5 ms·
I enjoy using clickhouse-local for parsing csv files. I generally hit situations where I need custom delimiter and custom parsing rules, I find it handles all
by pradeepchhetri 3y ago
I enjoy using clickhouse-local for parsing csv files. I generally hit situations where I need custom delimiter and custom parsing rules, I find it handles all of these edge cases very well. Recently I found that if my csv files are compressed, i don't even need to uncompress them, it auto-magically figures out the compression format and process it for me.
- tambourine_man 3y agoDidn’t know about them, seems interesting.
- magundu 3y agoWhich one is better for querying CSV from both command line and programmatically? 1. SQLite 2. DucksDB 3. clickhouse-local
- pradeepchhetri 3y agoDefinitely clickhouse ecosystem. - For querying csv data from command line, I use clickhouse-local. - For querying csv data programmatically using a library, I use chdb (embedded version of clickhouse) - For querying large amount of csv data programmatically, I offload it to clickhouse cluster which can do processing in distributed fashion. If you are looking from query performance perspective, this blog is useful: https://www.vantage.sh/blog/clickhouse-local-vs-duckdb https://www.vantage.sh/blog/clickhouse-local-vs-duckdb