Re: [HACKERS] GSOC'17 project introduction: Parallel COPY execution with errors handling

2017-04-11 Robert Haas : > There's a nasty trade-off here between XID consumption (and the > aggressive vacuums it eventually causes) and preserving performance in > the face of errors - e.g. if you make k = 100,000 you consume 100x > fewer XIDs than if you make k = 1000,

Re: [HACKERS] Bitmap index scans use of filters on available columns

2015-11-04 Antonin Houska : > While prefix expression > > y like 'abc%' > > can be converted to > > y >= 'abc' > > (see expand_indexqual_opclause()), I'm not sure any kind of expansion is > possible for '%abc%' which would result in a b-tree searchable condition. I think the

Re: [HACKERS] a funnel by any other name

2015-09-17 Robert Haas : > 1. Exchange Bushy > 2. Exchange Inter-Operator (this is what's currently implemented) > 3. Exchange Replicate > 4. Exchange Merge > 5. Interchange > 1. ? > 2. Gather > 3. Broadcast (sorta) > 4. Gather Merge > 5. Redistribute > 1. Parallel Child

Re: [HACKERS][PROPOSAL] Covering + unique indexes.

2015-09-16 Rod Taylor : > 2015-09-15 Anastasia Lubennikova : > >> - We have a table tbl(f1, f2, f3, f4). >> - We want to have an unique index on (f1,f2). >> - We want to have an index on (f1, f2, f3) which allow us to use index for >> complex

Re: [HACKERS][PROPOSAL] Covering + unique indexes.

2015-09-15 David Rowley : > I'm also a bit confused where f3 comes in here. If it's UNIQUE on (f1,f2) > and we include f4. Where's f3? Columns f1, f2, f3 are in the internal nodes of the tree (i.e., they are used to find the ultimate leaf nodes). f4 is only in the

Re: [HACKERS] Eliminating CREATE INDEX comparator TID tie-breaker overhead

2015-07-23 Robert Haas robertmh...@gmail.com: I think what you should do is go find out whether the second rationale is valid or not. Knowing how much impact on performance such “non TID ordered” entries have, would of course be very useful for future patch authors to know. Especially useful

Re: [HACKERS] Eliminating CREATE INDEX comparator TID tie-breaker overhead

2015-07-24 Nicolas Barbier nicolas.barb...@gmail.com: Especially useful would be to know whether interleaving a small number of TID ordered streams (as would probably be generated by parallel scans/processing) would result in an ordering that performs significantly worse or not. I assume

Re: [HACKERS] Is it possible to have a fast-write Index?

2015-06-05 deavid deavidsed...@gmail.com: Mode 3: on aminsert, put the new entry on a second btree; leaving the first one untouched. Because the second btree is new, will be small, and writes should be faster. When doing a index scan, read tuples from both at same time (like merge sort). On

Re: [HACKERS] Re: Patch to add functionality to specify ORDER BY in CREATE FUNCTION for SRFs

2015-01-05 Tom Lane t...@sss.pgh.pa.us: What would make sense to me is to teach the planner about inlining SQL functions that include ORDER BY clauses, so that the performance issue of a double sort could be avoided entirely transparently to the user. Another way of getting to the point

Re: [HACKERS] Materialized views don't show up in information_schema

2014-10-18 Stephen Frost sfr...@snowman.net: * Peter Eisentraut (pete...@gmx.net) wrote: More subtly, if we claim that a materialized view is a view, then we cannot have asynchronously updated materialized views, because then we have different semantics. This is, at least, a reason I can

Re: [HACKERS] Materialized views don't show up in information_schema

2014-10-16 Stephen Frost sfr...@snowman.net: Alright, coming back to this, I have to ask- how are matviews different from views from the SQL standard's perspective? Matviews that are always up to date when you access them are semantically exactly the same as normal views. Matviews that can get

Re: [HACKERS] Minmax indexes

2014-08-07 Oleg Bartunov obartu...@gmail.com: +1 for BRIN ! +1, rolls off the tongue smoothly and captures the essence :-). Nicolas -- A. Because it breaks the logical sequence of discussion. Q. Why is top posting bad? -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)

Re: [HACKERS] Minmax indexes

2014-08-06 Claudio Freire klaussfre...@gmail.com: So, I like blockfilter a lot. I change my vote to blockfilter ;) +1 for blockfilter, because it stresses the fact that the physical arrangement of rows in blocks matters for this index. Nicolas -- A. Because it breaks the logical sequence of

