Re: [sqlite] SqlAlchemy 0.9 release note complaint

[big stone]

Python 3.4 last beta3 is planned for January 26, 2014

Current SQLite in Python3.4 beta2 is 3.8.1.

Unless a miracle occurs these next few days, we will not see CTE in
standard Python before 2 years more.
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Lightweight integrity_check?

[Tim Streater]

On 15 Jan 2014 at 19:20, Steven Fisher  wrote: 

> After opening, I want to do some basic checking of my database.
>
> I’m finding integrity_check much too slow for some of my users (especially
> with large amounts of data) but I don’t want to just blindly trust the
> database either.
>
> A few options:
>
> - Rely on what sqlite does on its own when I start actually using the database
> - Test for particular tables I expect
> - Something else
>
> Any thoughts?

On startup of my app (an email client) I do a number of checks on each mailbox 
(each is an SQLite db). I do these things:

1) open it
2) try to select something from the main table
3) try to read and write back a value from a metadata table (it has just the 
one row)

That ensures that the mailbox exists, has the right tables, and is 
read/writable. You could do something similar, appropriate for your situation.



--
Cheers  --  Tim
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Case Insensitive Defect with New Query Planner

[Richard Hipp]

On Wed, Jan 15, 2014 at 6:13 PM, Chris Swinefurth wrote:

> The query:
> SELECT   terms.term, terms.id, terms.type, terms.product_id FROM   terms
>   JOIN products   ON terms.product_id = products.id   JOIN
> sub_categories ON products.category_id = sub_categories.id WHERE
> terms.term LIKE 's%' collate nocase AND   sub_categories.parent_id = 89133;
>

Thank you for the query.  Can you also send the schema so that we can run
it?

-- 
D. Richard Hipp
d...@sqlite.org
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Case Insensitive Defect with New Query Planner

[Chris Swinefurth]

Guys,
I’ve identified a detrimental query planner change between 3.7.12 and 
3.8.1.  It appears to be a change from the 3.8 NGQP.  A query with a like where 
clause with “collate nocase” now uses a full table scan instead of a covering 
index.  This occurs even if the case_sensitive_like pragma is left at default / 
off.  Later it was found removing the “collate nocase" reverted to the previous 
plan using a covering index.  Removing the modifier on an already case 
insensitive should have been a no-op?

Test was run with downloaded sqlite binary running on OSX.  The primary 
table (terms) has ~1.6 million rows, the others, hundreds.  Analyze was run 
before queries on both versions.  Original .sqlite file was created with 3.7.?? 
and a pristine copy was made before each test / analyze.

The query:
SELECT   terms.term, terms.id, terms.type, terms.product_id FROM   terms   JOIN 
products   ON terms.product_id = products.id   JOIN sub_categories ON 
products.category_id = sub_categories.id WHERE   terms.term LIKE 's%' collate 
nocase AND   sub_categories.parent_id = 89133;

3.7.12 Explain Plan
sele  order  from  deta
  -    
0 0  0 SEARCH TABLE terms USING COVERING INDEX 
term_id_type_product_id (term>? AND termhttp://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Insert multiple values as an economy of coding time and quicker upload process ?

[Keith Medcalf]

>I'm not an expert in SQLite extension.
>Unless I missed something, Wholenumber seems apparently not embeded
>into SQLite for python :

>CREATE VIRTUAL TABLE nums USING wholenumber;
>SELECT value FROM nums WHERE value<1;
>SELECT value FROM nums WHERE value BETWEEN 100 AND 500

>gives :
>
>OperationalError: no such module: wholenumber
>Error on sql CREATE VIRTUAL TABLE nums USING wholenumber;

That is true ... you would have to build the replacement amalgamation yourself 
and include the ext/misc extensions you want to use, or build the wholenumber 
extension as a separate DLL and load it yourself.




___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] AUTO: Oren Bar is out of the office - He is in vacation (returning 27/02/2014)

[Oren Bar]

I am out of the office until 27/02/2014.

I will be unavailable.
For CDS issues, please contact Asaf Levy.
For PT issues, please contact Elena Drobchenko.


Note: This is an automated response to your message  "Re: [sqlite] How to
bind parameters to LIKE with sqlite3_bind_text?" sent on 15/01/2014
19:14:32.

This is the only notification you will receive while this person is away.

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SqlAlchemy 0.9 release note complaint

[Keith Medcalf]

Yes, just changing the DLL is safe.  I believe the default DLL might need to be 
compiled with SQLITE_ENABLE_COLUMN_METADATA.  I replace the sqlite3.dll in 
Python all the time to keep it up-to-date with the latest head.

