into  neptunus.(none):/home/msvensson/mysql/mysql-4.1-maint

into  naruto.:C:/cpp/mysql-4.1-maint


sql/item_timefunc.cc:
  Auto merged

into  zippy.cornsilk.net:/home/cmiller/work/mysql/bug17583/my41-bug17583


client/mysql.cc:
  Auto merged

When the client program had its stdout file descriptor closed by the calling
shell, after some amount of work (enough to fill a socket buffer) the server 
would complain about a packet error and then disconnect the client.

This is a serious security problem.  If stdout is closed before the mysql is
exec()d, then the first socket() call allocates file number 1 to communicate
with the server.  Subsequent write()s to that file number (as when printing
results that come back from the database) go back to the server instead in 
the command channel.  So, one should be able to craft data which, upon being
selected back from the server to the client, and injected into the command
stream become valid MySQL protocol to do something nasty when sent /back/ to 
the server.

The solution is to close explicitly the file descriptor that we *printf() to, 
so that the libc layer and the OS layer both agree that the file is closed.


BitKeeper/etc/collapsed:
  BitKeeper file /home/cmiller/work/mysql/bug17583/my41-bug17583/BitKeeper/etc/collapsed
client/mysql.cc:
  If standard output is not open (specifically, if dup() of its file number 
  fails) then we explicitly close it so that future uses of the file descriptor
  behave correctly for a closed file.
mysql-test/r/mysql_client.result:
  Prove that the problem of writing SQL output to the command socket no longer
  exists.
mysql-test/t/mysql_client.test:
  Prove that the problem of writing SQL output to the command socket no longer
  exists.

strcmp -> strncmp

Improve sort algorithm for reorder, all test with smae name except for ending digit should be run after each other. 
Example of that is ndb_autodiscover[1-3]

Two implementations of "get_err_code_from_name", one use if mysqld_ername.h is available and oe if it's not
Declare "to_ptr" first in block 

mysql-test/lib/mtr_report.pl:
  Use tinfo's comment as primary place to look for error message

Fix problem with specifying vardir, mysql_version_id was not yet available when vardir arguments is checked
Move the code to look for exe_mysqld earlier => to initial_setup
Fix warnings detected by running with "diagnostics"
Remove unused option "opt_result_ext"
Init "path_ndb_examples_dir"


mysql-test/lib/mtr_cases.pl:
  Set default number of slave to 0
  Remove unused/uninitialized "$opt_result_ext"
mysql-test/lib/mtr_report.pl:
  Remove unused/uninitialized "$opt_result_ext"

into  neptunus.(none):/home/msvensson/mysql/mysql-4.1-maint

the "mysqld --version" command will print "/path/.libs/lt-mysqld  Ver x.x.x"


mysql-test/mysql-test-run.pl:
  Modify regex for parsing mysqld version as the mysqld is sometimes a libtool wrapper and
  the "mysqld --version" command will print "/path/.libs/lt-mysqld"

into  neptunus.(none):/home/msvensson/mysql/mysql-4.1-maint


mysql-test/r/subselect.result:
  Auto merged
mysql-test/t/ps.test:
  Auto merged
mysql-test/t/subselect.test:
  Auto merged

Improve 'run_testcase_need_slave_restart' to detect if a slave restart really is necessary. 
So far all rpl test requires a slave restart, but for all other tests it can be skipped
Improve the sort order used by --reorder


mysql-test/lib/mtr_cases.pl:
  Improve the sort order used by reorder
mysql-test/mysql-test-run.pl:
  Improve 'run_testcase_need_master_restart' to require restart if master is not already started
  Improve 'run_testcase_need_slave_restart' to detect if a slave restart really is necessary. 
  So far all rpl test requires a slave restart, but for all other tests it can be skipped

Move call f dunction into "initialize_servers" so it's called also
in stress and bench mode

Cleanup .progress, .reject, .log and .warnings files produced by mysqltest


client/mysqltest.c:
  Add printout of file in which warning was detected
mysql-test/include/ctype_like_escape.inc:
  Remove warnings, convert -- comments to # comments
mysql-test/mysql-test-run.pl:
  Cleanup all files produced by mysqltest before starting mysqltest again

mysql-test/mysql-test-run.pl:
  Add policy directive about keeping mysqltest framework tools identical in all versions
  Cleanup the initial comment to reflect current state

creating an environment variable is handled directly from the perl code.  

but having it on tmpfs gives a big speedup. 


mysql-test/mysql-test-run.pl:
  Make use of opt_mem and let 4.1 allow vardir to be set. Still relies on the var/ directory
  but having it on tmpfs gives a big speedup.

mysql-test/mysql-test-run.pl:
  Use same location for slave-load-tmpdir in all versions
mysql-test/mysql-test-run.sh:
  Use same location for slave-load-tmpdir in all versions
mysql-test/r/rpl_loaddata.result:
  Update result after changing slave-load-tmpdir to use a shorter path
mysql-test/r/rpl_loaddatalocal.result:
  Update result after changing slave-load-tmpdir to use a shorter path
mysql-test/r/rpl_log.result:
  Update result after changing slave-load-tmpdir to use a shorter path
mysql-test/t/rpl_loaddatalocal.test:
  Use MYSQLTEST_VARDIR when specifying path to load from(backport from 5.0)
  Use new command "remove_file" instead of s"ystem rm"

- When --record is passed to mysqltest the whole testcase should be executed
as it normally is while the output form the test is acumulating
in ds_res. When test has finished ds_res should simply be written to the
specified result file(if any) instead of comapring it against the result file.
Simplify handling of --require and also the cecking of result files by splitting
check_result function into one  function 'check_require' that is specialised in
checking require's and leave 'check_result' to do just that.
- "mysqltest --record" has been considered unsafe, but with this really simple
logic, it should be safe to use.


client/mysqltest.c:
  When --record is passed to mysqltest the whole testcase should be executed
  as it normally is while the output form the test is acumulating
  in ds_res. When test has finished ds_res should simply be written to the
  specified result file(if any) instead of comapring it against the result file.
  Simplify handling of --require and also the cecking of result files by splitting
  check_result function into one  function 'check_require' that is specialised in
  checking require's and leave 'check_result' to do just that.

from var/ to a tmpfs area and thereby speed up the execution of the testsuite
 significantly


mysql-test/mysql-test-run.pl:
  Add new option --mem to mysql-test-run.pl. It will automatically setup a symlink
  from var/ to a tmpfs area and thereby speed up the execution of the testsuite
  significantly

Some variable names in 5.1 is soo long that there is no space inbetween 
name and value. Outputing a warning about it, as that should probably be fixed.
At least one space should be put between the name and value.

Improve check of wheter test case need slave cluster

Set the environment variable NDB_STATUS_OK to indicate that cluster is installed ok. This is actually not needed for this script as it will
skip ndb tests if cluster install failed. But provided for compatibility


mysql-test/include/have_ndb.inc:
  Revert

I.e take advantage of the fact that a # comment is always terminated by a new line 
Add tests for the above


client/mysqltest.c:
  Improve "check_eol_junk" to detect junk although there are multi line comments in the way.
  I.e take advantage of the fact that a # comment is always terminated by a new line
mysql-test/r/mysqltest.result:
  Update resut file
mysql-test/t/mysqltest.test:
  Add test for improved check_eol_junk

Fix detection of debug binaries. If mysqld ouputs "debug xx" in the variable list it's a debug binary

 - add missing DBUG_RETURN
 - Change type of st_error->code to uint, that should be more than enough
   to hold the error numbers between 1000 and around 2000