I'm new to SQLite but have some experience with MS SQL Server.  I am looking 
for 
a quick confirmation from expert that my planed use of SQLite is reasonable and 
maybe point out if there are any gotchas.  I've done some prototyping and tried 
most of this out but it is based on how I've done things in the past so maybe 
not best way in SQLite.  My prototypes have only been very small.

I will have two general usage patterns.  1) Few and infrequent single row 
updates.  2) I will start a secondary processes that will scan through a table 
(1000-20000 rows) and for each row run an analysis that will take 1 to 90 
seconds and then update another table with the results.

I'm on windows.  I have compiled SQLite into a dll and wrapped it in some C++ 
classes.  SQLite is not exported from the dll.  I plan to use this in both 32 
bit and 64 bit applications to access potentially the same files.  


I will use the following options when I compile SQLite:
SQLITE_THREADSAFE=2
SQLITE_DEFAULT_FILE_FORMAT=4

I have my own compilation of sqlite3.exe that I compile the code of sqlite3.c 
and shell.c into the same application using the same options as above.

I call my dll and exe: sqlutils.dll and sqlutil.exe 

Generally, in the past I have always used normalized data but thought I read 
this might not always be most efficient in SQLite.  Is there any "rule of 
thumb"?

My tables will be relativly small.  Most probably in the 1000-20000 rows range.

I will call stored procedure from C++ for any actually changes to the data.

I will use views to make queries simply from c++ code.

Thank you for you time in looking into this,
Jeff
_______________________________________________
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Reply via email to