Re: [sqlite] Implementing a statement cache

On 16/12/2019 13:38, carsten.muencheberg wrote: > I am working on a generic cache for prepared statements I really wish SQLite had a statement cache behind the scenes, so that devs don't have to keep re-implementing a statement cache. I would be delighted to delete the statement cache code in

[sqlite] sqlite3_complete and comments

I use sqlite3_complete in my shell in order to determine when a complete statement has been input and can be run. (Otherwise a continuation "sqlite> " prompt is shown.) If the line entered is: -- hello Then the sqlite shell does not issue a continuation and "executes" the text. However

Re: [sqlite] Best practices for forward conversion of database formats?

On 30/08/18 09:51, Randall Smith wrote: > is how to convert existing data from one DB format to another, given some > arbitrary set of changes in the database schema in the interim. I use SQLite's user pragma. It starts at zero. https://sqlite.org/pragma.html#pragma_user_version My code ends

Re: [sqlite] Size of the SQLite library

On 06/06/18 09:24, Bob Friesenhahn wrote: > A local tool which makes it easy to configure sqlite from local files > sounds useful ... It already exists. It is what the SQLite team uses to produce the amalgamations etc, and is part of the SQLite code base. > but depending on a "web site"

Re: [sqlite] Size of the SQLite library

On 05/06/18 15:07, Warren Young wrote: > All right, so include [multi-component source control and build process] ... I'm still not sure what point you are trying to make. Yes *you* can do that. Should *every* SQLite user who wants non-default options *have* to go through a similar amount of

Re: [sqlite] Size of the SQLite library

On 01/06/18 13:46, Warren Young wrote: > Your jQuery example later on doesn’t much apply here, for several reasons: Note that I was showing how the site let you choose whatever features you want, and then gave you a download matching exactly that. > 1. JavaScript is a dynamic language, while C

Re: [sqlite] Size of the SQLite library

On 31/05/18 10:15, Richard Hipp wrote: > Size is still important. But having useful features is important too. > I'm continuing to work to find the right balance between these > competing goals. A pattern used in other projects is to have standard downloads, as well as custom ones. With the

Re: [sqlite] SQLite between forks

On 26/03/18 13:30, Simone Mosciatti wrote: > However I fail to see how this can be a problem for purely in-memory > database. When a process forks, only the thread that called fork is kept in the new child process. Also note that semaphores (and locks in general) are left in the same state as at

Re: [sqlite] Linux top command and sqlite

On 21/02/17 10:22, Kevin O'Gorman wrote: > Some of my stuff takes a while to run, and I like to keep tabs on it. > Right now, I'm running one of those, and the Linux top command shows > extremely small CPU usage, and a status ("S" column) of "D" which the man > page defines as "uninterruptable

Re: [sqlite] Beginning of release testing for version 3.17.0

On 08/02/17 11:41, Cecil Westerhof wrote: > ​OK, glad to help. What should I do? It is nicest if whatever software/tools you already have also has some sort of testing (ideally automated, but a manual checklist works too). Then run the testing with the existing version of SQLite, and repeat with

Re: [sqlite] [sqlite-dev] Beginning of release testing for version 3.17.0

On 07/02/17 08:56, James K. Lowden wrote: > I must be having a bad day. Both Google and cscope fail to turn up any > reference to NULL_TRIM. What are you referring to, and what does it > do? I started at the changelog posted at the beginning of this thread which makes everything

Re: [sqlite] Why this query plan?

On 11/01/17 16:49, Richard Hipp wrote: > For years I have threatened to make it a feature of SQLite that it > really does output the rows in some random order if you omit the ORDER > BY clause - specifically to expose the common bug of omitting the > ORDER BY clause when the order matters. And

Re: [sqlite] Speaking of Ruby & Sqlite...a weekend oddity for someone

On 09/12/16 14:09, Don V Nielsen wrote: > However, > it fails using the sqlite3 gem. The specific exception is "in > 'initialize': near "with": syntax error That will be a SQLite version issue - older SQLite's didn't support with. While you made some effort around versions, whatever is happening

Re: [sqlite] Interactive shell tab completion for queries

On 09/12/16 14:18, Dominique Devienne wrote: > How did you implement completion in APSW? [Long answer] at bottom. You get a callback from the readline or equivalent library when completion is needed, with it expecting you to provide 0 or more completions. You can get the input line, the current

Re: [sqlite] Interactive shell tab completion for queries

On 09/12/16 14:18, Dominique Devienne wrote: > How did you implement completion in APSW? [Long answer] at bottom. You get a callback from the readline or equivalent library when completion is needed, with it expecting you to provide 0 or more completions. You can get the input line, the current

Re: [sqlite] Interactive shell tab completion for queries

On 09/12/16 06:29, daveparr...@tutanota.com wrote: > I'm writing to ask if it is possible to have table completion for SQL queries > in the interactive shell? The APSW shell (compatible with the SQLite one) has completion and colouring: https://rogerbinns.github.io/apsw/shell.html

Re: [sqlite] Scope of sqlite3_update_hook?

