Hello All, Wouldn't it make sense to write a program that reads it in, one line at a time, splits and inserts the data into the proper tables? Even creating the table on the fly? That's what I'd do, a little command line utility.
It's basically the "convert to SQL" technique you suggest where the SQL conversion's done programatically in the utility. Tuesday, December 6, 2005, 11:50:50 PM, you wrote: RLC> I create an SQL file that has contents like this: RLC> [EMAIL PROTECTED] elections]$ cat insert_precinct.sql RLC> BEGIN TRANSACTION; RLC> INSERT INTO "precinct" VALUES(1, 'Community Center 15 Crescent Road', 3, RLC> 'Greenbelt', 'Maryland', 0); RLC> INSERT INTO "precinct" VALUES(2, 'Police Station 550 Crescent Road', 6, RLC> 'Greenbelt', 'Maryland', 0); RLC> INSERT INTO "precinct" VALUES(3, 'Springhill Lake Recreation Center 6111 RLC> Cherrywood Lane', 8, 'Greenbelt', 'Maryland', 0); RLC> COMMIT; RLC> Then I fire up sqlite3 on the command line, and issue RLC> .read insert_precinct.sql RLC> I realize this will probably make you unhappy because it means editing RLC> your CSV file so that each line is transformed into an sql statement. RLC> This can be done most easily with sed (if you are a Linux or Unix RLC> person), but you need to know sed commands and you need to be willing to RLC> patiently experiment until the sed script applies exactly the right edits. RLC> Bob Cochran RLC> ronggui wong wrote: >>I have a very large CSV file with 10000 rows and 100 columns.and the >>file looks like the following: >>"a","b","c","d", >>"1","2","1","3" , >>"3","2","2","1", >>...... >> >>If I use .import,It seems I have to set the variable names manually . >>Is there any way to import the whole data file into SQLite quickly? >>Thank you! >> >>ronggui >> >> >> >> -- Best regards, Teg mailto:[EMAIL PROTECTED]
