Parsing CSV files in Apex
In my grousing about the lack of easy CSV parsing in Apex, I began to feel like Paul Rudd’s character in Web Hot American Summer. Usually it’s not something you would need to do in SalesForce – the Import Wizard or Data Loader are built to do this easily for one table at a time.
I needed to build a widget that would operate on data in potentially 5 tables simultaneously, hence the consternation. Well, it wasn’t all that hard to roll my own. The code sample below (edit: code plugin seems to have died at some point, code is here) does not handle newline characters in the cells/fields, but should do fine dealing with commas and double quotes. I even got to leave the computer for a precious few seconds to sketch it out on the whiteboard:
Thank you for providing the code sample for parsing a CSV file with Apex.
http://wiki.developerforce.com/index.php/Code_Samples#Parse_a_CSV_with_APEX
However, I would like to provide feedback that I don’t think the code correctly handles quote text delimiters, because the function returns a quoted value with the quotes instead of returning just the value.
For example, take a file with the following contents:
“My Value”
The value is returned as ‘”My Value”‘ instead of ‘My Value’.
I want to know how to get the data from the csv to the function?
Since the string takes the content from CSv file. it throws limit exceeded exception. So we cant read or upload many records for example 150 at a time in a file. IS there any alterantive where we can read and store it in a temp file like Java and then upload to salesforce.
Please help if there is any solution for this.
Vanessen, you can get the CSV to the function in a number of ways: Upload it to a Document and parse the Document, do a callout to an external web service, or paste the CSV contents into a textarea in a VisualForce page (in a controller member variable’s field).
Marty, I haven’t tested it but you may be right. You could add a line that looks for fields starting with “, and ending with “, with no ” between. Regex like this? /^”.*”$/. Then chop the first and last character.
Amith, we used a Document and then read from the Document. Try that?