On 01/12/16 16:51, Jens Alfke wrote: > If so, then does that include connections in other OS processes? (I'm looking > for a way to detect this.) You can't get a callback when other processes change the database for many reasons. However it is possible to detect if the database has changed:

Re: [sqlite] Pragma to flag unknown pragma?

On 22/11/16 16:08, Scott Hess wrote: > Something like "PRAGMA pedantic_pragma = on" would be much slicker. SQLite lets you do "natrual" joins too. I'd argue all this falls under a lint mode that helps conscientious developers make sure everything is working correctly under the hood.

Re: [sqlite] problem or not problem around 'begin immediate'

On 25/11/16 12:02, Richard Hipp wrote: > Rather, I presume > that Python has recently started using the sqlite3_stmt_readonly() > interface in a new way. The bigger picture may be helpful. There is a third party module developed under the name "pysqlite" which has a long and storied history. At

Re: [sqlite] I keep getting seg faults building my database using python sqlite3

On 19/11/16 08:08, Kevin O'Gorman wrote: > System with problems: Running Xubuntu Linux 16.04.1, Python 3.5.2. [...] > System without this problem: Running Ubuntu Linux 14.04.5, Python 3.4.3. You are good on Python versions then. My remaining recommendation is to make the process that does SQLite

Re: [sqlite] I keep getting seg faults building my database using python sqlite3

On 18/11/16 15:19, James K. Lowden wrote: > Good catch, Roger. It's a liability, but I slightly disagree with your > characterization. > >> - Running any Python code (destructors can be called which then run in >> the parent and child) > > Yes, if those destructors affect shared resources.

Re: [sqlite] I keep getting seg faults building my database using python sqlite3

On 18/11/16 08:55, Kevin O'Gorman wrote: >> I am not. All of the python code is a single thread. The closest I come > is a few times where I use subprocess.Popen to create what amounts to a > pipeline, and one place where I start a number of copies of a C program in > parallel, but each is a

Re: [sqlite] I keep getting seg faults building my database using python sqlite3

On 17/11/16 19:14, Kevin O'Gorman wrote: > SO: I need help bifurcating this problem. For instance, how can I tell if > the fault lies in SQLite, or in python? Or even in the hardware, given that > the time to failure is so variable? Are you using threads, threading related settings etc in any

Re: [sqlite] Database malformed after 6000 Inserts?

On 04/10/16 03:11, Werner Kleiner wrote: > ... after 6000 records. > > Is there a limitation with huge inserts? While there may be "Lite" in the name, SQLite copes very well with "huge" stuff. That means many many gigabytes in database sizes, many many millions of rows, up to 2GB per row etc.

Re: [sqlite] client app crashes frequently and points sqlite DLL

On 02/09/16 03:11, Frantz FISCHER wrote: > I'm almost out of tracks to follow. Any idea on what I could check next? Are you using threads? Note that in this case whatever is happening is making SQLite be the victim. The cause is almost certainly some other C code in your process. Roger

Re: [sqlite] The upcoming "pi" release of SQLite

On 23/07/16 08:16, Richard Hipp wrote: > Draft change log: https://www.sqlite.org/draft/releaselog/3_14_0.html Please please don't make the new trace/profile API expand the SQL by default. There are two problems with expanding by default: - The text no longer matches what the programmer had in

Re: [sqlite] builtin functions and strings with embedded nul characters

On 01/07/16 05:04, Simon Slavin wrote: > On 1 Jul 2016, at 10:18am, Rob Golsteijn wrote: > >> For the tests below I assumed that the intention is that a string ends at >> the first embedded nul character. > > I'm not sure that this is the intent. > > The idea that

Re: [sqlite] page checksums (was Re: Bad db feature request)

