Commit graph

1050 commits

Author SHA1 Message Date
Sergei Golubchik
8ddcd0cda8 post-review changes 1
include/my_time.h:
  remove duplicate defines.
  cast to ulonglong to avoid overflow
sql/field.cc:
  perform sign extension when reading packed TIME values
sql/item_cmpfunc.cc:
  when converting a string to a date for the purpose of comparing it with another date,
  we should ignore strict sql mode.
sql/item_timefunc.cc:
  better error message
sql/item_timefunc.h:
  limit decimals appropriately
sql/share/errmsg.txt:
  don't refer to an object as a "column" in error messages that are used not only for columns.
2011-05-19 19:01:46 +02:00
Sergey Glukhov
b7ed314fab 5.1 -> 5.5 merge 2011-05-18 10:59:04 +04:00
Sergey Glukhov
a48bff1d90 5.1 -> 5.5 merge 2011-05-18 10:59:04 +04:00
Sergey Glukhov
3201f92cb3 Bug#12403504 AFTER FIX FOR #11889186 : ASSERTION FAILED: DELSUM+(INT) Y/4-TEMP > 0
There are two problems:
1. There is a missing check for 'year' parameter(year can not be greater than 9999) in
   makedate function. fix: added check that year can not be greater than 9999.
2. There is a missing check for zero date in from_days() function.
   fix: added zero date check into Item_func_from_days::get_date()
   function.


mysql-test/r/func_time.result:
  test case
mysql-test/t/func_time.test:
  test case
sql/item_timefunc.cc:
  --added check that year can not be greater than 9999 for makedate() function
  --added zero date check into Item_func_from_days::get_date() function
2011-05-18 10:47:43 +04:00
Sergey Glukhov
a7cd008e12 Bug#12403504 AFTER FIX FOR #11889186 : ASSERTION FAILED: DELSUM+(INT) Y/4-TEMP > 0
There are two problems:
1. There is a missing check for 'year' parameter(year can not be greater than 9999) in
   makedate function. fix: added check that year can not be greater than 9999.
2. There is a missing check for zero date in from_days() function.
   fix: added zero date check into Item_func_from_days::get_date()
   function.
2011-05-18 10:47:43 +04:00
Kent Boortz
789aa8c485 Updated/added copyright headers 2011-07-04 01:25:49 +02:00
Kent Boortz
02e07e3b51 Updated/added copyright headers 2011-06-30 17:46:53 +02:00
Kent Boortz
0f03af653c Updated/added copyright headers 2011-07-04 01:25:49 +02:00
Kent Boortz
9da00ebec9 Updated/added copyright headers 2011-06-30 17:46:53 +02:00
Mattias Jonsson
10e3313366 merge 2011-06-13 11:57:47 +02:00
Michael Widenius
f09f1c7c7d Merge with dynamic column code 2011-05-12 14:30:34 +03:00
Georgi Kodinov
c4c808d606 weave merge of mysql-5.5->mysql-5.5-security 2011-05-10 17:20:26 +03:00
Georgi Kodinov
712f2d3833 weave merge of mysql-5.5->mysql-5.5-security 2011-05-10 17:20:26 +03:00
Michael Widenius
5ab92b1f85 Adding support for Dynamic columns (WL#34):
- COLUMN_CREATE(column_nr, value, [column_nr,value]...)
- COLUMN_ADD(blob,column_nr, value, column_nr,value]...)
- COLUMN_DELETE(blob, column_nr, column_nr...)
- COLUMN_EXISTS(blob, column_nr)
- COLUMN_LIST(blob, column_nr)
- COLUMN_GET(string, column_nr AS type)

Added cast(X as DOUBLE) and cast(x as INT)
Better warning and error messages for wrong cast's
Created some sub functions to simplify and reuse code.
Added a lot of conversation functions with error/warnings for what went wrong.
Fixed some issues when casting time to datetime.
Added functions to dynamic strings and Strings to allow one to move a string buffer from dynamic strings to String (to save malloc+ copy)
Added dynamic columns library to libmysqlclient


