ic/install-procedure.html
If you are not installing from source them whomever compiled the package
you are installing will have specified where they are placing files and you
will need to inquire of them. The PostgreSQL docs don't cover file
locations in depth because each of the many platforms we support have their
own conventions.
David J.
es.
The core project doesn't control, and thus does not document, those
procedures.
David J.
t;named" - "...while
connected to the named database."
Given that, aside from 3 release note instances, we never use the word
"victim" in the docs leads me to want to change this if just for internal
consistency.
David J.
a patch wondering whether it would be accepted is
not particularly productive.
David J.
he complexity that would be
introduced explaining how it differs from other reserved words.
I wouldn't be opposed to adding a sentence and cross-reference link to the
system columns page and pointing out that specific column names are also
already reserved for system use.
David J.
y.csv' DELIMITER
> ',' FORMAT
> csv;
> ERROR: syntax error at or near "FORMAT"
> LINE 1: ...entity FROM '/tmp/podl_db_city.csv' DELIMITER
> ',' FORMAT csv...
>
The parentheses surrounding the options are not optional. Personally I
like to include the WITH.
COPY gis_entity FROM 'file.csv' WITH ( FORMAT csv );
David J.
) for the
server, while the later, despite having to be configured by the server
admin, plays more of an application role.
David J.
On Monday, September 18, 2017, wrote:
>
> But if the variable is not set, I want the script to prompt for
> it.
In version 10 this should be doable, we introduced \if conditionals. You
might be able to hack up something via "\!" in 9.6 and earlier (haven't
tried it myself).
David J.
on.
>
>
The to_char examples in that table show examples from a variety of locales
without specifying which one they are using. The result shown can be
obtained if your locale uses space and comma for group and decimal
separators. Yours does not so you get the result you see. This could be
improved.
David J.
casted to text. The text
representation of a composite type is documented.
I cannot think of anything that the docs need to become more complete in
this area; and they don't purport to be a "teacher of useful techniques" in
the application of SQL. While I've complained about the lack of "why"
discussion in the docs there is merit in focusing them on "what" and "how"
and leaving why to alternative mediums - ones whose release cycles are more
flexible and don't require a hacker to commit.
David J.
lt User-Database Configuration (GUC)
value (which in this case applies to all users connecting to this
database); which is what the last 4 syntax specifications pertain to - as
noted in the description section on that page.
18.1.3 describes this dynamic in more detail:
https://www.postgresql.org/docs/9.5/static/config-setting.html#CONFIG-SETTING-SQL-COMMAND-INTERACTION
David J.
t on the current sequence value*; it simply
sets the value that future ALTER SEQUENCE RESTART commands will use."
(emphasis
mine)
https://www.postgresql.org/docs/9.5/static/sql-altersequence.html
David J.
l
are use if you wish to convert the warm standby into a hot standby. In
either case the relevant wal_level needs to be "replica". It was deemed
unnecessary to distinguish between "archive (cold/warn) and hot_standby" in
9.6 - in terms of what information was recorded in the WAL.
David J.
t the privileges of
the rule owner, not the user invoking the rule."
It does feel like an additional blurb about views and a link to the above
page would be warranted on the ddl-rowsecurity.html page.
David J.
On Wed, Aug 30, 2017 at 5:09 PM, Bruce Momjian wrote:
>
> Stephen, is there any
t grammatically
wrong and fails to fix the underlying problem. "about" it is right word
but the sentence fails to provide "what", what must only appear in the
result list or having clause?
A meaningful example might help too - I'm having trouble imaging when one
would choose to write a subquery whose effect would end up being treated as
part of the main query.
David J.
about ALTER SEQUENCE is so low that the
motivation to dream up and add new examples just isn't there. If a patch
was submitted, though, it would likely be committed.
David J.
7;m not aware of any recent benchmark
runs that would indicate whether improvements to arrays in recent versions
of PostgreSQL have narrowed or eliminated that gap.
David J.
ll fail to
> understand it.
>
>
Table 9-26 on that page says "(uses locale)" for D and G, the user is left
to deduce that since comma and period do not have that qualifier that they
are not locale dependent.
> I guess there should be some more explanation.