Re: [HACKERS] 9.4 release notes

2014-05-05 Bruce Momjian br...@momjian.us: On Mon, May 5, 2014 at 10:40:29AM -0700, Josh Berkus wrote: * ALTER SYSTEM SET Lemme know if you need description text for any of the above. OK, great! Once I have the markup done, I will beef up the descriptions if needed and copy the text up

Re: [HACKERS] Patch: iff - if

2014-04-17 Michael Paquier michael.paqu...@gmail.com: Is there no equivalent in German? For example in French there is ssi. gdw (genau dann, wenn) Nicolas -- A. Because it breaks the logical sequence of discussion. Q. Why is top posting bad? -- Sent via pgsql-hackers mailing list

Re: [HACKERS] Proposal for Merge Join for Non '=' Operators

2014-04-09 Dilip kumar dilip.ku...@huawei.com: I would like to propose a New merge join algorithm for optimizing non ‘=’ operators. (‘’, ‘=’, ‘’, ‘=’) Do you have a real-world example use case of such joins, to offset the extra planner time that will likely have to be paid (even for queries

Re: [HACKERS] [PATCH] Negative Transition Aggregate Functions (WIP)

2013/12/15 David Rowley dgrowle...@gmail.com: I've been working on speeding up aggregate functions when used in the context of a window's with non fixed frame heads. 1. Fully implement negative transition functions for SUM and AVG. I would like to mention that this functionality is also

Re: [HACKERS] Fast insertion indexes: why no developments

2013/11/12 Claudio Freire klaussfre...@gmail.com: On Tue, Nov 12, 2013 at 6:41 PM, Nicolas Barbier nicolas.barb...@gmail.com wrote: (Note that K B-trees can be merged by simply scanning all of them concurrently, and merging them just like a merge sort merges runs. Also, all B-trees except

Re: [HACKERS] Fast insertion indexes: why no developments

2013/11/12 Simon Riggs si...@2ndquadrant.com: On 12 November 2013 21:41, Nicolas Barbier nicolas.barb...@gmail.com wrote: Look-up speed is as follows: Each look-up must look through all B-trees. That can be optimised by using a min max approach, so we need only look at sub-trees that may

Re: [HACKERS] Fast insertion indexes: why no developments

2013/11/2 Simon Riggs si...@2ndquadrant.com: On 29 October 2013 16:10, Peter Geoghegan p...@heroku.com wrote: Presumably someone will get around to implementing a btree index insertion buffer one day. I think that would be a particularly compelling optimization for us, because we could avoid

Re: [HACKERS] Fast insertion indexes: why no developments

2013/11/12 Nicolas Barbier nicolas.barb...@gmail.com: In conclusion, use a “B-forest” when: * The index entries are small (large fan-out). * The insertion throughput is high. * It’s OK for look-ups to be slow. * Extra points when the storage medium has high seek times. Oops, forgot

Re: [HACKERS] Chinese in Postgres

[ Could you please trim your citations, i.e., please don’t top-post: https://en.wikipedia.org/wiki/Posting_style#Top-posting ] 2013/8/16 Francesco ciifrance...@tiscali.it: Thanks for your answer. Yes, the client is also UTF8: MyDB=# show client_encoding; client_encoding -

Re: [HACKERS] Documentation/help for materialized and recursive views

2013/6/27 Magnus Hagander mag...@hagander.net: Is there a particular reason why CREATE RECURSIVE VIEW is part of the help for CREATE VIEW, but CREATE MATERIALIZED VIEW doesn't show up there? I realize the technical reason (they're different man pages, and that also controls what's in \h in

Re: [HACKERS] Hash partitioning.

2013/6/26 Heikki Linnakangas hlinnakan...@vmware.com: On 26.06.2013 16:41, Yuri Levinsky wrote: Heikki, As far as I understand the height of the btree will affect the number of I/Os necessary. The height of the tree does not increase linearly with the number of records. Now let's compare

Re: [HACKERS] Hash partitioning.

2013/6/27 Nicolas Barbier nicolas.barb...@gmail.com: When each index requires one extra I/O (because each index is one level taller), that is 50 extra I/Os. In the partitioned case, each index would require the normal smaller amount of I/Os. [..] Using those other indexes (both for look-ups

Re: [HACKERS] Hash partitioning.

