Gyanangshu

5 ms·

1. 1.▲
   Ask HN: How do you handle wide-to-long without Pandas/R?

   1 points by Gyanangshu 7mo ago|1 comments
2. 2.▲
   by Gyanangshu 7mo ago

   That's the ideal, but unfortunately not always an option when you're on the receiving end. We're building a data cleaning tool, so the whole point is dealing with messy user-uploaded CSVs where we don't control the expor
3. 3.▲
   by Gyanangshu 7mo ago

   Thanks for sharing this! The context-based approach is interesting. Maintaining state from past successful parses to resolve ambiguity is essentially what we do too, though we batch it (scan first 50 values) rather than doing it incremental
4. 4.▲
   Ask HN: What's your strategy for inconsistent date formats?

   2 points by Gyanangshu 7mo ago|5 comments