Huh. Completely the wrong post to reply to. Sorry folks... this of course was
meant for the "Oracle Sequence" question.
Brett
Quoting Brett Connor <[EMAIL PROTECTED]>:
> selecting max... isn't a viable solution, because of other sessions as you
> say.
> Assuming y
(In case my reply to completely the wrong post didn't get read!...)
selecting max... isn't a viable solution, because of other sessions as you say.
Assuming you're at least using Java, you might want to look at
java.sql.Statement.getGeneratedKeys()
Oracle database returns generated keys in the '
selecting max... isn't a viable solution, because of other sessions as you say.
Assuming you're at least using Java, you might want to look at
java.sql.Statement.getGeneratedKeys()
Oracle database returns generated keys in the 'C' i/f, my memory's rusty for
JDBC but I know there is a way somewher
", " _ ", " @ " characters.
This is definately incorrect. Certainly + for example is legal in the local name
part. I've been unable to give money to one or two companies who's email
registration have had this validation bug. Shame.
Brett Connor
Quoting Bart
(the part before
the @). So there are many other characters that can legally appear. If I
have to do my own regexp I typically use something like
[EMAIL PROTECTED]@[A-Za-z0-9_-]. etc
Regards
Brett Connor
-
To unsubsc
Javadoc will automatically generate parameter names and types anyway.
Just my humble opinion.
Brett Connor
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Hyrum wrote:
I am trying to use the Ant 'cactus' optional task in my build.xml. This
fails because Eclipse must have the optional task set in the preferences,
but I can't figure out how to do this.
If I have understood your issue correctly, you need add the jar with the
optional task to t
e already had to take steps to change banks
because they only supported proprietry browsers rather than suport the
standards. We should not be allowing these things to slip through IMHO.
Brett Connor
Victor Grazi wrote:
Not sure I understand.
What's a CVS? Are u referring to a version
[EMAIL PROTECTED] wrote:
Brett Connor <[EMAIL PROTECTED]> wrote on 09/09/2004 12:22:39 PM:
Taking a step back though, the locale is required for presentation only
(in this example), it belongs in the view rather than any business
components. It is not the job of the business layer
is as useful to my
mum as UnsupportedOperationException. One nicety I like in these cases
is to include a reference to a log file or system error report or
something that people can use in an error report, but I'm wandering a
bit of topic here. My 2c.
Brett Connor
--
d up here. I think I would
have a peruse of the tomcat source if you can get it for your port, see
if there's anything helpful in there. I know it's not the most enjoyable
of tasks wading through a ton of code you don't know (I'm assuming you
don't), but I've been wat
r meaningful context available at the point where it's
reported. In some ways, the further removed and isolated the reporting
is from the sources of error the less easy it is without some effort,
yet again lots of IMHO all through this.
Brett Connor
---
I usually wack log and assert false; if there is nothing better to do
with it at the time. In situtations where you know an exception can
never be thrown, I still would leave in at least this.
Of course this view also comes I'm sure from being in your position
before now. I don't recall but it's
k
is required, probably more to make sure things are thread safe rather
than distinguish between different windows.
Brett Connor
Leandro Melo wrote:
I could save the reference to stateful session ejb and
jsessionid in a map in HttpSession.
This way, things wouldn't get counfusing, am i right?
xxx needs to be somewhere accessible to the c:forEach tag. Do a search
on scoped variables. This can be done in your scriptlet or in the action
that forwards to this jsp.
Brett
andy wix wrote:
Hi,
In trying to resolve another (more Struts related!) problem someone
suggested using a
<%@ taglib
x27;re using I doubt holding a
reference in the action is really what you want to do, unless I've
misunderstood your situation here.
Brett Connor
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Craig McClanahan wrote:
On Wed, 11 Aug 2004 10:32:04 -0700, Wiebe de Jong <[EMAIL PROTECTED]> wrote:
I had a similar problem, which I discovered when one of my users tried to
enter a street address containing an apostrophe. Since I use apostrophes to
delineate my text strings in my SQL statement
Woodchuck wrote:
hihi,
this is my scenario:
- container managed pooling (Tomcat 4.1.24)
- jdbc (PreparedStatement objects)
- wrapping my own transactions by autoCommit(false), and issuing
commit() manually
i'm running into a table-lock situation after DELETE statements. i'm
using the same Connecti
Brett Connor wrote:
David Stevenson wrote:
On 11/8/04 6:18, "Michael McGrady" <[EMAIL PROTECTED]> wrote:
I loved those competitions in perl that tried to
be the hardest to understand.
Competitions? I've seen C ones (the 12 days of Xmas one is a
classic), but
truly i
David Stevenson wrote:
On 11/8/04 6:18, "Michael McGrady" <[EMAIL PROTECTED]> wrote:
I loved those competitions in perl that tried to
be the hardest to understand.
Competitions? I've seen C ones (the 12 days of Xmas one is a classic), but
truly idiomatic Perl is self-obfuscating!
Let's not
7;s good for you you will stay right away from javascript..
-Original Message-
From: Brett Connor [mailto:[EMAIL PROTECTED]
Sent: Wednesday, July 28, 2004 4:47 PM
To: Struts Users Mailing List
Subject: Re: Not Logged in with the same loggin-password if it is
already registered.
AFAIK session
/her window? Are you saying
this isn't something that a container will be notified about..?/ ..hmm... maybe not... ok,
Robert scratch my earlier note..(like I said i ought to go home..)
-Original Message-
From: Brett Connor [mailto:[EMAIL PROTECTED]
Sent: Wednesday, July 28, 2004 4:47 PM
AFAIK session timeout is the only notification you can get here. There
are undoubtedly some wacky things we can do with javascript - reduce the
session timeout to a small value and include some javascript polling in
every page or something along those lines, but at the end of the day it
comes d
bean:write has a format property. I haven't seen any documentation on
what the format syntax is, but there's every chance it supports
something along the lines of
.
Alternatively, create a derived property on your form bean and reference
that, eg
class MyForm extends ActionForm {
S
is a separate
thread). There seems to be no way for the server to know that the browser
was closed. If I do not reset the session variable then I cannot start
generating another report until the first report thread dies on the server.
-Original Message-
From: Brett Connor [mailto:[EMAIL PROT
Surely the session variable will be reset by the report generation
process, independent of any browser windows open, therefore no problem,
you never need to know if the browser has been closed or not. Or have I
misunderstood?
Brett
Rizwan Merchant wrote:
Thanks for the previous input. I moved t
rs will be less helpful.
There was some effort a while ago to standardize some of the common
database error codes, I don't know how far that got, but as always these
standards are the lowest common subset anyway.
HTH
Brett Connor
Research labs wrote:
Thanks for your prompt respon
g on and you can take the exception at its word.
Dunno if that helps at all.
Brett Connor
Lucas Gonzalez Pearson wrote:
Hi,
I am still trying to find the root of this problem but I don´t seem to be
able to do it.
Has anyone experienced this problem before?
copying a struts .war file to jboss´ dep
28 matches
Mail list logo