include/Makefile.am:
  Added ma_dyncol.h
include/decimal.h:
  Added 'const' to arguments for some functions.
include/my_sys.h:
  Added dynstr_reassociate()
include/my_time.h:
  Added TIME_SUBSECOND_RANGE
  Added double_to_datetime()
  Added flag argument to str_to_time()
libmysql/CMakeLists.txt:
  Added mysys/ma_dyncol.c
libmysql/Makefile.shared:
  Added ma_dyncol
libmysql/libmysql.c:
  Added argument to str_to_time()
mysql-test/r/bigint.result:
  Better error messages
mysql-test/r/cast.result:
  Better warning and error messages
  A lot of new cast() tests
mysql-test/r/func_math.result:
  Better warning messages
mysql-test/r/func_str.result:
  Better warning messages
mysql-test/r/func_time.result:
  Better warning messages
mysql-test/r/sp-vars.result:
  Better warning messages
mysql-test/r/strict.result:
  Better warning messages
  New test result
mysql-test/r/type_newdecimal.result:
  Better warning messages
mysql-test/r/warnings.result:
  Better warning messages
mysql-test/suite/funcs_1/r/innodb_func_view.result:
  Updated results after better cast warnings
mysql-test/suite/funcs_1/r/memory_func_view.result:
  Updated results after better cast warnings
mysql-test/suite/funcs_1/r/myisam_func_view.result:
  Updated results after better cast warnings
mysql-test/suite/optimizer_unfixed_bugs/t/bug43448.test:
  Added begin...commit to speed up test.
mysql-test/suite/parts/inc/part_supported_sql_funcs_delete.inc:
  Added begin...commit to speed up test.
mysql-test/suite/parts/inc/partition_supported_sql_funcs.inc:
  Added begin...commit to speed up test.
mysql-test/suite/parts/r/part_supported_sql_func_innodb.result:
  Added begin...commit to speed up test.
mysql-test/suite/parts/r/part_supported_sql_func_myisam.result:
  Added begin...commit to speed up test.
mysql-test/suite/parts/r/rpl_partition.result:
  Added begin...commit to speed up test.
mysql-test/suite/parts/t/part_supported_sql_func_innodb.test:
  Removed duplicated --big_test
mysql-test/suite/parts/t/rpl_partition.test:
  Added begin...commit to speed up test.
mysql-test/suite/pbxt/r/cast.result:
  Updated results after better cast warnings
mysql-test/suite/pbxt/r/func_str.result:
  Updated results after better cast warnings
mysql-test/suite/pbxt/r/type_newdecimal.result:
  Updated results after better cast warnings
mysql-test/suite/rpl/r/rpl_innodb_bug28430.result:
  Added begin...commit to speed up test.
mysql-test/suite/rpl/t/rpl_innodb_bug28430.test:
  Added begin...commit to speed up test.
mysql-test/suite/vcol/r/vcol_supported_sql_funcs_innodb.result:
  More warnings
mysql-test/suite/vcol/r/vcol_supported_sql_funcs_myisam.result:
  More warnings
mysql-test/t/cast.test:
  A lot of new cast() tests
mysql-test/t/strict.test:
  Added new test
mysys/CMakeLists.txt:
  Added ma_dyncol.c
mysys/Makefile.am:
  Added ma_dyncol.c
mysys/string.c:
  Added dynstr_reassociate() to move a buffer from dynamic_strings to some other allocator
sql-common/my_time.c:
  Added 'fuzzydate' flag to str_to_time()
  Added support for microseconds to my_time_to_str() and my_datetime_to_str()
  Reset second_parts in number_to_datetime()
  Added double_to_datetime()
sql/field.cc:
  Added double_to_longlong() and truncate_double() to simplify and reuse code
sql/field.h:
  New prototypes
sql/item.cc:
  Changed Item::get_date(MYSQL_TIME *ltime,uint fuzzydate) to be aware of type of argument.
  (Needed to make it microsecond safe and get better warnings).
  Updated call to str_to_time_with_warn()
sql/item.h:
  Added struct st_dyncall_create_def used by dynamic columns
  Added virtual bool dynamic_result() to tell if type of argument may change over calls.
sql/item_cmpfunc.cc:
  Added Item_func_dyncol_exists()
sql/item_cmpfunc.h:
  Added class Item_func_dyncol_exists
sql/item_create.cc:
  Added get_length_and_scale() to simplify other functions
  Simplified and extended create_func_cast()
  Added support for cast(X as double(X,Y))
  Added functions to create dynamic column functions.
sql/item_create.h:
  Added prototypes
sql/item_func.cc:
  Extended cast functions Item_func_signed() and Item_func_unsigned() to work with dynamic types
  Added Item_double_typecast()
sql/item_func.h:
  Added class Item_double_typecast()
sql/item_strfunc.cc:
  Added functions for COLUMN_CREATE(), COLUMN_ADD(), COLUMN_GET() and COLUMN_LIST()
sql/item_strfunc.h:
  Added classes for COLUMN_CREATE(), COLUMN_ADD(), COLUMN_GET() and COLUMN_LIST()
sql/item_timefunc.cc:
  Added flag argument to str_to_time_with_warn()
  Updated Item_char_typecast() to handle result type that may change between calls (for dynamic columns)
  Added Item_time_typecast::get_date() to ensure that we cast a datetime to time properly.
sql/item_timefunc.h:
  Added get_date() to Item_time_typecast() to allow proper results for casting time to datetime
sql/lex.h:
  Added new SQL function names
sql/my_decimal.cc:
  Added 'const' to some arguments.
  Better error message in case of errors (we now print out the wrong value)
  Added my_decimal2int()
sql/my_decimal.h:
  Moved some constants to my_decimal_limits.h
  Updated prototypes.
  Made my_decimal2int() a function as it's rather long (no reason to have it inline)
  Added decimal2my_decimal() function.
sql/mysql_priv.h:
  Prototypes for new functions
sql/share/errmsg.txt:
  New error messages for wrong casts and dynamic columns
sql/sql_acl.cc:
  Fixed indentation
sql/sql_base.cc:
  Added dynamic_column_error_message()
sql/sql_string.h:
  Added reassociate() to move a buffer to be owned by String object.
sql/sql_yacc.yy:
  Added syntax for COLUMN_ functions.
sql/time.cc:
  Updated str_to_datetime_with_warn() flag argument to same type as other functions
  Added conversion flag to str_to_time_with_warn() (Similar to all datetime functions)
  Added conversion functions with warnings: double_to_datetime_with_warn() and decimal_to_datetime_with_warn()
strings/decimal.c:
  Added 'const' to arguments for some functions.
unittest/mysys/Makefile.am:
  Added test for dynamic columns code
2011-05-08 13:24:06 +03:00
unknown
db4c6f56e2 Merge from mysql-5.5.12-release 2011-05-06 10:27:04 +02:00
bjorn.munch@oracle.com
f152d4cf05 Merge from mysql-5.5.12-release 2011-05-06 10:27:04 +02:00
Karen Langford
83f19ef457 Merge from mysql-5.1.57-release 2011-05-06 10:03:02 +02:00
Karen Langford
74afcca8f2 Merge from mysql-5.1.57-release 2011-05-06 10:03:02 +02:00
Michael Widenius
e415ba0fb2 Merge with MySQL 5.1.57/58
Moved some BSD string functions from Unireg
2011-05-02 20:58:45 +03:00
Tor Didriksen
ae154449f1 Bug#12340997 - DATE_ADD/DATE_SUB WITH INTERVAL CRASHES IN GET_INTERVAL_VALUE()
get_interval_value() was trying to parse the input string,
looking for leading '-' while skipping whitespace.
The macro my_isspace() does not work for utf16 character set,
since my_charset_utf16_general_ci.ctype == NULL.