2013/6/27 Markus Wanner mar...@bluegap.ch: On 06/27/2013 11:12 AM, Nicolas Barbier wrote: Imagine that there are a lot of indexes, e.g., 50. Although a lookup (walking one index) is equally fast, an insertion must update al 50 indexes. When each index requires one extra I/O (because each

Re: [HACKERS] refresh materialized view concurrently

2013/6/17 Heikki Linnakangas hlinnakan...@vmware.com: +errmsg(concurrent refresh requires a unique index on just columns for all rows of the materialized view))); Maybe my english is failing me here, but I don’t understand the “just” part. Nicolas -- A.

Re: [HACKERS] counting algorithm for incremental matview maintenance

2013/5/17 Kevin Grittner kgri...@ymail.com: During calculation of the deltas to apply to the matviews, it must be possible to query the referenced tables from the perspective of both the before and after versions of the data. [..] I don't think the process applying the deltas needs to do

Re: [HACKERS] counting algorithm for incremental matview maintenance

2013/5/17 Kevin Grittner kgri...@ymail.com: Nicolas Barbier nicolas.barb...@gmail.com wrote: Note that the basic count algorithm assumes real-serializable transactions for correctness. Example: [..] Good point. It might be hard to detect when this type of race condition exists, since

Re: [HACKERS] counting algorithm for incremental matview maintenance

2013/5/17 Claudio Freire klaussfre...@gmail.com: On Fri, May 17, 2013 at 4:25 PM, Kevin Grittner kgri...@ymail.com wrote: (3) The count algorithm must be implemented in a way that understands MVCC internals: Reading the base tables must be done using a technique that reads all rows (i.e.,

Re: [HACKERS] Drastic performance loss in assert-enabled build in HEAD

2013/4/5 Noah Misch n...@leadboat.com: On Thu, Apr 04, 2013 at 12:28:01PM +0200, Nicolas Barbier wrote: +1. Having unlogged matviews without having incremental updates yet, isn't super useful anyway. I would have surmised the opposite: since an unlogged MV requires a full refresh

Re: [HACKERS] Drastic performance loss in assert-enabled build in HEAD

2013/4/3 Tom Lane t...@sss.pgh.pa.us: Kevin Grittner kgri...@ymail.com writes: To be honest, I don't think I've personally seen a single use case for matviews where they could be used if you couldn't count on an error if attempting to use them without the contents reflecting a

Re: [HACKERS] REFRESH MATERIALIZED VIEW locklevel

2013/3/8 Andres Freund and...@anarazel.de: On 2013-03-07 15:21:35 -0800, Josh Berkus wrote: This limitation is in no way crippling for this feature, or even a major detraction. I still intend to promote the heck out of this feature. Thats scaring me. Because the current state of the

Re: [HACKERS] Materialized views WIP patch

2013/3/5 Kevin Grittner kgri...@ymail.com: Perhaps it would be worth looking for anything in the patch that you think might be painting us into a corner where it would be hard to do all the other cool things. While it's late enough in the process that changing anything like that which you

Re: [HACKERS] Materialized views WIP patch

2013/3/5 Robert Haas robertmh...@gmail.com: All that having been said, it's hard for me to imagine that anyone really cares about any of this until we have an incremental update feature, which right now we don't. Actually, I'm betting that's going to be significantly harder than

Re: [HACKERS] Materialized views WIP patch

2013/3/5 Kevin Grittner kgri...@ymail.com: Exactly. I predict that we will eventually have some special sort of trigger for maintaining MVs based on base table changes to handle the ones that are just too expensive (in developer time or run time) to fully automate. But there is a lot of

Re: [HACKERS] Materialized views WIP patch

2013/3/3 Kevin Grittner kgri...@ymail.com: Rewriting queries using expressions which match the MV's query to pull from the MV instead of the underlying tables is the exception. While that is a sexy feature, and I'm sure one can construct examples where it helps performance, it seems to me

Re: [HACKERS] Materialized views WIP patch

2013/3/3 Kevin Grittner kgri...@ymail.com: Nicolas Barbier nicolas.barb...@gmail.com wrote: I think that automatically using materialized views even when the query doesn’t mention them directly, is akin to automatically using indexes without having to mention them in the query. That way

Re: [HACKERS] Materialized views WIP patch

2013/2/19 Robert Haas robertmh...@gmail.com: In the department of crazy ideas, what about having pg_dump NEVER refresh ANY materialized views? It's true that the job of pg_dump and pg_restore is to put the new database in the same state that the old database was in, but I think you could