On 29/06/16 19:13, Scott Robison wrote: > Given the nature of VFS, it is trivial* for anyone to create a module to > provide this very functionality. So you can write it yourself! > > *Not really trivial, but probably not horribly difficult either. VFS is one way you can't reasonably do it. The

Re: [sqlite] Bad db feature request

On 29/06/16 09:45, Drago, William @ CSG - NARDA-MITEQ wrote: > Aren't there things like that [checksums] already built in to the hard disk > controllers (CRC, Reed Solomon, etc.)? They are at a different level and can only detect issues in what they see. For example SQLite can create a page of

Re: [sqlite] Bad db feature request

On 29/06/16 07:51, Dominique Devienne wrote: > I wish for the day SQLite has page checksums to detect any such random > corruption. Agreed. The SQLite team rejected doing so: http://www.sqlite.org/src/tktview/72b01a982a84f64d4284 > Yes, I know, it's a format change, and will likely slow

Re: [sqlite] WAL: no schema after close

On 03/06/16 18:56, Gelin Yan wrote: >APSW is great, I have used it for years. I want to know whether it > support pypy. I have switched to pypy for a while. APSW at its heart is very much a CPython extension and uses that C API to bind to SQLite. I did port APSW to pypy a few years ago which

Re: [sqlite] WAL: no schema after close

On 03/06/16 08:28, Максим Дементьев wrote: > Thank you, I'll try this "yet another python SQLite wrapper" in the near > future, it looks interesting. Disclosure: I am the author of APSW. I recommend looking at the page showing the differences between APSW and pysqlite:

[sqlite] Podcast with Dr Hipp: SQLite history, success and funding

On 16/05/16 10:36, Richard Hipp wrote: > I find that when you are thinking long-term, it changes your > perspective on which patches land on trunk. In addition to your (plural) fantastic work, saying yes/no is probably by far the most important piece. There are constant calls for things to be

[sqlite] 2 different SQLite versions inside the same process space

On 10/05/16 23:43, Dominique Devienne wrote: > That explains how to avoid symbol mixups, to have two or more SQLite > "instances" > in a single process, but doesn't address the concern about POSIX locks > DRH mentioned. > if more than one of those same-process instances access the same file. --DD

[sqlite] 2 different SQLite versions inside the same process space

On 10/05/16 10:42, Andrey Gorbachev wrote: > I am a bit worried that the initialisation of 2 different versions of SQLite > would interfere with one another. Any advice? There is a way to do it, and I use it for my Python APSW extension as the recommended build instructions. It is especially

[sqlite] Incremental backup/sync facility?

On 06/05/16 05:32, Stephan Buchert wrote: > Having the copies of the file synced becomes increasingly tedious > as their sizes increase. Ideal would be some kind of > incremental backup/sync facility. Out of curiousity, would an approach of using multiple databases and using ATTACH to "unify"

[sqlite] SQLite vs GPLed software

On 02/05/16 20:00, Richard Hipp wrote: > I said I won't participate in that debate and I mean it. The relative > merits of GPL vs whatever is *not* the question on the floor. Apologies for not being clear. I am in no way debating the merits of either, nor expressing my opinion on the merits,

[sqlite] SQLite vs GPLed software

On 02/05/16 16:42, Richard Hipp wrote: > But some of the discussion did get me > thinking about the extent of GPLed software versus SQLite. Something that needs be made abundantly clear is the GPL is *not* about popularity. The GPL is about freedom (think freedom of speech, not price). Even

[sqlite] Recent downloads gone? Bad http returns

On 28/04/16 12:56, Richard Hipp wrote: > I intentionally removed 3.12.0 and 3.12.1 because they can (under > obscure circumstances) generate incompatible databases. I have no problems with removing the links to those downloads, but removing the downloads themselves seems a bit extreme. Don't

[sqlite] Recent downloads gone? Bad http returns

I can download 3.12.2 as http://sqlite.org/2016/sqlite-autoconf-3120200.tar.gz but 3.12.1 and 3.12.0 are now gone, giving a 404 error. URLs were http://sqlite.org/2016/sqlite-autoconf-3120100.tar.gz and http://sqlite.org/2016/sqlite-autoconf-312.tar.gz which used to work. Is this

[sqlite] Understanding table and database locking mechanism in shared cache mode and normal mode

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 30/03/16 16:58, Simon Slavin wrote: > In both modes (whether you're using 'shared cache' or not) use > either > > https://www.sqlite.org/c3ref/busy_timeout.html The last time I dealt with shared cache mode, the busy timeout did not apply. You

[sqlite] Reserved column names

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 21/03/16 03:32, Dominique Devienne wrote: > SQLite tries to be compatible with non-standard extensions from > various popular RDBMS', but when a standard alternative exists, it > should be preferred IMHO. --DD That depends on the code and project.