>-Original Message-
>From: sqlite-users-boun...@sqlite.org [mailto:sqlite-users-
>boun...@sqlite.org] On Behalf Of Richard Hipp
>Sent: Wednesday, 15 January, 2014 11:55
>To: General Discussion of SQLite Database
>Subject: Re: [sqlite] SqlAlchemy 0.9 release note complaint
>
>On Wed, Jan 15, 2014 at 1:53 PM, big stone  wrote:
>
>> Ok,
>>
>> The problem is fixed starting SQLite 3.7.16,
>> But Python 3.3 series (up to 3.3.3) seems to be shipped with the very
>old
>> SQLite 3.7.12, by default.
>>
>> Is there a simple trick to "upgrade" SQLite3 version used by Python
>(under
>> windows) ?
>> Just changing the sqlite3.dll is safe ?
>>
>
>That's really a question for the Python people, but my guess is that
>changing the sqlite3.dll would be safe.
>
>
>
>--
>D. Richard Hipp
>d...@sqlite.org
>___
>sqlite-users mailing list
>sqlite-users@sqlite.org
>http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users



___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Lightweight integrity_check?

[Dan Kennedy]

On 01/16/2014 02:20 AM, Steven Fisher wrote:

After opening, I want to do some basic checking of my database.

I’m finding integrity_check much too slow for some of my users (especially with 
large amounts of data) but I don’t want to just blindly trust the database 
either.

A few options:

- Rely on what sqlite does on its own when I start actually using the database
- Test for particular tables I expect
- Something else

Any thoughts?


You could try this:

  http://www.sqlite.org/pragma.html#pragma_quick_check

Dan.

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Lightweight integrity_check?

[Steven Fisher]

After opening, I want to do some basic checking of my database.

I’m finding integrity_check much too slow for some of my users (especially with 
large amounts of data) but I don’t want to just blindly trust the database 
either.

A few options:

- Rely on what sqlite does on its own when I start actually using the database
- Test for particular tables I expect
- Something else

Any thoughts?
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SqlAlchemy 0.9 release note complaint

[Richard Hipp]

On Wed, Jan 15, 2014 at 1:53 PM, big stone  wrote:

> Ok,
>
> The problem is fixed starting SQLite 3.7.16,
> But Python 3.3 series (up to 3.3.3) seems to be shipped with the very old
> SQLite 3.7.12, by default.
>
> Is there a simple trick to "upgrade" SQLite3 version used by Python (under
> windows) ?
> Just changing the sqlite3.dll is safe ?
>

That's really a question for the Python people, but my guess is that
changing the sqlite3.dll would be safe.



-- 
D. Richard Hipp
d...@sqlite.org
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SqlAlchemy 0.9 release note complaint

[big stone]

Ok,

The problem is fixed starting SQLite 3.7.16,
But Python 3.3 series (up to 3.3.3) seems to be shipped with the very old
SQLite 3.7.12, by default.

Is there a simple trick to "upgrade" SQLite3 version used by Python (under
windows) ?
Just changing the sqlite3.dll is safe ?
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Insert multiple values as an economy of coding time and quicker upload process ?

[big stone]

I'm not an expert in SQLite extension.
Unless I missed something, Wholenumber seems apparently not embeded
into SQLite for python :

CREATE VIRTUAL TABLE nums USING wholenumber;
SELECT value FROM nums WHERE value<1;
SELECT value FROM nums WHERE value BETWEEN 100 AND 500

gives :

OperationalError: no such module: wholenumber
Error on sql CREATE VIRTUAL TABLE nums USING wholenumber;
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] How to bind parameters to LIKE with sqlite3_bind_text?

[Stephan Beal]

On Wed, Jan 15, 2014 at 4:41 PM, Clemens Ladisch  wrote:

> sqliteuser wrote:
> > sqlite3_bind_text(stmt, 1, qry.c_str(), -1, 0);
>
> The last parameter must not be zero.  For string::c_str(), you
> should use SQLITE_TRANSIENT.
>


Depending on the context, SQLITE_STATIC could be legally used, but only if
the string in question is guaranteed to survive (_unchanged_) for the life
of the statement.

-- 
- stephan beal
http://wanderinghorse.net/home/stephan/
http://gplus.to/sgbeal
"Freedom is sloppy. But since tyranny's the only guaranteed byproduct of
those who insist on a perfect world, freedom will have to do." -- Bigby Wolf
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Serialization and threading

[Clemens Ladisch]

Reza Housseini wrote:
> The execution looks as follow:
>
> sqlite3_step(beginstmt);
> sqlite3_step(vectorstmt);
> sqlite3_step(intrstmt);
> sqlite3_step(commitstmt);

So your program does not check error codes?