Solution: convert input to ASCII before parsing.


mysql-test/r/ctype_utf16.result:
  New test case.
mysql-test/t/ctype_utf16.test:
  New test case.
sql/item_timefunc.cc:
  Use val_string_ascii() rather than val_string()
  so that we can safely use my_isspace() for skipping whitespace.
2011-04-28 09:03:56 +02:00
Tor Didriksen
1019b95877 Bug#12340997 - DATE_ADD/DATE_SUB WITH INTERVAL CRASHES IN GET_INTERVAL_VALUE()
get_interval_value() was trying to parse the input string,
looking for leading '-' while skipping whitespace.
The macro my_isspace() does not work for utf16 character set,
since my_charset_utf16_general_ci.ctype == NULL.

Solution: convert input to ASCII before parsing.
2011-04-28 09:03:56 +02:00
Sergey Glukhov
a60c39a2ff Bug#11889186 60503: CRASH IN MAKE_DATE_TIME WITH DATE_FORMAT / STR_TO_DATE COMBINATION
calc_daynr() function returns negative result
if malformed date with zero year and month is used.
Attempt to calculate week day on negative value
leads to crash. The fix is return NULL for
'W', 'a', 'w' specifiers if zero year and month is used.
Additional fix for calc_daynr():
--added assertion that result can not be negative
--return 0 if zero year and month is used