[sqlite] Reserved column names

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 19/03/16 03:30, Paul Sanderson wrote: > I know that keywords can be quoted but I am interested in just > those that can be used unquoted (even if not advisable) Out of curiousity, why? My rule of thumb is to always quote (using square brackets)

[sqlite] Changing the default page_size in 3.12.0

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 04/03/16 07:48, Richard Hipp wrote: > The tip of trunk (3.12.0 alpha) changes the default page size for > new database file from 1024 to 4096 bytes. ... This seems like a > potentially disruptive change, so I want to give you, the user > community,

[sqlite] applyng schema changes to several databases

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 17/02/16 03:38, Luca Ferrari wrote: > The above syntax> does not look familiar to me, It is made up to be concise and get the point across. > and in the meantime I've wrapped the user_version pragma get/set > into a shell script. You can't do if

[sqlite] SQLite tuning advice for caching scenario

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 17/02/16 08:20, Dave Baggett wrote: > One issue, though is that I'd have to run two FTS searches to > search -- one on the disk-based database, and one on the > memory-based one You see issues, I see features :-) The memory based cache would

[sqlite] SQLite tuning advice for caching scenario

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 17/02/16 06:37, Dave Baggett wrote: > I'd welcome any suggestions How about two databases? Create an in memory database for the cache. Then whenever it hits a certain size (eg 64MB) or time passed (eg 5 minutes), copy/move data from the memory

[sqlite] applyng schema changes to several databases

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 15/02/16 01:01, Luca Ferrari wrote: > While I'm pretty sure a simple sheel script that will execute, > file per file, the alter table (within a transaction) will do the > job I'm wondering if there's a better approach or a more automated > one.

[sqlite] Why SQLite does not use a web-based forum?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 14/02/16 17:19, admin at shuling.net wrote: > Why SQLite does not utilize a web-based forum for all users > discuss problems? I think that will be more convenient and can help > more people. Here you go in several different formats:

[sqlite] Use of __builtin_expect in SQLite

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/02/16 22:39, Matthias-Christian Ott wrote: > Amdahl's law is not applicable here and describes a completely > different problem. SQLite does not involve concurrency. Amdahl's law very much applies, and doesn't explicitly only involve

[sqlite] Use of __builtin_expect in SQLite

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/02/16 00:56, Dominique Pell? wrote: > I'm curious about the outcome on SQLite benchmarks. About a year ago I tried them out on some tight code (non-SQLite) that absolutely had to use less CPU time. I couldn't get them to make any difference

[sqlite] Slight problem with sqlite3_compileoption_get

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 24/01/16 13:12, Bart Smissaert wrote: >>> I think that unless the argument is going to be altered I have >>> to pass > them always ByVal. That wasn't quite right, eg: int > sqlite3_close(sqlite3*); sqlite3* is not going to be altered, but I > can

[sqlite] Slight problem with sqlite3_compileoption_get

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 24/01/16 02:19, Bart Smissaert wrote: > You didn't mention it but how arguments are called from VB6 is also > very important, that is ByVal or ByRef. I think that unless the > argument is going to be altered I have to pass them always ByVal. Sort

[sqlite] Slight problem with sqlite3_compileoption_get

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 23/01/16 17:36, Bart Smissaert wrote: > What is different though about sqlite3_compileoption_get, so that > long in the IDL causes the mentioned problems and int doesn't? Absolutely nothing is different about that API. If something was then one

[sqlite] Slight problem with sqlite3_compileoption_get

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 23/01/16 13:14, Bart Smissaert wrote: >> I am somewhat confused about what you wrote. > > This has to do with making a .tlb (type library) to access > sqlite3.dll from a VB6 ActiveX dll. That much was clear. > Sofar I have mapped SQLite int with

[sqlite] Slight problem with sqlite3_compileoption_get

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 23/01/16 11:41, Bart Smissaert wrote: > My question is how could I have foreseen this problem, so how could > I have known that using long is no good here? I am somewhat confused about what you wrote. SQLite provides a C level api. You can