>
I don't really agree but I wouldn't discourage someone from submitting a
patch.
David J.
ar with the feature knows that they have
options other than just writing out the comma and period literally. If one
wants to try the example in a language other than in which the example was
written they should modify it so that the literal number being parsed in
written in conformance to the locale definition for the language you are
using.
David J.
If you do re-post in the appropriate list I would suggest you at least
demonstrate how it is that you know what you claim.
David J.
what is the problem
>
> Are you saying that ALTER SEQUENCE ... OWNER TO doesn't work? It
> appears to work for everyone else.
>
Sequences can have two owners - a role and, optionally, a table.column.
The name following "OWNED BY" is a table.column (NONE to clear).
The name following "OWNER TO" is a role (mandatory).
David J.
On Wed, Aug 9, 2017 at 4:03 PM, David G. Johnston <
[email protected]> wrote:
> On Wed, Aug 9, 2017 at 3:21 PM, Jym Morton wrote:
>
>> When I write software, and use a database I don’t need to escape literals
>> if I have a Prepared Statement. This is a maj
evel of knowledge
for the reader so that they can avoid being repetitive. You may disagree
with what is assumed but I stand by my opinion that whomever wrote this
example did not need to consider that their example would be improved if
they used a value that required escaping.
Ultimately, if someone wanted to act on your suggestion it wouldn't bother
me. But I see no systematic problem with this example.
David J.
e different kinds of commands are
or, more commonly, needs a refresher as to the exact syntax for less often
used commands.
More precisely, the vast majority of the docs assume you've read "Chapter
4. SQL Syntax".
David J.
Reading the entire
paragraph, including the first introduction of the term "JSON documents",
leads me to think that the best solution here is to remove the
double-quotes surrounding "documents" and remove the parenthetical
entirely. I'm not seeing how it adds or clarifies anything here - the
reader has been told that an alias for "json value" is "document".
Introducing the concept of "datum" here is just going to prompt questions
such as this.
David J.
On Tue, Aug 1, 2017 at 7:23 AM, Jeremy Finzel wrote:
>
> I *think* I follow now: before the checkpoint, the dirty data pages had
> already been written to the WAL files, but not necessarily flushed to
> disk. The checkpoint flushes them all to disk as well.
>
Yes, that it correct.
David J.
ys a checkpoint has occurred
and that every change noted in the WAL files written before the marker
entry and now known to exist in the actual data files on disk. Until that
happens some of those change may only exist in memory (in the form of dirty
data pages).
David J.
ity' to
> be summed in the primary SELECT statement. I think this is a mistake.
It isn't.
> I do
> not think this field exists in the WITH virtual table
>
It wouldn't matter if it did since in the main query the only from item is
the orders table which can be reasonably assumed to contain a quantity
column.
David J.
cause/solution would not be covered. Plus, frankly, the documentation
doesn't go to any great efforts to deal with run-time environment problems
(and its update schedule is not really suited for such). A more
appropriate forum would be our wiki.
https://wiki.postgresql.org/wiki/Main_Page
David J.
is not mechanically equivalent since, as you
say, the contents of postgres.conf are not actually changed.
David J.
;
All -W does is force a prompt. There is no way to actually supply the
password on the command line, that is what environment variables or pgpass
are for.
David J.
t how things should work that are not
true.
Your welcome to your venting but all I see here is a specific case of
learning having gone into spiral. The community here is great at
helping people get themselves out these kinds of spirals. That the
documentation cannot do so all by itself is not
g system's shell. You run "create
database" with psql or some other SQL executing client. From your
description you appear to be in an SQL executing program hence createdb and
other command line applications will not work.
David J.
t second sentence is part of
the overall policy being described. The policy has two components which is
why both FK constraints were changed.
David J.
ct? false
== off when dealing with a boolean parameter as described at:
https://www.postgresql.org/docs/9.5/static/config-setting.html#CONFIG-SETTING-NAMES-VALUES
David J.
On Thu, Apr 6, 2017 at 6:14 PM, Bruce Momjian wrote:
> On Thu, Apr 6, 2017 at 03:53:24PM -0700, David G. Johnston wrote:
>
> > hypothetical-set aggregates as being distinct from ordered-set
> aggregates yet
> > the tables and the docs here do.