Re: [HACKERS] [PATCH] Docs: Make notes on sequences and rollback more obvious

2012/8/7 Kevin Grittner kevin.gritt...@wicourts.gov: I also think it's a problem that one can get through the entire Concurrency Control chapter (mvcc.sgml) without a clue that sequences aren't transactional. It is possible to say that they *are* transactional when considering the following

Re: [HACKERS] Missing optimization when filters are applied after window functions

2012/5/17 Volker Grabsch v...@notjusthosting.com: Also, is there any chance to include a (simple) attempt of such an optimiztation into PostgreSQL-9.2 beta, or is this only a possible topic for 9.3 and later? For 9.2, you’re about 4 months late :-). The last commitfest was in Januari:

Re: [HACKERS] Temporary tables under hot standby

2012/4/25 Simon Riggs si...@2ndquadrant.com: On Wed, Apr 25, 2012 at 4:55 AM, Noah Misch n...@leadboat.com wrote: I do not see a clean behind-the-scenes fix for points 1, 4 and 5.  We can resolve those by adding a new variety of temporary table, one coincidentally matching the SQL standard's

Re: [HACKERS] Temporary tables under hot standby

2012/4/25 Simon Riggs si...@2ndquadrant.com: So you are saying it is OK to not be able to *create* them on HS, just *use* pre-defined tables? That's almost useless IMHO. Applications expect to be able to do this all in the same transaction on one session CREATE TEMP TABLE x;  ...DML

Re: [HACKERS] Temporary tables under hot standby

2012/4/25 Nicolas Barbier nicolas.barb...@gmail.com: is the reason of existence for the PG-like temporary transactions. s/transactions/tables/ Nicolas -- A. Because it breaks the logical sequence of discussion. Q. Why is top posting bad? -- Sent via pgsql-hackers mailing list (pgsql

Re: [HACKERS] VACUUM ANALYZE is faster than ANALYZE?

2012/2/22 Robert Haas robertmh...@gmail.com: On Tue, Feb 21, 2012 at 9:00 AM, Pavel Stehule pavel.steh...@gmail.com wrote: I had to reply to query about usage VACUUM ANALYZE or ANALYZE. I expected so ANALYZE should be faster then VACUUM ANALYZE. But is not true. Why? I'm pretty sure

Re: [HACKERS] 16-bit page checksums for 9.2

2012/1/4 Simon Riggs si...@2ndquadrant.com: On Wed, Jan 4, 2012 at 9:20 AM, Andres Freund and...@anarazel.de wrote: I wonder if CRC32c wouldn't be a good alternative given more and more cpus (its in SSE 4.2) support calculating it in silicon. We're trying to get something that fits in

Re: [HACKERS] 16-bit page checksums for 9.2

2011/12/30 Ants Aasma ants.aa...@eesti.ee: On Thu, Dec 29, 2011 at 6:44 PM, Kevin Grittner kevin.gritt...@wicourts.gov wrote: positives.  To get this right for a checksum in the page header, double-write would need to be used for all cases where full_page_writes now are used (i.e., the

Re: [HACKERS] Inlining comparators as a performance optimisation

2011/12/5 Tom Lane t...@sss.pgh.pa.us: What is bothering me is that this approach is going to cause substantial bloat of the executable code, and such bloat has got distributed costs, which we don't have any really good way to measure but for sure micro-benchmarks addressing only sort speed

Re: [HACKERS] Patch Review: Bugfix for XPATH() if text or attribute nodes are selected

2011/6/29, Florian Pflug f...@phlo.org: Secondly, there is little point in having an type XML if we don't actually ensure that values of that type can only contain well-formed XML. +1. The fact that XPATH() must return a type that cannot depend on the given expression (even if it is a

Re: [HACKERS] SSI modularity questions

2011/6/28, Robert Haas robertmh...@gmail.com: You know, it just occurred to me while reading this email that you're using the term predicate lock in a way that is totally different from what I learned in school. What I was taught is that the word predicate in predicate lock is like the word

Re: [HACKERS] XPATH evaluation

2011/6/17, Andrew Dunstan and...@dunslane.net: On 06/17/2011 10:55 AM, Radosław Smogura wrote: XML canonization preservs whitespaces, if I remember well, I think there is example. In any case if I will store image in XML (I've seen this), preservation of white spaces and new lines is

Re: [HACKERS] XPATH evaluation

2011/6/17, Andrew Dunstan and...@dunslane.net: On 06/17/2011 11:29 AM, Nicolas Barbier wrote: CDATA sections are just syntactic sugar (a form of escaping): Yeah. OTOH doesn't an empty CDATA section force a child element, where a pure empty element does not? Wow, some Googling around shows