[sqlite] Incompatible change in unix vfs xCurrentTime

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 20/01/16 20:11, Roger Binns wrote: > SQLite 3.10 changed the xCurrentTime (note not -64 version) method > for From the documentation it is hard to tell who is at "fault" > here. xCurrentTime is documented as optional now, but wa

[sqlite] Incompatible change in unix vfs xCurrentTime

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 SQLite 3.10 changed the xCurrentTime (note not -64 version) method for the "unix" VFS from a function to a NULL pointer. This has broken things for me, because I have a VFS that calls back into the "unix" vfs. A SQLite shared library upgrade will

[sqlite] Using sqlite3.exe as a subprocess

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 15/01/16 13:16, Matthew Allen wrote: > Yeah in hindsight it wasn't the best. I was trying to do: > > while still running: p.communicate etc > > Poll is not the right method to call. It especially is the wrong method in your scenario since you

[sqlite] Using sqlite3.exe as a subprocess

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 14/01/16 19:53, Matthew Allen wrote: > It seems that sqlite3.exe (console) doesn't work as a subprocess > with pipes. There is a bit of a problem with using apps via pipes. Generally when stdout is a terminal, output will be line buffered (ie you

[sqlite] Database Corrupt While Disk Full

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 14/01/16 00:44, sanhua.zh wrote: > Recently, my monitoring system showed that the error code > SQLITE_FULL and SQLITE_CORRUPT increasing in same trend. Just as another data point, I had SQLite using code in a library that was used across a bunch

[sqlite] Incorrect missing download behaviour on SQLite.org website

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 If you request a download that doesn't exist, the SQLite website behaves in a very non-standard way. For example the current download should be from a 2016 directory, but if you instead request from a 2015 directory, you instead get HTTP 302

[sqlite] [AGAIN] SQLite on network share

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/13/2015 11:55 AM, A. Mannini wrote: > About VistaDB it support use on network share look at > http://www.gibraltarsoftware.com/Support/VistaDB/Documentation/WebFram e.html#VistaDB_Introduction_SupportedPlatforms.html > > and confirmed from its

[sqlite] [AGAIN] SQLite on network share

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/13/2015 10:46 AM, A. Mannini wrote: > 1) is there a list of FS where SQLite works fine? I don't know of any. Network filesystems are very hard to implement (so many corner cases), and there is a lot of complexity if you also want them to be

[sqlite] [AGAIN] SQLite on network share

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/13/2015 10:31 AM, Richard Hipp wrote: > Just be warned that there are many network filesystems that claim > to implement locks correctly, and do most of the time, but > sometimes mess up It is also worth mentioning that SQLite trusts the

[sqlite] Feature Request: Support for MMAP beyond 2GB

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 https://sqlite.org/mmap.html SQLite supports memory mapping databases, but only does so for the first 2GB of the file. My databases are around 4 to 20GB, and completely fit within RAM on my 64 bit systems. The 2GB mmap limit means that only a

[sqlite] Question about Style

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 10/30/2015 11:08 AM, Ramar Collins wrote: > I'm working on a static site generator and want to use SQLite to > store metadata. I'm using C and a small library to get the > majority of the work done. My question is, do you have any > suggestions

[sqlite] SIGSEGV in sqlite3_open

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 10/28/2015 01:48 AM, Lohmann, Niels, Dr. (CQTN) wrote: > #0 0x010385c4 in SignalProcmask_r () from > C:\QNX650\target\qnx6/armle-v7/lib/libc.so.3 Have you tried turning off all compiler optimisations? Some older/embedded system compilers can

[sqlite] SQLITE_CANTOPEN after days of operation

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 10/11/2015 12:56 PM, Andrew Miles wrote: >>> 5) I ran lsof on the db, only one process (this one) had the >>> file open Have you run lsof on your monitoring process? You may be running out of file descriptors that are accessing other files or

[sqlite] Potential bug: Database on gvfs mount cannot be committed to

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 09/06/2015 11:13 AM, Florian Weimer wrote: > Surely that's not true, and NFS and SMB are fine as long as there > is no concurrent access? And no program crashes, no network glitches, no optimisation in the protocols to deal with latency, nothing

[sqlite] Potential bug: Database on gvfs mount cannot be committed to

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 09/06/2015 10:20 AM, Markus Weiland wrote: > I see. Since this was working under Ubuntu 14.04, I assume this is > a regression with gvfs. I'll check over there. Nope. SQLite can not maintain data integrity when used with *any* network filesystem.

