Closed SoftCircuits closed 11 months ago
Also, I wanted to add a comment. Using my benchmarks, my parser showed a performance boost of better than 100%. However, using your benchmarks, the improvements are much more modest.
Examining your test data, I have an idea why this might be. Your test data contains no special characters (quotes, commas, newlines), which require special handling in CSV files. My test data included many random punctuation characters and multi-line field values. And I guess I had really optimized for those cases.
Which makes me think your results might be drastically different with different data.
I tried running your benchmarks with my test file but it looks like you're expecting a particular format in the results so it failed. Also, I tried uploading my test file here so anyone who wanted to play with it could. But it exceeds the size GitHub will allow. But if anyone's interested in my test data, I'd be happy to make it available.
In SoftCircuits.CsvParser 4.0.0, the
ReadRow()
method is now a wrapper for the preferredRead()
method.On your next update, could you update SoftCircuits_CsvParser.cs to use the following
while
loop?Thanks.