed off by one error in array literal element length
o Fixed handling of array indexing in text expansion (e.g.
in echo() strings) when array indexed by a variable value
o Fixed auto casting for int and uint in printf formats
Misc
====
o Fixed a couple of bugs in the time library for date/time handling
o Added errno.h to the headers searched in setup when looking for
sys_errlist (for NetBSD 1.2)
o Added stdlib.h to the headers searched in setup when looking for
ssize_t (for Irix 6.2)
o Started total re-working of the documentation kit.
o Code has been "linted" within and inch of its life (using
gcc -Wall)
1 February 1997 2.0 Beta 4
-----------------------------------
Lite / W3-mSQL
==============
o Added support for Multi-Select form data to W3-mSQL
o Added typeof() function to Lite standard module
o Fixed bug in expandText() handling mis-defined symbols
o Added work-around for the gnu getopt() library
o Added work-around for the gnu atoi() function
o Added 'uint' type handling and "%u" to printf()
o Fixed problem with fuction call arg count on some machines
o Fixed last character match problem in tr()
o Fixed a few doco bugs in the mSQL module doc's
o Fixed up type checking during a cast
o Fixed a problem if the first element of an array is not a CHAR val
o Fixed urlEncode problem for the + character
mSQL
====
o Fixed token destruction in error reporting from the parser
o Fixed documentation example for sequence creation
o Fixed core dump in mSQL monitor on overly large error results
o Fixed inaccurate initialisation of ident segments in msqlCreateIdent
o Added support for MSQL_CONF_FILE environment variable
o Fixed looping problem in wild-card delete
o Fixed some _more_ doco bugs
o Fixed problem with comparison of TEXT fields with NULL
o Fixed problem with the matchVarChar() function when comparing
values with TEXT fields
o Fixed bug in LIKE and CLIKE for matching multiple % wildcards
to an empty string
o Fixed a possible problem with the static size of the connection
array
14 January 1997 2.0 Beta 3
-----------------------------------
o Fixed bugs in table join logic
o Fixed bug that allowed an update to set a non-unique value
for a unique index
o Completed dynamic loading of module object code in Lite / W3-mSQL
o Fixed some documentation errors in the HTML manual
o Fixed non-flushed output in user prompts from msqladmin and msql
o Fixed problem with fields being inserted with the NULL flag set
o Added a 0 exit val to msqldump
o Fixed some API function prototypes
o Added strdup() to the malloc test code (although we don't
use the malloc library anymore for in-house code testing)
o Fixed memory leaks in parser, msqlServerDelete, QUIT processing,
join condition lists, and join ident fields.
o Wrote the mod_graph dynamic loaded graphing module for Lite &
W3-mSQL.
o Added optional W3-auth security. The file .w3-auth must now
be created in the directory containing the requested file to
enable W3-auth access control.
6 January 1997 2.0 Beta 2
-----------------------------------
o Fixed problem with deletion using an index for row selection
(reshuffle of dup values in tree node caused getNext to skip
a valid entry)
o Fixed bug in bulk index creation where it was including
non-active rows in the newly created index.
o Added strncasecmp() for SVR4 people (so they can dodge the BSD lib)
o Fixed bug in the AVL node swapping code wrt dup node lists.
o Fixed core dump bug when DISTINCT used with an empty result
table.
o Fixed bug in the updating of CHAR fields.
o Increased listen queue for UNIX socket to 128
o Fixed bug in updateValues() where by a full field width char
value would trash the active flag of the next field (i.e. NULL
termination of the string).
o Fixed param count test in Lite function calling.
o Increased Lite VM performance by up to 20% by caching stack
frames and other reuseable items.
o mSQL Server performance increases
- Removed some more malloc & bzero overhead
- Added speedup for boxes without hardware multiply (like
lots of the early SPARCs).
- The above modifications produced a 20% faster "killer"
result on a SPARC without hardware math.
o Added support for symbolic Macros in modules
o Started support for dynamic loading shared modules
16 December 1996 2.0 Beta 1
-----------------------------------
o Initial public beta release of the 2.0 code base.