>
> Do you have
On Thu, Apr 6, 2017 at 2:49 PM, Bruce Momjian wrote:
> As part of writing a talk about window functions, I have done some
> cleanups of the window function docs, attached, that I would like to
> apply to head.
+1 Overall. Some minor technical items noted below. The only conceptual
item is whe
at would also be permitted to destroy data. Having such a dependency
listing query distinguish between data-loss and other would be a good
intermediate step.
I haven't thought this out in depth, and its probably an unlikely area for
improvement, but having recently spent a couple hours re-learning this
(then confirming its working as intended via the docs) at least makes me
want to bring it up.
David J.
hould probably add some kind of warning or something.
>
Yeah, it should probably be modified to something like:
Automatically drop objects (tables, functions, etc...) that are contained
in the schema. Each object dropped is also done with the CASCADE option.
This means that others schemas will be affected if they depend on objects
in the named schema.
David J.
s textually as the empty string.
I'd maybe modify the table to read:
void : A value that has no content: used to indicate that a function
returns no value.
https://www.postgresql.org/docs/devel/static/datatype-pseudo.html
David J.
The second closing bracket after "connlimit" should be moved after
"istemplate ]"
The IS_TEMPLATE option can be prefixed by "WITH" when used alone.
i.e., create database newdb with is_template false;
David J.
On Saturday, May 21, 2016, Tom Lane wrote:
> "David G. Johnston" > writes:
> > Based upon what you've said I would soften it a bit. Given my own
> > experience I'd probably point out what is now obvious to me - that the
> > allowance of the ORDER BY
On Fri, May 20, 2016 at 5:07 PM, Tom Lane wrote:
> "David G. Johnston" writes:
> > http://www.postgresql.org/docs/9.5/static/functions-aggregate.html
> > """
> > SELECT xmlagg(x) FROM (SELECT x FROM test ORDER BY y DESC) AS tab;
> > But this s
CT x FROM test ORDER BY y DESC))
The example seems expressly permitted by the standard and other database
systems.
I'll believe that said ordering in the example is not guaranteed but that
isn't what it says.
David J.
On Thursday, May 12, 2016, Peter Eisentraut <
[email protected]
> wrote:
> On 5/7/16 11:16 PM, David G. Johnston wrote:
>
>> While this may be in the postgresql.conf it's absence in the docs is made
>> noticeable by the fact both the immediately precedi
using max_worker_processes - provoking the very question as to how many is
2 relative to the default pool size.
http://www.postgresql.org/docs/devel/static/runtime-config-resource.html#GUC-MAX-PARALLEL-DEGREE
David J.
uld it be hateful to make it so that both entries appear on both tables?
array_to_string(anyarray, text [,text]) could be lumped in with this as
well.
string_to_array not being on the string functions page is the more glaring
omission for me.
David J.
ere feature a query example show the use of a
> wildcard for columns
> >> as well as allowing you to assign names to as many of the leading
> columns as you wish:
> >>
> >>
> >> WITH T0 as ( SELECT 1,2,3 )
> >> SELECT T0.* from T0 as T0(foo,bar) ;<
rs to it using *
In this case the FROM clause is what is being alised. It is documented
though I'd need to look to identify the specific location. It would not be
documented in a section regarding the select-list.
David J.
On Thursday, August 27, 2015, Bruce Momjian wrote:
> On Thu, Aug 27, 2015 at 11:50:53AM -0400, David G. Johnston wrote:
> > Yes, I read that as "\pset tuples_only" and "\t" are alternate forms of
> the
> > same thing
> > and thought that "\pset
On Thu, Aug 27, 2015 at 11:13 AM, Bruce Momjian wrote:
> On Wed, Aug 26, 2015 at 10:51:09PM -0400, David G. Johnston wrote:
> > On Wed, Aug 26, 2015 at 10:02 PM, Bruce Momjian
> wrote:
> >
> > On Wed, Jun 24, 2015 at 06:11:20PM -0400, David G. Johnst
On Wed, Aug 26, 2015 at 10:02 PM, Bruce Momjian wrote:
> On Wed, Jun 24, 2015 at 06:11:20PM -0400, David G. Johnston wrote:
> > http://www.postgresql.org/docs/devel/static/app-psql.html
> >
> > The other pset keywords with black-slash aliases include a "(or
> )
s of arguments but in the reverse order.
