> "Henryk Szal" <[EMAIL PROTECTED]> writes:
> > YES, this feature should affect ALL locks.
> > 'Timeout on lock' parameter says to server "I CAN'T WAIT WITH THIS
> > TRANSACTION TOO LONG BECAUSE OF (ANY) LOCK",
>
> It still seems to me that what such an application wants is not a lock
> timeout
> > In short, I think lock timeout is a solution searching in vain for a
> > problem. If we implement it, we are just encouraging bad > application
> > design.
>
> I agree with Tom completely here.
>
> In any real-world application the database is the key component of a
> larger system: the w
Thus spake Lamar Owen
> This one probably needs the 'iron hand and the velvet paw' touch. The
> iron hand to pound some sense into the author, and the velvet paw to
> make him like having sense pounded into him. Title of article is 'Open
> Source Databases Won't Fly' --
> http://www.dqindia.com/c
> In latest 7.1 (checked out 2 days ago from CVS), I see following
> behaviour:
>
> create table foo(x int4);
> create function xx(foo) returns int4 as ' return 0;' language 'plpgsql';
> create view tv2 as select xx(foo) from foo;
regression=# create function xx(foo) returns int4 as ' return 0;
On Tue, Apr 17, 2001 at 07:46:19PM +0200, Mario Weilguni wrote:
> I noticed a quite strange behaviour of to_char() in 7.0 and 7.1. It treats
> abbreveated forms of a date completely wrong. Example:
>
> -- this one is ok
> mario=# select to_date('04.01.2001', 'dd.mm.');
> to_date
>
I can't seem to get at the original anymore, but we talked to Dr.
Soparkar, and is posted a 'followup' of the article to:
http://linuxtoday.com/news_story.php3?ltsn=2001-04-16-009-21-PS-EL-HE-0038
Since I can't seem to get to the original on dqindia.com, I can't comment
on what's changed ...
O
> Perhaps I'm stuck in the eighties when I did my thesis in LaTeX, but
> I was of the impression that what's considered good style in LaTeX *is*
> content-based markup. Sure, a LaTeXer may occasionally be forced to
> throw in low-level stuff like a \pagebreak to get nice looking results
> ... but
Thomas Lockhart <[EMAIL PROTECTED]> writes:
>> This is a step forward?
> Not true. If you embed pagebreak commands *in the source* then those
> breaks *must* be reevaluated every time the docs change. If content is
> added or removed, the appropriate place for a page break will likely
> change, s
Zeugswetter Andreas SB <[EMAIL PROTECTED]> writes:
> It is not something that makes anything unrelyable or less robust.
How can you argue that? The presence of a lock timeout *will* make
operations fail that otherwise would have succeeded; moreover that
failure will be pretty unpredictable (at
Alex Pilosov <[EMAIL PROTECTED]> writes:
> In latest 7.1 (checked out 2 days ago from CVS), I see following
> behaviour:
> create table foo(x int4);
> create function xx(foo) returns int4 as ' return 0;' language 'plpgsql';
> create view tv2 as select xx(foo) from foo;
> users=# \d tv2
> ERROR:
> > It is not something that makes anything unrelyable or less robust.
>
> How can you argue that? The presence of a lock timeout *will* make
> operations fail that otherwise would have succeeded; moreover that
> failure will be pretty unpredictable (at least from the point of view
> of the app
> "A Z" == Zeugswetter Andreas SB <[EMAIL PROTECTED]> writes:
PS: where can I find more on the distributed txn plans for PostgreSQL? Thanks.
A Z> BTW: for distributed txns you need a lock timeout feature
A Z> anyway, because detecting remote deadlocks between two or
A Z> more dif
> This is the real reason why I've been holding out for restricting the
> feature to a specific LOCK TABLE statement: if it's designed that way,
> at least you know which lock you are applying the timeout to, and have
> some chance of being able to estimate an appropriate timeout.
As I pointed be
On Wed, 18 Apr 2001, Tom Lane wrote:
> The ruleutils.c bug cannot explain this however, since ruleutils won't
> even be invoked. Can you find a sequence to reproduce it?
Sorry, I was mistaken. The error I get for select is this:
ERROR: cache lookup for type 0 failed
This is a far harder to tri
D. Hageman writes:
> The postgresql interactive terminal will dump core on any script that is
> run via the -f command line option if their exists a connect line without
> a valid user. An example connect line is in one of the attached files.
Okay, I've found the problem. When the connection f
tgconstrrelid (in pg_trigger) holds table references in a RI trigger.
The value in this field is not successfully recreated after a
dump/restore.