Re: [HACKERS] Cube Index Size

2011/5/30, Nick Raj nickrajj...@gmail.com: 3. When tuples are 5 lakh For the benefit of the others: 5 lakh seems to mean 500,000. URL:http://en.wikipedia.org/wiki/Lakh Nicolas -- A. Because it breaks the logical sequence of discussion. Q. Why is top posting bad? -- Sent via pgsql-hackers

Re: [HACKERS] Please test peer (socket ident) auth on *BSD

2011/5/31, Tom Lane t...@sss.pgh.pa.us: Unless maybe there's a kFreeBSD-like project out there with NetBSD as the kernel?) There used to be an attempt by Debian (called GNU/NetBSD), but that has since long been abandoned. I don't know of any other similar projects.

Re: [HACKERS] FailedAssertion

[ Forgot the list. ] 2011/5/12, Yves Weißig weis...@rbg.informatik.tu-darmstadt.de: I'm currently debugging my developed AM and are running into this problem: TRAP: FailedAssertion(!(((bool) (((void*)(tid) != ((void *)0)) ((tid)-ip_posid != 0, File: indexam.c, Line: 488) Can anybody

Re: [HACKERS] the big picture for index-only scans

2011/5/11, Bruce Momjian br...@momjian.us: FYI, because the visibility map is only one _bit_ per page, it is 8000 * 8 or 64k times smaller than the heap, e.g. one 8k page covers 64MB of heap pages. Actually, that would be one 8kB block covers 512MB of heap: 1 block of visibility map (8kB) =

Re: [HACKERS] Predicate locking

2011/4/27 Vlad Arkhipov arhi...@dc.baikal.ru: I'm currently need predicate locking in the project, so there are two ways to get it by now: implement it by creating special database records to lock with SELECT FOR UPDATE or wait while they will be implemented in Postgres core. Is there

Re: [HACKERS] PostgreSQL sourcecode

2011/3/30 aaronenabs aaronen...@btconnect.com: Hi all i have been trying to compile the sourcecode for postgresql but keep getting an error message when running it in cygwin. it states: dllwrap: gcc exited with status 1 make[3]: *** [cygpq.dll] Error 1 make[3]: Leaving directory

Re: [HACKERS] PostgreSQL sourcecode

2011/3/30 aaronenabs aaronen...@btconnect.com: Can you alos advise how i change the the HeapTupleSatisfiesVisibility() to true within the source code: [..] #define HeapTupleSatisfiesVisibility(tuple, snapshot, buffer) \        ((*(snapshot)-satisfies) ((tuple)-t_data, snapshot, buffer)) As

Re: [HACKERS] I am confused after reading codes of PostgreSQL three week

2011/3/20 hom obsidian...@gmail.com: I trace into scan.c because I want to known how the paser tree is built and I debug the source step by step. I suggest you learn how flex/bison work first. The contents of the *.c files generated by flex/bison are not generally supposed to be interpreted by

Re: [HACKERS] Fwd: Multi-language glossary of PostgreSQL terms.

2011/3/15 Dmitriy Igrishin dmit...@gmail.com: Oleg Bartunov and I intend to begin the project of translation of PostgreSQL documentation on Russian. Just to make sure that you are aware of a recently-started effort to create a German translation, by Susanne and others:

Re: [HACKERS] Read uncommitted ever possible?

2011/3/10 Jesper Krogh jes...@krogh.cc: On 2011-03-10 19:25, Bruce Momjian wrote: Sure, anyone can add text to that wiki;  you create a community account here:        http://www.postgresql.org/community/signup Suggestion: Add this url to the login box on the wiki. +1, Adrian von Bidder

Re: [HACKERS] What happens If a table changes during a query/procedure execution

2011/3/9 Vlad Arkhipov arhi...@dc.baikal.ru: Let there are two transactions that were created with read commited isolation level. In the first one we're executing a SELECT query: SELECT * FROM t UNION ALL SELECT * FROM t; In the second transaction we're modifying the same table: INSERT INTO

Re: [HACKERS] What happens If a table changes during a query/procedure execution

2011/3/9 Nicolas Barbier nicolas.barb...@gmail.com: Note that the standard defines things that must never happen in the case of READ COMMITTED, it does not specify that one *must* be able to see the stuff as committed by previous transactions, for example. Hmm, make that stuff as committed

Re: [HACKERS] Native XML

2011/3/1 Andrew Dunstan and...@dunslane.net: I think hierarchical data really only scratches the surface of the problem. It would be nice to be able to specify all sorts of context for searches:   * foo after bar   * foo near bar   * foo and bar in the same paragraph   * foo as a

Re: [HACKERS] Sorting. When?

