I don't think it is because the big file causes the problem. For debugging, print the data on console using tLogRow and check the 'die on error' option on tFileInputExcel, the job will stop to run once an error occurs, all the right data will be printed on the console, then you will check the next row which has the bad data.
Hello Shong, thank you for the answer. The problem is that the word "FALSE" is not present anywhere in the Excel document, that's why I need to identify the row with the problem. I've tried to create another excel file using some of the rows of the first document, but I have no error.
The file has a lot of columns and about 20000 rows: is there any function on talend that can tell me the cell, or simply the row, that causes the problem ?
Hello, an update, maybe can be useful to someone: the file giving problems to my talend job is an xlsx generated by an Access DB; I took this file, loaded with libreoffice, saved as xlsx (I know, it seems pointless) and substituted to the original file. The error "java.lang.NumberFormatException: For input string: "FALSE"" disappeared and all is working now
Just to be sure I tried several times, using different files, keeping the same job procedure
The file saved by libreoffice has the same data than the original, but a smaller size; it seems like there's a bug on the java library used to read the excel file, maybe big files, I've never had problems with small excel files.
However, it would be great if Talend had a way to tell more details error about the row giving error
I don't think it is because the big file causes the problem. For debugging, print the data on console using tLogRow and check the 'die on error' option on tFileInputExcel, the job will stop to run once an error occurs, all the right data will be printed on the console, then you will check the next row which has the bad data.