[sqlite] Potential bug: Database on gvfs mount cannot be committed to

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 09/06/2015 06:16 AM, Markus Weiland wrote: > I've discovered a potential bug in handling of SQLite database > files on gvfs mounted network shares. SQLite doesn't support being stored on the network for several reasons, including that network file

[sqlite] ESC_BAD_ACCESS when using a background thread

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/31/2015 11:28 PM, Jeff M wrote: > All my bad -- I'm fessing up. Can you tell us how you found the root causes of the problems? It would be nice to know what tools and techniques worked. Roger -BEGIN PGP SIGNATURE- Version: GnuPG v2

[sqlite] ESC_BAD_ACCESS when using a background thread

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/25/2015 02:38 AM, Jeff M wrote: > I tried to install valgrind (on Mac OS Yosemite), but I can't get > it to compile. I don't have the fortitude to work out the install > issues. Generally you would be better off using something like homebrew:

[sqlite] ESC_BAD_ACCESS when using a background thread

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/24/2015 03:08 AM, Jeff M wrote: > I've checked all of your suggestions and nothing is amiss. You ran valgrind and it said everything is fine? That would be shocking. > I don't understand how the main thread can run before the > background

[sqlite] ATTACH DATABASE statement speed

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/23/2015 06:48 PM, Simon Slavin wrote: > Paolo is using Linux which does not do read-ahead optimization like > some versions of Windows. Therefore if he really is using an SSD > then fragmentation is not an issue. You are confusing things. The

[sqlite] ATTACH DATABASE statement speed

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/19/2015 05:56 PM, Paolo Bolzoni wrote: > I left running the pragma quick check during the night and finished > in 2 hours and 46 minutes, so it is about 8 times slower than in > ext4. Zfs is an advanced filesystem plenty of features, but this >

[sqlite] ESC_BAD_ACCESS when using a background thread

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/23/2015 03:31 AM, Jeff M wrote: > sqlite3_step(); // occasionally crashes here (showing > ESC_BAD_ACCESS on main thread) That has three very likely causes. The first is that your internal state gets messed up, and the statement has actually

[sqlite] Site unavailable.

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/11/2015 04:18 PM, ??? ??? wrote: > In Russia when I'm trying to visit sqlite.org site, I'm getting > conection timeout. Use down for everyone or just me to check sites: http://www.downforeveryoneorjustme.com/sqlite.org In this

[sqlite] how to detect when a table is modified ?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/07/2015 04:23 PM, Nicolas J?ger wrote: > I'm writing a program using sqlite3 and gtk to screen a database, > there is a way to get a signal from sqlite3 when something is > modified inside a database (values, add/delete row...), in the way > to

[sqlite] MMAP performance with databases over 2GB

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/07/2015 02:55 AM, Dan Kennedy wrote: >> http://www.sqlite.org/src/info/3a82c8e6cb7227fe > >> Does that improve performance any in your case? I tested 3a82c8e6 (fix above) against e596a6b6 (previous changeset), in both cases starting with an

[sqlite] Determine query type

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/06/2015 09:46 AM, Ben Newberg wrote: > Is there a good way to determine if a sql query is either > attempting to modify a database, or simply querying it? You most like want to use the authoriser. It tells you exactly what the query is doing,

[sqlite] MMAP performance with databases over 2GB

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/06/2015 09:27 AM, Dan Kennedy wrote: >> Is it using more CPU cycles in mmap mode or just taking longer? >> If the former, does [time] attribute them to "user" or "system"? It is taking longer. I have 3 XML dumps which I turn into JSON

[sqlite] MMAP performance with databases over 2GB

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 In my testing I am finding that using SQLite's mmap support with larger databases (final db size is ~10GB) to create the database (WAL mode) is considerably (about three times) slower than no mmap. The default max mmap limit is 2GB (sadly). Has

[sqlite] CSV excel import

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/30/2015 10:58 AM, Sylvain Pointeau wrote: > is it possible? in a lot of cases, I cannot use sqlite (executable) > because of the lack of a good CSV import. It would really great if > this could be addressed. Use the APSW shell:

[sqlite] why does SQLite support VARCHAR(X) when the doc says it ignores it?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/16/2015 02:46 PM, Hayden Livingston wrote: > It seems that > > CREATE TABLE A( something varchar(255) ) > > it just > > CREATE TABLE A( something text ) > > Why have it at all? Because other code and other humans look at the database too.