[ Please don't top-post. URL:http://en.wikipedia.org/wiki/Posting_style ] 2011/2/11 mac_man2...@yahoo.it mac_man2...@yahoo.it: So, is there any precise way to discover when sorting is invoked? EXPLAIN shows how a query would be executed; explicit sorts should be mostly obvious.

Re: [HACKERS] Sorting. When?

2011/2/11 Kevin Grittner kevin.gritt...@wicourts.gov: mac_man2...@yahoo.it mac_man2...@yahoo.it wrote: I need to know, from an algorithmic point of view, in which cases sorting is invoked. [..] Are your really looking to categorize the types of queries where sorting is *invoked*, or the

Re: [HACKERS] Sorting. When?

2011/2/10 mac_man2...@yahoo.it mac_man2...@yahoo.it: Which operations invoke the sorting algorithms implemented in the sorting module (tuplesort.c) ? Of course, one of those operations invoking sorting is the ORDER BY clause and the DISTINCT too. Moreover, the Merge Join should be

Re: [HACKERS] query execution question

2011/2/9 amit sehas cu...@yahoo.com: Lets say that the cost based optimizer determines that the order of the joins should be   T1.a=T2.b followed by T2.c = T3.d followed by T3.e = T4.f the question we have is during query execution are the joins evaluated completely one by one in that order,

Re: [HACKERS] bad links in messages from commits

2011/2/1 Magnus Hagander mag...@hagander.net: On Tue, Feb 1, 2011 at 07:56, Pavel Stehule pavel.steh...@gmail.com wrote: 2011/2/1 Magnus Hagander mag...@hagander.net: On Tue, Feb 1, 2011 at 05:53, Pavel Stehule pavel.steh...@gmail.com wrote: Hello There are broken links inside messages

Re: [HACKERS] Re: In pg_test_fsync, use K(1024) rather than k(1000) for write size units.

2011/1/27 Bruce Momjian br...@momjian.us: Bruce Momjian wrote: Peter Eisentraut wrote: We use small k in postgresql.conf, so pg_test_fsync should use the same.  Using kB would be more accurate in any case. OK, done with the attached applied patch. FYI, I had used 'k' because this page

Re: [HACKERS] SSI and Hot Standby

2011/1/21 Anssi Kääriäinen anssi.kaariai...@thl.fi: Sorry for bothering all of you, but I just don't get this. What if T2 rolls back instead of committing? Then the snapshot of T3 would have been valid, right? Now, for the snapshot of T3 it doesn't matter if T2 commits or if it doesn't,

Re: [HACKERS] SSI and Hot Standby

2011/1/21 Robert Haas robertmh...@gmail.com: On Fri, Jan 21, 2011 at 8:05 AM, Nicolas Barbier nicolas.barb...@gmail.com wrote: 2011/1/21 Anssi Kääriäinen anssi.kaariai...@thl.fi: Sorry for bothering all of you, but I just don't get this. What if T2 rolls back instead of committing

Re: [HACKERS] ToDo List Item - System Table Index Clustering

2011/1/16 Simone Aiken sai...@ulfheim.net:        is there a way to make queries on the system tables show me what        is actually there when I'm poking around?  So for example:                Select * from pg_type limit 1;        tells me that the typoutput is 'boolout'.  An english

Re: [HACKERS] join functions

2011/1/5 Zotov zo...@oe-it.ru: Why doesn`t work this query? select table1.field1, func1.field2 from table1 left outer join func1(table1.field1) on true where func1.field3 in (20, 100); If i have other than LEFT OUTER JOIN I can understand why   ERROR:  invalid reference to FROM-clause entry

Re: [HACKERS] proposal : cross-column stats

2010/12/24 Florian Pflug f...@phlo.org: On Dec23, 2010, at 20:39 , Tomas Vondra wrote:   I guess we could use the highest possible value (equal to the number   of tuples) - according to wiki you need about 10 bits per element   with 1% error, i.e. about 10MB of memory for each million of  

Re: [HACKERS] Extensions, patch v20 (bitrot fixes)