mysql-test/r/func_time.result:
  test case
mysql-test/t/func_time.test:
  test case
sql-common/my_time.c:
  --added assertion that result can not be negative
  --return 0 if zero year and month is used
sql/item_timefunc.cc:
  eturn NULL for 'W', 'a', 'w' specifiers
  if zero year and month is used.
2011-04-27 11:35:57 +04:00
Sergey Glukhov
735bab72c5 Bug#11889186 60503: CRASH IN MAKE_DATE_TIME WITH DATE_FORMAT / STR_TO_DATE COMBINATION
calc_daynr() function returns negative result
if malformed date with zero year and month is used.
Attempt to calculate week day on negative value
leads to crash. The fix is return NULL for
'W', 'a', 'w' specifiers if zero year and month is used.
Additional fix for calc_daynr():
--added assertion that result can not be negative
--return 0 if zero year and month is used
2011-04-27 11:35:57 +04:00
Georgi Kodinov
efe4af973b merge 5.5->5.5-security 2011-04-07 12:59:51 +03:00
Georgi Kodinov
9d5aac9463 merge 5.5->5.5-security 2011-04-07 12:59:51 +03:00
Georgi Kodinov
2db44e8bb4 merged mysql-5.1->mysql-5.1-security 2011-04-07 12:24:10 +03:00
Georgi Kodinov
8fbd9e4ca1 merged mysql-5.1->mysql-5.1-security 2011-04-07 12:24:10 +03:00
Sergey Glukhov
197bef3876 5.1 -> 5.5 merge 2011-04-27 11:46:23 +04:00
Sergey Glukhov
75f02c36ce 5.1 -> 5.5 merge 2011-04-27 11:46:23 +04:00
Sergey Glukhov
8cca70e036 5.1 -> 5.5 merge 2011-03-28 17:33:35 +04:00
Sergey Glukhov
2bcaf27144 5.1 -> 5.5 merge 2011-03-28 17:33:35 +04:00
Sergey Glukhov
d6125b27b3 Bug#11765216 58154: UNINITIALIZED VARIABLE FORMAT IN STR_TO_DATE FUNCTION
Valgrind warning happens due to uninitialized cached_format_type field
which is used later in Item_func_str_to_date::val_str method.
The fix is to init cached_format_type field.


mysql-test/r/func_time.result:
  test case
mysql-test/t/func_time.test:
  test case
sql/item_timefunc.cc:
  init cached_format_type field
2011-03-28 17:24:25 +04:00
Sergey Glukhov
9893d60e4f Bug#11765216 58154: UNINITIALIZED VARIABLE FORMAT IN STR_TO_DATE FUNCTION
Valgrind warning happens due to uninitialized cached_format_type field
which is used later in Item_func_str_to_date::val_str method.
The fix is to init cached_format_type field.
2011-03-28 17:24:25 +04:00
Sergey Glukhov
b1b17093c8 5.1->5.5 merge 2011-03-28 12:03:53 +04:00
Sergey Glukhov
315ce7bf4f 5.1->5.5 merge 2011-03-28 12:03:53 +04:00
Sergey Glukhov
d499851be0 Bug#11766112 59151:UNINITIALIZED VALUES IN EXTRACT_DATE_TIME WITH STR_TO_DATE(SPACE(..) ...
Valgrind warining happens due to missing
'end of the string' check. The fix is to
check if we reached the end of the string.


mysql-test/r/func_time.result:
  test case
mysql-test/t/func_time.test:
  test case
sql/item_timefunc.cc:
  check if we reached the end of
  the string after leading spaces skipping.
2011-03-28 11:53:18 +04:00
Sergey Glukhov
e9935c6363 Bug#11766112 59151:UNINITIALIZED VALUES IN EXTRACT_DATE_TIME WITH STR_TO_DATE(SPACE(..) ...
Valgrind warining happens due to missing
'end of the string' check. The fix is to
check if we reached the end of the string.
2011-03-28 11:53:18 +04:00
Sergei Golubchik
e016a2f5f0 lp:705210 Compiling with BUILD/compile-pentium64-debug fails 2011-03-26 11:59:34 +01:00
Vladislav Vaintroub
8250ceced6 Fix compilation on Windows:
- Fixes for type-conversion
  (time_t   is not interchangeable  with  my_time_t on Windows as time_t s 64 bit while my_time_t is long)
- BIGENDIAN-> ARCH_BIGENDIAN . 
  BIGENDIAN constant  is defined in winsock2.h (as 0)
- added explicit cast for longlong->double conversion in sql/item.h  (fixed many warnings) 

