[sqlite] Retrieving constraint name

[Igor Korot]

Hi,
The CREATE TABLE statement supports the following syntax:

CREATE TABLE( , CONSTRAINT  FOREIGN
KEY() REFERENCES (ref_column_list>);

However, the statement "PRAGME foreign_key_list;" does not list the
foreign key name ("fk_name" in the statement above).

Does the info for the aforementioned PRAGMA stored somewhere?
If yes - does it include the key name and it just not printed with the PRAGMA?
If not - does this mean that the only way to get the name is to parse the sql
from sqlite_master? Or there is a better way?

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

Re: [sqlite] Unable to store 500MB size of row data even after define macro SQLITE_MAX_LENGTH

[Simon Slavin]

On 9 Dec 2017, at 6:29am, Dianne Dunn  wrote:

> How,do I get off this email list?? What is it for??

Use the. link on the last line of every post, including this one.

Discussing SQLite.  Why and how did you join it if you didn’t know ?

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

Re: [sqlite] Virtual table row deletition on trigger

[Tibor Balog]

Hi,

I have realized that this quetion went awry so I give it a second run.

My scenario:

   Enviroment:

DotNet, no 3d party library just pinvoke.
SQLite:
Sqlite library version:3022000
-COMPILER=msvc-1911
-ENABLE_FTS5
-LIKE_DOESNT_MATCH_BLOBS
-MAX_EXPR_DEPTH=0
-OMIT_DECLTYPE
-OMIT_DEPRECATED
-OMIT_PROGRESS_CALLBACK
-OMIT_SHARED_CACHE
-TEMP_STORE=3
-THREADSAFE=1

Using:Synchronous connection.
"PRAGMA journal_mode=MEMORY"

   My assumption:
Opening just one (Read/Write|Create) conection SQLite should never signal 
busy to that connection after it successfuly acquired it.


Since you can use thight loops and it should be up to SQLite how long its 
takes for the individual command execution and how many threads its opens up 
to accomplish the job at hand,
it should come back sync. otherwise it will break the loop forcing to launch 
a signal handling thread.


   My task setup:
A table "ART"
+
A virtual fts5 contentless table:
CREATE VIRTUAL TABLE "OCR" using fts5 (content='',FullText)
+
Trigger:
CREATE TRIGGER "ART_AD" AFTER DELETE ON "ART" BEGIN INSERT INTO "OCR" 
("OCR",rowid) VALUES('DELETE',old.rowid);END

to keep the virtual table sync with row deletation of "ART"

   My task.:
0 Collecting user input from grid selection into a list of SQL DELETE range 
statements (BETWEEN value1 AND value2;)


1 Open connection
2 BEGIN_TRANSACTION
3 the delete loop on the list.
4 COMIT

SQLite breaks the loop with signal "busy" if one or more of the delete 
ranges are bigger than ~20.000 rows.


Without the transaction frame seems to be ok.

   My quetion:
Is my "assumption" above wrong and I allways have to take care of eventual 
interraptions from SQLite?
Is this a limitation because of the trigger forces a separate thread for the 
virtual table, which is maybe not that thightly integrated?

Is this a bug?

KR,
Tibor

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