[sqlite] Schema-less JSON SQLite DB?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/15/2015 08:22 AM, Sergej Jure?ko wrote: > What do you guys think? Is it stupid, could it be improved? I recommend looking at Mongodb and Postgres first to see how they do queries. It would be better to be compatible with them where practical,

[sqlite] Fwd: Usenix paper: Korean researchers invent sqlite WALDIO mode to circumvent redundant journaling by EXT4 on eMMC

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/14/2015 06:53 AM, Mikael wrote: > Does SQLite do the same, what about SQLite do the same, patch? The SQLite authors rejected checksumming SQLite database pages. The existing integrity check will only catch issues that happen in sufficiently

[sqlite] Schema-less JSON SQLite DB?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/13/2015 08:00 PM, Hayden Livingston wrote: > Does your code also map object hierarchies in json? Yes, but thankfully I don't have much of them. Essentially the top level of the object has a unique id (SQLite allocated), and then other tables

[sqlite] Schema-less JSON SQLite DB?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/13/2015 05:43 PM, Hayden Livingston wrote: > Is there a concept of a schema-less JSON SQLite DB? What exactly is it you want? A schema-less database? A JSON database? Using SQLite to store JSON? (It is unclear if you mean something like

[sqlite] SIGSEGV rolling back in-memory DB‏‎

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/08/2015 10:33 AM, Brian Soby wrote: > Nothing appears to happen but it's not running very faithfully. > Specifically, I get a bunch of problems on the rollbacks about the > savepoints not existing even though it's replaying the original >

[sqlite] SIGSEGV rolling back in-memory DB‏‎

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/08/2015 09:55 AM, Brian Soby wrote: > It's single threaded. Good - that eliminates a whole bunch of complications. > I have a log of all of the queries run through the database up > until the crash. It crashes on a rollback command. What

[sqlite] SIGSEGV rolling back in-memory DB‏‎

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/07/2015 02:06 PM, Brian Soby wrote: > I don't believe it's a python specific issue since the core problem > occurs in a rollback command internal to SQLlite. Unless you configure it otherwise, pysqlite likes to parse your SQL and do extra

[sqlite] SIGSEGV rolling back in-memory DB‏‎

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/07/2015 09:47 AM, Richard Hipp wrote: > Can you send instructions on how to recreate the problem? Please > bare in mind that none of the core SQLite developers do much with > python. Alternatively, you might send emails to Ro

[sqlite] Extending VFS documentation

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 05/24/2015 02:39 AM, Philip Bennefall wrote: > I have looked at the example VFS implementations, but it is hard > to determine which parts of the code that are implementation > details as opposed to being part of the stable public API so to >

[sqlite] xBestIndex() implementation question

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 05/13/2015 08:06 PM, Jilong Kuang wrote: > Sure, that is one option. But I'm just wondering if inside the > xBestIndex() we can get the value field, we can do a much better > job to give the cardinality info. Note that the value could be something

[sqlite] Suggestion for .import improvement

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 05/11/2015 12:24 PM, tonyp at acm.org wrote: > It?s a real nuisance to have to edit out the headers every time, > especially when the files are very large and they do not load fast > (or at all) into a text editor. Plus, it?s semi-automatic. I

[sqlite] Issue: Inconsistency of sqlite3_dbstat_register declaration

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 05/11/2015 01:34 AM, Jan Nijtmans wrote: > 2) If sqlite is not compiled with -DSQLITE_ENABLE_DBSTAT_VTAB=1, > or the function is renamed (as it should) it wouldn't even work. That was my scenario. I just -DSQLITE_ENABLE_DBSTAT_VTAB and the code

[sqlite] Issue: Inconsistency of sqlite3_dbstat_register declaration

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 There is a forward declaration of sqlite3_dbstat_register (currently line 130871 in the 3.8.10.1 amalgamation): int sqlite3_dbstat_register(sqlite3*); Later is the actual implementation (currently line 155838 in the 3.8.10.1 amalgamation):

[sqlite] Bug: sqlite3_memory_used/highwater truncate to 32 bits

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 The sqlite3_memory_used and highwater interfaces are defined to return a 64 bit value. They carefully call the 32 bit limited sqlite3_status method and then cast to 64 bit. Instead they should call sqlite3_status64. Reported indirectly via

  1   2   3   4   5   6   7   8   9   10   >