Regards,
Clemens
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Load spatialite-extension through System.Data.SQLite

[a . furieri]

On Tue, 14 Jan 2014 20:54:54 +0100, Markus Dibo wrote:

I am trying to use SQLite/Spatialite from a .Net-4.0-Application. To
get easy access to SQLite I am using System.Data.SQLite.
I try to load version 4.1.1 of Spatialite via the command-text
"**SELECT load_extension("libspatialite-4.dll")" which leads to an
Access Violation.
Other users seem to have the same problem:
https://groups.google.com/forum/#!topic/spatialite-users/fOb_J8djVP4



Hi Markus,

more or less similar issues were reported about using splite 4.1.1
on Java/JDBC and Python: so for sure it's not a System.Data.SQLite
specific problem.

you can find more recent news related to this topic on the
SpatiaLite Users list:
https://groups.google.com/forum/#!forum/spatialite-users

and specifcally:
https://groups.google.com/forum/#!topic/spatialite-users/EqJAB8FYRdI
https://groups.google.com/forum/#!topic/spatialite-users/JbcX_k4S1m4

bye Sandro
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Serialization and threading

[Igor Tandetnik]

On 1/14/2014 2:41 AM, Reza Housseini wrote:

But now it seems that not every interaction is logged into the database (I
also log the data in a std container). As I understand with the
serialization option multiple threads can have access to the same database
connection and statement, and sqlite will sort out any data races and dead
locks.


There can't be any data races or deadlocks to sort out when you only 
have a single connection. Within a single connection, SQLite simply 
serializes all API calls on a mutex. It doesn't matter if a single tread 
executes two BEGIN statements in a row, or two threads execute one each 
(on the same connection): the second statement will fail. It doesn't 
matter if a single thread calls sqlite3_step on an INSERT statement 
twice in a row, or two threads execute one each (on the same statement 
handle): the second call has no effect (without an intervening 
sqlite3_reset).

--
Igor Tandetnik

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] How to bind parameters to LIKE with sqlite3_bind_text?

[Igor Tandetnik]

On 1/14/2014 8:10 AM, sqliteuser wrote:

what is the proper way to bind string to LIKE statement?


The same way you bind any other string parameter. The fact that the 
parameter happens to be used as an operand of a LIKE operator is irrelevant.



std::string qry = "SELECT name "
   " FROM Person "
   " WHERE name LIKE ? ";

sqlite3_prepare_v2 (...)

sqlite3_bind_text(stmt, 1, qry.c_str(), -1, 0);


You are binding the text of the query as a parameter to that same query. 
The query you are effectively executing looks like this:


select name from Person where name LIKE 'select name from Person where 
name LIKE ?';


Is that really what you meant to do? I would hate to be a person whose 
name matches that pattern (little Bobby Tables, we call him: 
http://xkcd.com/327/).

--
Igor Tandetnik

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Urdu Language Support

[Igor Tandetnik]

On 1/15/2014 8:25 AM, Rehan Siddiqui wrote:

Does Sqlite support Urdu language. We want to put a lot of urdu text data in 
our software.


Depends on what you mean by "support". SQLite can store arbitrary 
Unicode strings, whether containing characters used in Urdu or any 
others. If you need, say, correct sorting for Urdu language strings, 
that is not supported out of the box and requires the use of extensions 
(such as 
http://www.sqlite.org/src/artifact?ci=trunk=ext/icu/README.txt)

--
Igor Tandetnik

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Urdu Language Support

[Rehan Siddiqui]

Hi, 

Does Sqlite support Urdu language. We want to put a lot of urdu text data in 
our software.

Regards
  
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] How to bind parameters to LIKE with sqlite3_bind_text?

[sqliteuser0000]

Hi,
what is the proper way to bind string to LIKE statement? For some reason i
get no results:
I can bind int's just fine (not with LIKE though).

std::string qry = "SELECT name "
  " FROM Person "
  " WHERE name LIKE ? ";

sqlite3_prepare_v2 (...)

sqlite3_bind_text(stmt, 1, qry.c_str(), -1, 0);

I need to use injection because the names may include ' characters

How to get this to work?