At least 9.3 appears not to...
David J.
http://www.postgresql.org/docs/devel/static/app-psql.html
The other pset keywords with black-slash aliases include a "(or )"
paren after the keyword. "title" does not (should be "C")
David J.
e public API for the application.
Adding mention of pg_reload_conf may be warranted (probably is but I
haven't made an independent evaluation) but removing SIGHUP is a separate
concern and one that doesn't seem necessary.
David J.
On Saturday, April 25, 2015, David G. Johnston
wrote:
> On Saturday, April 25, 2015, Kevin Grittner > wrote:
>
>> Bruce Momjian wrote:
>> > On Sat, Apr 25, 2015 at 07:45:35PM +, Kevin Grittner wrote:
>>
>> >> We could perhaps have the column heade
possible (contra-SQL)" and emphasize our
implementation and footnote the two differences.
David J.
On Saturday, April 25, 2015, Kevin Grittner wrote:
> Bruce Momjian > wrote:
> > On Sat, Apr 25, 2015 at 07:45:35PM +, Kevin Grittner wrote:
>
> >> We could perhaps have the column header say "Non-Serializable
> >> Behavior" or some such; but I think we need to define whatever
> >> term we use
http://momjian.us/expire/transaction-iso.html
Need to add "Serialization Anomalies" to the previous section's definitions
list.
Pondering whether something like: "Possible (not in PG)" and avoiding the
additional rows would make reading the table easier.
David J.
On Fri, Apr 24, 2015 at 9:57 AM, Peter Eisentraut wrote:
> On 4/17/15 7:36 PM, David G. Johnston wrote:
> > diff --git a/doc/src/sgml/mvcc.sgml b/doc/src/sgml/mvcc.sgml
> > index f88b16e..5002138 100644
> > --- a/doc/src/sgml/mvcc.sgml
> > +++ b/doc/src/sgml/mvcc
ould think to go there for additional information. No
like though - just a pointer to it or the Internet generally.
It is not obvious to me what means...I suspect 1=yes...
David J.
On Wed, Apr 15, 2015 at 6:26 PM, David G. Johnston <
[email protected]> wrote:
> On Wednesday, Ap
On Wednesday, April 15, 2015, Bruce Momjian wrote:
> On Mon, Apr 13, 2015 at 08:00:38PM -0700, David G. Johnston wrote:
> > http://www.postgresql.org/docs/9.4/static/transaction-iso.html
> >
> > Table 13-1 shows the SQL standard isolation levels and what is and is not
>
, probably right after the paragraph, with the same
columns, three rows, and the corresponding possible/not-possible cell
values?
David J.
Y idx BETWEEN 5
PRECEDING AND CURRENT ROW" isn't obviously wrong until you get a syntax
error...
David J.
that it's a page targeted at
> beginners.
That example uses the numeric type for currency; which is considerably less
convoluted then forcing everything to have hundredths resolution and
multiplication if you want to show the nominal unit (e.g. dollars).
David J.
--
View this message in context:
sting forms, while correct, are repetitive
and that is what is being targeted but combining them into a single syntax
block. But if that is too hard to read and understand then maybe some other
means of simplification, like my example above, would work better.
David J.
--
View this messa
parts - which \drds does not do.
I am imagine there might be a reason that the catalog pages do not
reference the relevant psql meta-commands, or provide human-friendly
examples, but I cannot think of what that might be and so here I provide.
David J.
diff --git a/doc/src/sgml/catalogs.sgml b/doc/src
On Mon, Dec 29, 2014 at 12:12 PM, Tom Lane wrote:
> David G Johnston writes:
> > Tom Lane-2 wrote
> >> There is a warning against combining IYYY with MM/DD, but it's buried
> >> in trivia far down the page.
>
> > Can we move this warning/notice into code
he typical mistake goes away.
David J.
