[datatable-help] fread statement for a long data table

[Frank S.]

Hi all, I've a data base in txt file, which contains approximately  1.5 million observations and 18 variables. The aspect of two of the observationsis the following: 345~X~99~30/10/1950~89000~784~ERVY LE CHATEL, RUE~9~AUXERRE~08500~410005~VISITE MEDICALE~20/03/1998~20~UROLOGIE~9~VISITES~5~ 67895~J~102~28/05/1967~89000~359~CHATILLON SUR SEIENE, RUE~10~AUXERRE~08340~560025~ASSURANCE-CRÉDIT~15/09/1997~25~UROLOGIE~1~VISITES~5~ So, as you can see, the columns are separated by "~" symbol, but there is also other symbols: ",", "-", written accents, ...  inside some descriptive variables. I execute: data <- fread('data.txt',autostart=60) The R gives an error message in one row Error en fread("data.txt", autostart = 60) :   Expected sep (',') but '' ends field 1 on line 1 when detecting types: Please, anyone can help me? 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/datatable-help/attachments/20140728/2b6ca0a6/attachment.html>