Regards



--
View this message in context: 
http://sqlite.1065341.n5.nabble.com/How-to-bind-parameters-to-LIKE-with-sqlite3-bind-text-tp73237.html
Sent from the SQLite mailing list archive at Nabble.com.
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Serialization and threading

[Reza Housseini]

Hello

I want to use sqlite to log some data into a simple table. My application
is multi-threaded but I only need to write to the database. I begin with
specifying the statements.

const char *vectorsql = "INSERT INTO vector (x0,x1,x2) VALUES (?,?,?);";
const char *intrsql = "INSERT INTO interaction
(bodyid1,bodyid2,iter,vectorid)" \
  "VALUES (?,?,?,last_insert_rowid());";
sqlite3_prepare_v2(db, "BEGIN EXCLUSIVE TRANSACTION;", -1, ,
NULL);
sqlite3_prepare_v3(db, vectorsql, -1, , NULL);
sqlite3_prepare_v2(db, intrsql, -1, , NULL);
sqlite3_prepare_v2(db, "END TRANSACTION;", -1, , NULL);

The execution looks as follow:

sqlite3_bind_int(vectorstmt, 1, x0);
sqlite3_bind_int(vectorstmt, 2, x1);
sqlite3_bind_int(vectorstmt, 3, x2);
sqlite3_bind_int(intrstmt, 1, bodyid1);
sqlite3_bind_int(intrstmt, 2, bodyid2);
sqlite3_bind_int(intrstmt, 3, iter);

sqlite3_step(beginstmt);
sqlite3_step(vectorstmt);
sqlite3_step(intrstmt);
sqlite3_step(commitstmt);

sqlite3_reset(beginstmt);
sqlite3_reset(vectorstmt);
sqlite3_reset(intrstmt);
sqlite3_reset(commitstmt);

sqlite3_clear_bindings(vectorstmt);
sqlite3_clear_bindings(intrstmt);

But now it seems that not every interaction is logged into the database (I
also log the data in a std container). As I understand with the
serialization option multiple threads can have access to the same database
connection and statement, and sqlite will sort out any data races and dead
locks. And within an exclusive transaction no other process can have read
or write access to the database. But the interaction table and the vector
table don't have the same number of rows, therefore some race conditions
must occur.

Can someone point out my errors?

Cheers Reza
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Bug Report SQLITE JDBC driver 3.7.2: problem calling "wasNull" function in the ResultSet.

[Burkhard Losch]

Hi,

if the user calls "getObject" before calling wasNull then wasNull returns 
either the "wasNull" value of the previous called get call, or 
it throws an Exception, if it is called the first time.


How to replicate this bug:

DBConnection conn = 
DatabaseMetaData md = conn.getMetaData();

ResultSet rs = md.getTables(null, null, null, new String[] {"TABLE});


if (rs.next()) {
   Object anyData = rs.getObject(1)
   // the following statement will throw an exception: "column -1 out of 
bounds [1,10]"
   if (rs.wasNull()) {
  ... 
   }
 
}


I saw in the source code that getObject does not set the lastCol variable 
in the ResultSet class "RS".


The expected behavior is that the getObject function sets the "lastCol" 
variable so that cosecutive function calls to "wasNull" do not fail with 
an exception.


Would be nice to hear from you.

Regards,
Burkhard.
E-world 2014
11.-13..2.2013 Essen/Germany

Besuchen Sie uns auf der E-world 2014 | Halle 3 | Stand 131
Wir freuen uns auf Ihren 
Besuch.
 Burkhard Losch - KISTERS AG - Charlottenburger Allee 5 - 52068 Aachen - Germany
Handelsregister Aachen, HRB-Nr. 7838 | Vorstand: Klaus Kisters, Hanns Kisters | 
Aufsichtsratsvorsitzender: Dr. Thomas Klevers
Tel.: +49 241 9671 -112 | Mobil: +49 176 19671041 | E-Mail: 
burkhard.lo...@kisters.de | WWW: http://www.kisters.de

Diese E-Mail enthält vertrauliche und/oder rechtlich geschützte Informationen. 
Wenn Sie nicht der richtige Adressat sind oder diese E-Mail irrtümlich erhalten 
haben, informieren Sie bitte sofort den Absender und vernichten Sie diese Mail. 
Das unerlaubte Kopieren sowie die unbefugte Weitergabe dieser Mail ist nicht 
gestattet. 
This e-mail may contain confidential and/or privileged information. If you are 
not the intended recipient (or have received this e-mail in error) please 
notify the sender immediately and destroy this e-mail. Any unauthorised 
copying, disclosure or distribution of the material in this e-mail is strictly 
forbidden.
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Insert multiple values as an economy of coding time and quicker upload process ?