---
If I create a simple relationship:
create table p (id int primary key);
create table c (pid int references p);
and query the system table
On Wed, 18 Apr 2001, Alex Pilosov wrote:
> This is a far harder to trigger bug, and actually, it doesn't happen in
> this simple case (oops), and the only test case I have involves 2 tables
> and 3 stored procedures. It is not related to views at all, just doing the
> underlying select causes the
I just checked the CRN PostgreSQL article at:
http://www.crn.com/Sections/Fast_Forward/fast_forward.asp?ArticleID=25670
I see no changes to the article, even though Vince our webmaster, Geoff
Davidson of PostgreSQL, Inc, and Dave Mele of Great Bridge have
requested it be fixed. Not sure what
I am in the middle of a rather nasty experience that I hope someone
out
there can help solve.
My hard disk partition with the postgres data directory got full. I
tried to shut down postgres so I could clear some space, nothing
happened. So I did a reboot. On restart (after clearing some
pg_sort
Joel Burton <[EMAIL PROTECTED]> writes:
> tgconstrrelid (in pg_trigger) holds table references in a RI trigger.
> The value in this field is not successfully recreated after a
> dump/restore.
Yes, this problem was noted a couple months ago. AFAIK it was not fixed
for 7.1, but I concur that it sh
> One idea Tom had was to make it only active in a transaction,
> so you do:
>
> BEGIN WORK;
> SET TIMEOUT TO 10;
> UPDATE tab SET col = 3;
> COMMIT
>
> Tom is concerned people will do the SET and forget to RESET
> it, causing all queries to be affected by the timeout.
> 5) We have been working for translating docs into Japanese using
>EUC_JP encoding. Converting to HTML is no problem, but we cannot
>get correct results for sgml-> RTF conversion at all. The
>translated docs are just not be able to read, showing random
>characters. It seems that o
[ Charset ISO-8859-1 unsupported, converting... ]
> > This is the real reason why I've been holding out for restricting the
> > feature to a specific LOCK TABLE statement: if it's designed that way,
> > at least you know which lock you are applying the timeout to, and have
> > some chance of being
Tom Lane writes:
> 1. "real user" = what you originally authenticated to the postmaster.
>
> 2. "session user" = what you can SET if your real identity is a superuser.
>
> 3. "current user" = effective userid for permission checks.
We could have a Boolean variable "authenticated user is superuse
Send over a context diff and we can get it into 7.2. You may want to
shoot it to the JDBC list too.
[ Charset ISO-8859-1 unsupported, converting... ]
>
> Hi,
>
> I have just modified the jdbc 7.1rc4 source to let the PreparedStatement
> handle null values in setXXX methods gracefully...
>
> A
Bruce Momjian writes:
> However, what it doesn't give you is much control over
> appearance except how to map the tags to appearance. You can't tweek
> appearance in SGML unless you make special tags for certain appearances.
How do you derive this conclusion? SGML gives you a boatload of ways
> On Tue, Apr 17, 2001 at 01:31:43PM -0400, Lamar Owen wrote:
> > This one probably needs the 'iron hand and the velvet paw' touch. The
> > iron hand to pound some sense into the author, and the velvet paw to
> > make him like having sense pounded into him. Title of article is 'Open
> > Source Da
[ Charset ISO-8859-1 unsupported, converting... ]
> > One idea Tom had was to make it only active in a transaction,
> > so you do:
> >
> > BEGIN WORK;
> > SET TIMEOUT TO 10;
> > UPDATE tab SET col = 3;
> > COMMIT
> >
> > Tom is concerned people will do the SET and forget to RESE
Joel Burton <[EMAIL PROTECTED]> writes:
> Do we know if the problem is in pg_dump, or is there no way
> to pass the tgconstrrelid value in the CREATE CONSTRAINT TRIGGER
> statement?
IIRC, pg_dump is just failing to transfer the value; it needs to emit
an additional clause in the CREATE CONSTRAINT
> Bruce Momjian writes:
>
> > However, what it doesn't give you is much control over
> > appearance except how to map the tags to appearance. You can't tweek
> > appearance in SGML unless you make special tags for certain appearances.
>
> How do you derive this conclusion? SGML gives you a boa
On Wed, 18 Apr 2001, Tom Lane wrote:
> Joel Burton <[EMAIL PROTECTED]> writes:
> > tgconstrrelid (in pg_trigger) holds table references in a RI trigger.
> > The value in this field is not successfully recreated after a
> > dump/restore.
>
> Yes, this problem was noted a couple months ago. AFAIK
On Wed, Apr 18, 2001 at 02:22:48PM -0400, Bruce Momjian wrote:
> I just checked the CRN PostgreSQL article at:
>
>http://www.crn.com/Sections/Fast_Forward/fast_forward.asp?ArticleID=25670
>
> I see no changes to the article, even though Vince our webmaster, Geoff
> Davidson of PostgreSQL, In
On Wed, Apr 18, 2001 at 09:54:11AM +0200, Zeugswetter Andreas SB wrote:
> > > In short, I think lock timeout is a solution searching in vain for a
> > > problem. If we implement it, we are just encouraging bad application
> > > design.
> >
> > I agree with Tom completely here.