--
View this message in context:
http://postgresql.nabble.com/The-IYYY-mess-again-tp5832272p5832322.html
Sent from the PostgreSQL - docs mailing list archive at Nabble.com.
--
Sent via pgsql-docs mailing list ([email protected])
To make change
tion parameters; see Section 31.1.2 for a list but note that you may
not include the "service" parameter within the pg_service file (i.e.,
aliases are not allowed).
Thoughts?
David J.
P.S.
Any chance on actually making this work? The recent comment about "\c
service=name" not
ation the second attempt succeeds. The alternative is to define an
overloaded function without a VARIADIC parameter but with the same
non-optional ones."
Thoughts? Am I missing something?
David J.
not
supported any more.
The last two sentences are redundant.
http://www.postgresql.org/docs/devel/static/protocol-flow.html#PROTOCOL-ASYNC
David J.
On Tuesday, December 16, 2014, David Johnston
wrote:
> Forwarding to -docs for now.
>
> I suggest simply dropping the "cmd.exe /c"part from the docs and mention
> that the command should be run from an active console. A recommendation
> for a batch file or similar
;t see one come across but for now just
reporting this. Would want to research methods for making both these items
persist and recommending such. See my other curiosities below...
David J.
-- Forwarded message ------
From: *David G Johnston [via PostgreSQL]* <
ml-node+s1045698n583104
On Thu, Sep 11, 2014 at 9:24 PM, johnlumby wrote:
> On 09/11/14 10:03, David Johnston wrote:
>
On Thu, Sep 11, 2014 at 9:24 AM, John Lumby wrote:
>
> As I mentioned before the fact that null is a literal/constant is
> assumed. My point here is that given that assumption
John, please respond to the list next time.
On Thu, Sep 11, 2014 at 9:24 AM, John Lumby wrote:
> Thanks David ... but ... (below)
>
>
> > Date: Wed, 10 Sep 2014 18:20:00 -0700
> > From: [email protected]
> > To
d you go looking for it and where, in
order, did you look?
I'll agree that it seems that a previous recognition that null can always
used as a valid data value is assumed much like it is assumed that 1000 is a
valid value for an integer.
David J.
--
View this message in context:
http://p
On Wed, Sep 10, 2014 at 4:20 PM, Bruce Momjian wrote:
> On Tue, Sep 9, 2014 at 03:40:45PM -0400, David Johnston wrote:
>
> > "In addition to setting global defaults or attaching" <- add a comma
> after
> > "defaults"
>
> I didn'
On Tue, Sep 9, 2014 at 3:13 PM, Bruce Momjian wrote:
> On Thu, Jun 12, 2014 at 11:32:49AM -0400, David Johnston wrote:
> > This addition still needs some work as well as the patch as a whole (see
> my
> > comments above). The attached PDF is the page that results from &quo
form. The goal would be to
let them focus on writing but then have some kind of editing process - which
is what is going on now - to aid in polishing the final publication.
David J.
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/PATCH-Various-documentation-typo-gramma
forgot the list...
On Thu, Jun 12, 2014 at 2:05 PM, David Johnston
wrote:
> On Thu, Jun 12, 2014 at 1:25 PM, Tom Lane wrote:
>
>> David Johnston writes:
>> > On Thursday, June 12, 2014, Tom Lane wrote:
>> >> I don't think that argument holds water ei
On Thursday, June 12, 2014, Tom Lane wrote:
> David G Johnston > writes:
> > On Thu, Jun 12, 2014 at 10:42 AM, Tom Lane-2 [via PostgreSQL] <
> > [email protected] > wrote:
> >> I wonder if we should round fractions up instead of down in
3 in Server Configuration?
http://www.postgresql.org/message-id/cakfquwbbozrqmnx-xryd+vit+mcah6tvc0yn1ajtg5mdsna...@mail.gmail.com
would be sufficient.
Green field maybe I'd say yes but given that the new behavior could turn
features on that are currently off it doesn't seem to be bene
David G Johnston wrote
>
> Tomonari Katsumata-2 wrote
>> Hi,
>>
>> I'm reading about log_rotation_age.
>> And I noticed that the setting for disabling the feature is not clear.