[Keith Medcalf]

>
>INSERT INTO t1 SELECT 1 UNION SELECT 2 UNION (...) SELECT ;
>
>It works out of the box on any DBMS you care to name, and of course in
>SQLite, too.  It is ever so slightly verbose, but better than the
>alternative of executing  INSERT INTO t1 VALUES(###);
>

Of course, if your objective is just to have a table with a wholenumber 
sequence of rows, you could use the wholenumber virtual table extension:

CREATE VIRTUAL TABLE nums USING wholenumber;
SELECT value FROM nums WHERE value<1;
SELECT value FROM nums WHERE value BETWEEN 100 AND 500;

wholenumber.c is part of the sqlite distribution in ext/misc under the source 
tree ...




___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] shell crashes when there is no column name

[Richard Hipp]

On Wed, Jan 15, 2014 at 6:48 AM, Clemens Ladisch  wrote:

>
> And a NULL column name is considered a bug:
> 
>
>
Not so much a "bug" as "undesirable".  Note that even with the change
referenced above, sqlite3_column_name() might still return NULL if there is
an out-of-memory error.  (This is very unlikely to happen, but it is
possible.)  So the shell has been updated to test that case too.
http://www.sqlite.org/src/info/ac15455abc

We run extension OOM testing on the core library and probably would have
caught the above problem there.  But the level of testing on the
command-line shell is much lower.  And as this problem was in the
command-line shell, not the core library, it managed to slip through.


-- 
D. Richard Hipp
d...@sqlite.org
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] shell crashes when there is no column name

[Clemens Ladisch]

Klaas V wrote:
> Clemens wrote:
>>  $ sqlite3
>>  SQLite version 3.8.3 2014-01-11 12:52:25
>>  Enter ".help" for instructions
>>  Enter SQL statements terminated with a ";"
>>  sqlite> .header on
>>  sqlite> values(1);
>>  (null)
>>  1
>>  sqlite> .mode html
>>  sqlite> values(1);
>>  Segmentation fault
>
> One runs the risk stumbling upon obscure bugs sing what seems to be a beta.
> Aforementioned becomes almost certain when one uses new features.

Indeed.  This is why I was playing with VALUES.

>>  And I guess NULL should not be returned in the first place if there is no 
>> error.
>
> Since you did not provide a column name returning (null) is quite logical.

"(null)" is just what fprintf outputs for a NULL string.

And a NULL column name is considered a bug:



Regards,
Clemens
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SqlAlchemy 0.9 release note complaint

[Richard Hipp]

On Wed, Jan 15, 2014 at 1:50 AM, big stone  wrote:

> Hello,
>
> Reading SQLAlchemy release note, I see a functionality complaint on SQLite.
>
> see :http://docs.sqlalchemy.org/en/rel_0_9/changelog/migration_09.html
>
>
> In its last release Note, SQLAlchemy complains that  SQLite doesn't allow
> this :
>
> SELECT a.*, b.*, c.* FROM a LEFT OUTER JOIN (b JOIN c ON b.id = c.id) ON
> a.id
>


That issue was addressed over a year ago, at
http://www.sqlite.org/src/info/99127a669c49f for SQLite version 3.7.16.

-- 
D. Richard Hipp
d...@sqlite.org
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] shell crashes when there is no column name

[Klaas V]

Clemens wrote:

>  $ sqlite3
>  SQLite version 3.8.3 2014-01-11 12:52:25
>  Enter ".help" for instructions
>  Enter SQL statements terminated with a ";"
>  sqlite> .header on
>  sqlite> values(1);
>  (null)
>  1
>  sqlite> .mode html
>  sqlite> values(1);
>  Segmentation fault


One runs the risk stumbling upon obscure bugs sing what seems to be a beta.
Aforementioned becomes almost certain when one uses new features.

With the official 'r.v.u b' out of the box (no special compiler options 
whatsoever):

sqlite3
SQLite version 3.8.2 2013-12-06 14:53:30
Enter ".help" for instructions
Enter SQL statements terminated with a ";"
sqlite> .he on
sqlite> values(1);
Error: near "values": syntax error
sqlite> .mode html
sqlite> values(1);
Error: near "values": syntax error

>  And I guess NULL should not be returned in the first place if there is no 
> error.

Since you did not provide a column name returning (null) is quite logical. What 
else? 

 

Kind regards | Cordiali saluti | Vriendelijke groeten | Freundlliche Grüße,
Klaas `Z4us` V, CIO & freelance software engineer
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users