Also, HAVE_SNPRINTF is now defined and snprintf is defined to _snprintf in config-win.h
2011-03-24 12:30:03 +01:00
Sergei Golubchik
5ff5c4b8a3 lp:737111 Different behavior for TIMESTAMPADD with 0000-00-00 argument in 5.1-micro
respect fuzzydate flags in Item_*::get_date() methods
2011-03-18 13:43:33 +01:00
Sergei Golubchik
1cda265457 * fix for ALTER TABLE ... MODIFY timestamp->timestamp.
Use dedicated  do_field_temporal() for Copy_field.
* check_time_range() needs to know TIME's precision to use the
  correct max value.
2011-03-17 14:13:03 +01:00
Mattias Jonsson
d53906e387 Bug#11746819:
Bug#28928: UNIX_TIMESTAMP() should be considered unary monotonic by partition pruning

Made UNIX_TIMESTAMP MONOTONIC_INCREASING when it have TIMESTAMP argument (only).
2011-03-16 11:59:01 +01:00
Mattias Jonsson
11ad61f8c4 Bug#11746819:
Bug#28928: UNIX_TIMESTAMP() should be considered unary monotonic by partition pruning

Made UNIX_TIMESTAMP MONOTONIC_INCREASING when it have TIMESTAMP argument (only).
2011-03-16 11:59:01 +01:00
Georgi Kodinov
ace88ca2dd auto-merge 2011-03-15 16:56:11 +02:00
Georgi Kodinov
bcbce343dd auto-merge 2011-03-15 16:56:11 +02:00
Georgi Kodinov
7a37a7c0c8 Bug #11765023: 57934: DOS POSSIBLE SINCE BINARY CASTING DOESN'T
ADHERE TO MAX_ALLOWED_PACKET

Added a check for max_packet_length in CONVERT(, BINARY|CHAR).
Added a test case.
2011-03-15 13:19:30 +02:00
Georgi Kodinov
405f7ca69a Bug #11765023: 57934: DOS POSSIBLE SINCE BINARY CASTING DOESN'T
ADHERE TO MAX_ALLOWED_PACKET

Added a check for max_packet_length in CONVERT(, BINARY|CHAR).
Added a test case.
2011-03-15 13:19:30 +02:00
Sergei Golubchik
3c6ff364ca lp:731815 Crash/valgrind warning Item::send with 5.1-micro
Two problems:
* Item_func_convert_tz() did not tell args[0] that it needs
  TIME_NO_ZERO_IN_DATE result
* Item_func_timediff did not respect fuzzy_date limitations,
  truncated seconds when casting to signed long, resulting in
  invalid time value (hours, seconds = (uint)-1).
2011-03-09 11:59:47 +01:00
Sergei Golubchik
8b7fd8f577 lp:731124 Loss of precision on DISTINCT
many changes:
* NOT_FIXED_DEC now create hires fields, not old ones.
  As a result, temp tables preserve microseconds (on DISTINCT, GROUP BY)
* I_S tables force decimals=0 on temporal types (backward compatibility)
* Item_func_coalesce calculates decimals for temporal types
* no precision for TIME/DATETIME in CAST means 0, not NOT_FIXED_DEC
* addtime/timediff calculate decimals from arguments (not NOT_FIXED_DEC)

sql/field.h:
  NOT_FIXED_DEC now create hires fields, not old ones
sql/item.h:
  force decimals=0 for I_S tables
sql/item_cmpfunc.cc:
  Item_func_coalesce calculates decimals for temporal types
sql/item_create.cc:
  no precision for TIME/DATETIME in CAST means 0, not NOT_FIXED_DEC
sql/item_timefunc.cc:
  addtime calculates decimals from arguments (not NOT_FIXED_DEC)
sql/item_timefunc.h:
  timediff calculates decimals from arguments (not NOT_FIXED_DEC)
2011-03-08 19:41:58 +01:00
Sergei Golubchik
30e5b4d719 lp:730637 - Valgrind warnings in 5.1-micro
sql/field.cc:
  initialize ltime completely
sql/filesort.cc:
  don't pack MYSQL_TIME if it's not initialized
  (safe here, but valgrind complains)
sql/item_cmpfunc.cc:
  don't pack MYSQL_TIME if it's not initialized
  (safe here, but valgrind complains)
sql/item_timefunc.cc:
  don't check MYSQL_TIME members if it's uninitialized
  (safe here, but valgrind complains)
sql/time.cc:
  use c_ptr_safe() instead of c_ptr()
  (make valgrind happy)
2011-03-07 21:57:17 +01:00