Sorry. I do not think any of this is affected, since the parser by design is
not accessible to the client via the official channels -- this is the idea
of the DBIx::MyParse module is to add such functionality.

Thanks again, any help in finding the required number of reviewers to push
the patch through will be greatly appreciated.

Philip Stoev

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

On Thu, 18 Jan 2007, Philip Stoev wrote:

Note that I was referring to the A*B*I - does it result in any change of
binary structures that may be available via, e.g., the client library?

In any case, the patch needs a reviewer on our side to decide how to proceed
with it. I will ask around of somebody can take a look. Thanks!

Thank you for the support.

The patch is quite short and does not result in a change in the API, that
is, in the public methods exported by the Item_* classes.

Philip Stoev

----- Original Message -----
From: "Lenz Grimmer" mysql.com>
To: "Philip Stoev" stoev.org>
Cc: lists.mysql.com>
Sent: Thursday, January 18, 2007 2:10 PM
Subject: Re: Some Item classes are too private

It

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Philip,

On Sat, 13 Jan 2007, Philip Stoev wrote:

Thanks for the reminder! I unfortunately can't comment on the technical
implications of your proposed patch, but would you change result in a
change of the ABI?

--0-497191409-1169092522=:87709
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

Hi,

After having compiled (debug and optimized version) of MySQL 5.1.14 on Itanium-2 using icc, I have run
mysql_install_db --user=...........
and started mysqld_safe in both scenarios (debug and optimized). In both cases mysqld starts without any errors.

Unfortunaly when running the optimized version of mysqld I am not able to connect and set the root password.

./bin/mysqladmin -u root password '***********'
./bin/mysqladmin: connect to server at 'localhost' failed
error: 'Access denied for user 'root'@'localhost' (using password: NO)'

The same procedure applied on the mysqld compiled with debug version (--with-debug)
has no problem and I am able to set up the root password without any problems.

Is there a bug in the MySQL 5.1.14 version when not compiled with --with-debug or am I missing something ?

P.S. I have also tried to set the root password using the debug version, shut-down the mysqld debug, replaced the "mysql" subdirectory in the data subdir in the optimized version and restarted the optimized mysqld with the new mysql subdir in the data subdir, but I got the same error.

Could you please help ?

Thanks in advance for your help,

Mariella

Joerg Bruehe mysql.com> wrote:

Hello,

Can I please ask that someone applies the patch attached to bug #25612. It
allows one to walk the parse tree and obtain enough information from some
currently restricted Item objects in order to reconstruct the original SQL
statement without having to call print().

Once I have that, I will be able to release a MySQL parser, a server-side
MySQL protocol implementation and a query rewriting engine in Perl, which I
hope will allow more rapid query experimentation vs. having to modify the
MySQL source directly.

Thank you in advance.

Philip Stoev

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Don,

Happy New Year to you!

On Wed, 10 Jan 2007, Don Cohen wrote:

Thanks for the suggestion! Mark, could you take a look at this and
comment?

We are, sorry about that. Jay and Mark, can we please get the ball rolling
on this one and Don's other contributions?

This is a note about the feature submission I sent on 26 Oct 2006.
When I try to compile it for java version 1.4 I find that it uses
a feature introduced in 1.5. The best way I see so far to make it
compile (and work) in 1.4 is to make the two changes below.
If anyone would like to offer a better solution I'd be happy to see
it.
BTW, I've not heard anything back about either this feature submission
or the bug fix I submitted in December. I begin to wonder whether
anyone is listening.

/src/com/mysql/jdbc/ExportControlled.java
My previous diff listed line 68 as
TrustingSSLSocketFactory.getDefault();
which I now change to
TrustingSSLSocketFactory.getDefault2();

and /src/com/mysql/jdbc/TrustingSSLSocketFactory.java line 39
which was previously listed as
public static javax.net.ssl.SSLSocketFactory getDefault() {
I now change to
public static javax.net.ssl.SSLSocketFactory getDefault2() {

--0-1738025919-1168365282=:53943
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

Hi Giuseppe,

Thanks a lot for your suggestions.
The system has already the autoconf, aclocal, ...)
I have tried to use the BUILD/compile-ia64-debug-max (the system architecture is IA-64 Itanium-2) but I still get the same set of errors.
The version of icc is 9.0.
Do you know or does anybody know whether the issues could be related to the version of the icc ? ( e.g. http://dev.mysql.com/doc/refman/5.1/en/porting.html, even though I am trying to compile 5.0.27 I get the same set of errors that I get when trying to compile 5.1.14).

Could you please help ?

Thanks in advance for your help,

Mariella

Mariella Petrini yahoo.com> wrote:
Dear All,

I am trying to compile MySQL 5.0.27 and 5.1.14 on a shared memory processors system with 128 IA-64 Itanium 2.

I have downloaded the binary version of MySQL-max 5.0.27 Linux IA64 and I have installed and configured it with no problems.

I need to install also the MySQL-max 5.1.x Linux IA64 version. I could not find a compiled binary version for MySQL 5.1.x for Linux IA64, so I have downloaded the source code for
MySQL 5.1.14.

Hi!