>>
>>
>> The document tells us to set it to ZERO if we want to
gest documenting, if correct, that maximum resolution is minutes
(rounding down) and thus any sub-1-minute value will be interpreted as zero
and disable the feature.
David J.
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/disabling-log-rotation-age-feature-tp5806936
Re-sending to hopefully avoid admin moderation rule. Fixing top-post as
well.
On Wed, May 21, 2014 at 4:37 PM, Tom Lane wrote:
>
>> David G Johnston writes:
>> > Section 18.1.3 is named "18.1.3. Other Ways to Set Parameters" yet does
>> not
>> &g
[^]" and "^$" match the newline character, or not.
[w] anchor newline-matching
[n] dot/inverse-bracket newline-matching
[s] newline-matching
These are precise, what-oriented, names compared to:
[w] record mode
[n] multi-line mode
[s] single-line mode
which are more descriptive, use-oriented, names.
Use of these label sets is not mutually exclusive...
David J.
On Thu, Jun 5, 2014 at 8:00 PM, Tom Lane wrote:
> David G Johnston writes:
> > I simplified ". and bracket expressions" to "wildcard" and "^ and $" to
> > "anchors" though did make use of ^ and $individual quite a bit. I did
> not
>
atch near an
embedded newline - you have to use a literal \n to do that and then deal
with the newline itself being part of the capture. This is best thought of
as a compatibility mode since you can get the same behavior, without losing
the unique behavior of ^ and $, in anchor-only mode with
link]set_config[link]
function.
Also, I would suggest modifying the function definitions in table 9-58 to
include the data types:
current_setting(setting_name text)
set_config(setting_name text, new_value text, is_local boolean)
David J.
--
View this message in context:
http://postgresql.10456
cated (since at least 8.4)...
"-b" is only needed for selective dumps...
"-v" you've turned on verbose mode - maybe include the last handful of
output lines in your complaint...
Given the apparent testing of different versions of PostgreSQL you should
also provide the pairing of
>
> Regards
>
>
> Ian Barwick
While providing additional warnings is good and necessary it may also help
to be more descriptive as to in what situations tablespaces are appropriate
and/or necessary so that people leave with a better understanding of why the
feature exists and no
lt) as
an outcome of MVCC should be included - in some form - in the introduction.
In the current, proposed, and my revisions it is indeed covered but to
various degrees of detail and low/high level focus.
David J.
--
View this message in context:
http://postgresql.1045698.n5.nabble
Bruce Momjian wrote
> On Tue, Nov 12, 2013 at 03:36:01PM -0800, David Johnston wrote:
>> Bruce Momjian wrote
>> > We just want to get across the MVCC concept in the intro --- we cover
>> > the snapshots later in the document.
>>
>> I just think we're be
regarding the
docs I try and put myself in the person's shoes and find not that the docs
are incorrect but that they could be improved - which is just a part of our
reality).
My $0.02
David J.
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/MVCC-snapshot-timing-tp5
David Johnston wrote
> This reads badly to my ears:
>> This means that while querying a database each SQL statement sees a
>> snapshot of data (a database version) as it was some time ago, regardless
>> of the current state of the underlying data.
> How about something clos
intro but I wanted to get
opinions on the approach first.
David J.
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/MVCC-snapshot-timing-tp559p5777852.html
Sent from the PostgreSQL - docs mailing list archive at Nabble.com.
--
Sent via pgsql-docs mailing
Sorry if you get this twice but I use Nabble and didn't subscribe to the list
so my originals got put into the verification queue. I've subscribed now
and am re-posting hoping it will go through clean.
See my self-quote comment and my direct comment at the end.
David Johnston wrot
+1 for PostgreSQL tutorials that actually tutor. Is it worth trying
to do a baseline general one first, or going directly to separate ones
for DBAs, app & DB developers, etc.?
Cheers,
David.
On Sat, Aug 03, 2013 at 06:45:49PM -0700, Craig Kerstiens wrote:
> Count me as both intrig
orrection is
desired. For my purposes I'll likely just leave it alone.
David J.
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/Create-Table-Like-Including-Default-Sequences-tp5759108.html
Sent from the PostgreSQL - docs mailing list archive at Nabble.com.
--
1 - 100 of 225 matches
Mail list logo