2010/12/20 Martijn van Oosterhout klep...@svana.org: On Mon, Dec 20, 2010 at 09:03:56AM +0900, Itagaki Takahiro wrote: UTF-8 is not a superset of all encodings. I think you mean Unicode is not a superset of all character sets. I've heard this before but never found what's missing. [citation

Re: [HACKERS] Extensions, patch v18 (merge against master, bitrot-only-fixes)

2010/12/16 Robert Haas robertmh...@gmail.com: On Thu, Dec 16, 2010 at 7:49 AM, Dimitri Fontaine dimi...@2ndquadrant.fr wrote: Please note that the SQL scripts seem to be encoded in latin9. Seems like an odd choice.  Why not UTF-8? Latin 9 = ISO 8859-15 = a more modern version of Latin 1

Re: [HACKERS] mvcc DML on the same row

2010/12/15 matteo durighetto desmodem...@gmail.com: But why we need all these versions of the same row on table, if for rollback we need only the original row X (X0) ? And the previous value of row X during the execution of a statement (because statements don't see their own changes, think

Re: [HACKERS] ALTER TABLE ... ADD FOREIGN KEY ... NOT ENFORCED

2010/12/13 Tom Lane t...@sss.pgh.pa.us: But allow me to harbor doubts that they really intend to allow someone to force a constraint to be considered valid without any verification. Table constraints are either enforced or not enforced. Domain constraints and assertions are always enforced.,

Re: [HACKERS] MULTISET and additional functions for ARRAY

2010/11/11 David E. Wheeler da...@kineticode.com: On Nov 11, 2010, at 10:05 AM, Tom Lane wrote: So are you planning to implement multisets? It's a feature I'd love to see What actual functionality does it buy?  AFAICT from Itagaki-san's description, it's an array only you ignore the

Re: [HACKERS] Hash support for arrays

2010/11/2 Kenneth Marshall k...@rice.edu: Given that our hash implimentation mixes the input data well (It does. I tested it.) then a simple rotate-and-xor method is all that should be needed to maintain all of the needed information. The original hash function has done the heavy lifting in

Re: [HACKERS] client socket TIME_WAIT state after PQfinish

2010/9/27 Guillaume Du Pasquier guillaume.dupasqu...@sensometrix.ch: In both cases, the client socket (pgadmin or my program) remains in TIME_WAIT state. I have used wireshark to sniff the TCP protocol. We have at the end of a connection: Client      Server    ---    FIN,ACK 

Re: [HACKERS] client socket TIME_WAIT state after PQfinish

2010/9/27 Robert Haas robertmh...@gmail.com: On Mon, Sep 27, 2010 at 12:56 PM, Guillaume Du Pasquier guillaume.dupasqu...@sensometrix.ch wrote: Our client runs on the same machine as the postgresql server. Would it be possible to use PF_UNIX sockets ? Yeah, actually that's the default, if

Re: [HACKERS] Serializable Snapshot Isolation

[ Forgot the list, resending. ] 2010/9/25 Greg Stark gsst...@mit.edu: On Thu, Sep 23, 2010 at 4:08 PM, Kevin Grittner kevin.gritt...@wicourts.gov wrote: One place I'm particularly interested in using such a feature is in pg_dump. Without it we have the choice of using a SERIALIZABLE

Re: [HACKERS] Deadlock bug

2010/8/25 Simon Riggs si...@2ndquadrant.com: referenced meaning by an RI constraint, which only ever refers to PKs in other tables. FK constraints can also point to non-PK UNIQUE columns. Nicolas -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your

Re: [HACKERS] Deadlock bug

2010/8/25 Simon Riggs si...@2ndquadrant.com: On Wed, 2010-08-25 at 16:14 +0200, Nicolas Barbier wrote: 2010/8/25 Simon Riggs si...@2ndquadrant.com: referenced meaning by an RI constraint, which only ever refers to PKs in other tables. FK constraints can also point to non-PK UNIQUE

Re: [HACKERS] About debug two versions of postgresql in eclipse

2010/8/24 Pei He hepeim...@gmail.com: I want to run two different versions of postgresql-8.2.5 under eclipse. But, it requires me to change PGDATA and LD_LIBRARY_PATH to switch. Moreover, to let eclipse know the changes, I need to modify .profile under my home folder, and log out and log

Re: [HACKERS] assertions and constraint triggers

2010/8/11 Marko Tiikkaja marko.tiikk...@cs.helsinki.fi: On 8/11/10 8:31 AM +0300, Peter Eisentraut wrote: Thinking about SQL assertions (check constraints that are independent of one particular table), do you think it would be reasonable to implement those on top of constraint triggers?  On

Re: [HACKERS] Online backup cause boot failure, anyone know why?