> >
> > In any re
> If _you_ had been deluged with that kind of vitriol, what kind of favors
> would you feel like doing?
Well, one person's opinion on the article that was perhaps expressed a
little harshly shouldn't cause the company to cover their ears and hum when
their article is in need of multiple corre
> What might be a reasonable alternative would be a BEGIN timeout: report
> failure as soon as possible after N seconds unless the timer is reset,
> such as by a commit. Such a timeout would be meaningful at the
> database-interface level. It could serve as a useful building block
> for appl
On Wed, Apr 18, 2001 at 07:33:24PM -0400, Bruce Momjian wrote:
> > What might be a reasonable alternative would be a BEGIN timeout: report
> > failure as soon as possible after N seconds unless the timer is reset,
> > such as by a commit. Such a timeout would be meaningful at the
> > database-
Here's more info on the bug:
background: function cust_name(customers) returns varchar;
Query in question:
SELECT
cust_name(a)
FROM customers AS a, addresses AS b
WHERE
b.cust_id=a.cust_id
and b.oid=get_billing_record(a.cust_id)
and cust_balance(a.cust_id)>0
First, my idea of what's happening:
> On Wed, Apr 18, 2001 at 07:33:24PM -0400, Bruce Momjian wrote:
> > > What might be a reasonable alternative would be a BEGIN timeout: report
> > > failure as soon as possible after N seconds unless the timer is reset,
> > > such as by a commit. Such a timeout would be meaningful at the
> > >
> D. Hageman writes:
>
> > The postgresql interactive terminal will dump core on any script that is
> > run via the -f command line option if their exists a connect line without
> > a valid user. An example connect line is in one of the attached files.
>
> Okay, I've found the problem. When th
> Tatsuo, when I added SGML reference pages to the back of my book, I took
> the HTML-generated output from SGML and loaded that into LaTeX. I did
> have to do a few things:
>
> convert SGML to HTML
> html2latex
> add * to \subsection* ?
> remove \newline
> remove \
On Wed, Apr 18, 2001 at 09:39:39PM -0400, Bruce Momjian wrote:
> > On Wed, Apr 18, 2001 at 07:33:24PM -0400, Bruce Momjian wrote:
> > > > What might be a reasonable alternative would be a BEGIN timeout:
> > > > report failure as soon as possible after N seconds unless the
> > > > timer is reset, s
Alex Pilosov <[EMAIL PROTECTED]> writes:
> Something (when the query is evaluated, before cust_name function is
> called) sets the tupdesc->natts=0,
Ugh. You verified the natts is wrong in the tupdesc?
> Question: Should SPI_gettypeid look at tuple->t_data->t_natts (to do that,
> it needs to be
Alex Pilosov <[EMAIL PROTECTED]> writes:
> Something (when the query is evaluated, before cust_name function is
> called) sets the tupdesc->natts=0,
FWIW, I have just looked through all the code that sets natts fields,
and I don't believe that any of it can set a tupdesc's natts field to
zero. T
> > It actually was pretty quick. The fixes were more cleaning up strange
> > conversion from HTML to LaTeX.
>
> Looks nice, but I'm afraid I have to do all the work above for 489
> HTML files:-)
>
> What I'm doing now is trying to fix openjade. It is written in C++,
> and I hate C++, no way...
At 16:30 18/04/01 -0400, Tom Lane wrote:
>
>IIRC, pg_dump is just failing to transfer the value; it needs to emit
>an additional clause in the CREATE CONSTRAINT command to do so.
>
>From memory, this is one of the non-standard SQL things that pg_dump still
does (ie. definining the constraint usin
At 16:25 18/04/01 -0400, Joel Burton wrote:
>
>Do we know if the problem is in pg_dump, or is there no way
>to pass the tgconstrrelid value in the CREATE CONSTRAINT TRIGGER
>statement?
>
It's because pg_dump is not designed to dump these constraints *as*
constraints. We just need to make pg_dump
Alex Pilosov <[EMAIL PROTECTED]> writes:
> Here's more info on the bug:
> background: function cust_name(customers) returns varchar;
> Query in question:
> SELECT
> cust_name(a)
> FROM customers AS a, addresses AS b
> WHERE
> b.cust_id=a.cust_id
> and b.oid=get_billing_record(a.cust_id)
> and cus
> > 5) We have been working for translating docs into Japanese using
> >EUC_JP encoding. Converting to HTML is no problem, but we cannot
> >get correct results for sgml-> RTF conversion at all. The
> >translated docs are just not be able to read, showing random
> >characters. It se
Tatsuo Ishii <[EMAIL PROTECTED]> writes:
>> They are residual markup for graphics from Stephan's Master's Thesis
>> which were never transcribed from the originals (gifs?) to a usable
>> format.
>>
>> Through disk crashes, system upgrades, and a failed backup device I
>> *may* no longer have his
49 matches
Mail list logo