2010/8/5 Richard husttrip...@vip.sina.com: All jods are done by client code, not manually. What is a jod? I still did't not understand what you said. What break what? The fact that you replaced CHECKPOINT_WAIT with CHECKPOINT_IMMEDIATE is the cause of your problem. You broke the correctness

Re: [HACKERS] GSoC - code of implementation of materialized views

2010/6/30 Robert Haas robertmh...@gmail.com: By the way, does the SQL standard say anything about materialized views? AFAIK, nope. Probably for the same reason that indexes are not mentioned by the standard: both are only performance enhancements, and one could easily imagine future SQL

Re: [HACKERS] Exposing the Xact commit order to the user

2010/5/25 Dan Ports d...@csail.mit.edu: On Mon, May 24, 2010 at 10:24:07AM -0500, Kevin Grittner wrote: Replicating or recreating the whole predicate locking and conflict detection on slaves is not feasible for performance reasons. (I won't elaborate unless someone feels that's not

Re: [HACKERS] Exposing the Xact commit order to the user

2010/5/25 Dan Ports d...@csail.mit.edu: On Tue, May 25, 2010 at 02:00:42PM +0200, Nicolas Barbier wrote: I don't understand the problem. According to me, in the context of SSI, a read-only slave can just map SERIALIZABLE to the technical implementation of REPEATABLE READ (i.e., the currently

Re: [HACKERS] Exposing the Xact commit order to the user

2010/5/25 Florian Pflug f...@phlo.org: On May 25, 2010, at 20:18 , Dan Ports wrote: T3, which is a read-only transaction, sees the incremented date and an empty list of receipts. But T1 later commits a new entry in the receipts table with the old date. No serializable ordering allows this.

Re: [HACKERS] Exposing the Xact commit order to the user

2010/5/25 Florian Pflug f...@phlo.org: Hm, but for there to be an actual problem (and not a false positive), an actual dangerous circle has to exist in the dependency graph. The existence of a dangerous structure is just a necessary (but not sufficient) and easily checked-for condition for

Re: [HACKERS] Row-level Locks SERIALIZABLE transactions, postgres vs. Oracle

2010/5/14 Greg Stark gsst...@mit.edu: On Thu, May 13, 2010 at 10:25 PM, Florian Pflug f...@phlo.org wrote: C1: BEGIN C1: SELECT * FROM t WHERE id = 1 FOR UPDATE C2: BEGIN C2: SET TRANSACTION ISOLATION LEVEL SERIALIZABLE C2: SELECT * FROM t -- Take snapshot before C1 commits C1: COMMIT

Re: [HACKERS] Partitioning/inherited tables vs FKs

2010/5/11 Marko Tiikkaja marko.tiikk...@cs.helsinki.fi: On 2010-05-11 14:29 +0200, Robert Haas wrote: On Tue, May 11, 2010 at 2:16 AM, Dmitry Fefelov fo...@ac-sw.com wrote: The referential integrity triggers contain some extra magic that isn't easily simulatable in userland, and that is

Re: [HACKERS] Partitioning/inherited tables vs FKs

2010/5/11 Marko Tiikkaja marko.tiikk...@cs.helsinki.fi: This is getting way off topic, but: On 5/11/10 3:55 PM +0300, Nicolas Barbier wrote: T2  SELECT i FROM a WHERE i = 1 FOR SHARE; -- Lock a with i = 1 FOR SHARE.  i ---  1 (1 Zeile) T2  SELECT a_id FROM b WHERE a_id = 1; -- Check

Re: [HACKERS] system table/view and sequence

2010/4/7 Olivier Baheux olivierbah...@gmail.com: i'm trying to find where are stored sequence definition (increment,minvalue,maxvalue,start,cache) in system tables. Atm I found everything exept sequence. It's in the sequence itself (which can be accessed like a table). The fact that this

Re: [HACKERS] WIP: preloading of ispell dictionary

2010/3/23 Pavel Stehule pavel.steh...@gmail.com: 2010/3/23 Takahiro Itagaki itagaki.takah...@oss.ntt.co.jp: The fundamental issue seems to be in the slow initialization of dictionaries. If so, how about adding a pre-complile tool to convert a dictionary into a binary file, and each backend

  1   2   >