Problemsby IbisPolars costs outgrow the value it gives
Cloud and tool cost2 posts from 2 people2d active+00 posts in the last 7 days, 1 the 7 days before (fading)Unverified
Posts per day
The posts behind it
2, newest first`sink_iceberg`: support the `bucket` partition transform### Description `sink_iceberg` refuses tables partitioned by `bucket`: The `sink_iceberg` docstring lists bucket (and decimal truncate) as unsupported. Bucket partitioning is common on high-cardinality keys, and these tables can only be written through PyIceberg's `append`, not...pola-rs/polars · djouallah1 reactions, 1 repliesOpen ↗Parquet writer: dictionary-encodes high-cardinality columns where plain is much smaller; no way to opt out per column### Description `write_parquet` / `sink_parquet` dictionary-encode every eligible column. For high-cardinality columns that aren't fully unique (IDs that repeat 2-5×, block numbers, timestamps), the result is much larger than plain encoding, and there's no option to turn the...pola-rs/polars · bartek671Open ↗
Companies and products named
About this problem
Evidence
2 posts from 2 people in 1 place, about 2 a week over 4 days. Mostly on pola-rs/polars. Tools named alongside: DuckDB, GitHub.
Frustration Frustration 0 of 3· Seen on GitHub issues
How it was grouped
Posts that state a pain and match the "observability-cost" rule about Polars. First post Oct 6, 2026, latest Oct 9, 2026. Unverified: fewer than 3 posts, or one person or place only. Method
The same problem elsewhere
Other Polars problems
Polars users keep filing bugsPolars upgrades break working setupsTrouble with Polars streamingTrouble with Polars columnTrouble with Polars parquetPolars runs out of memoryTrouble with Polars joinBad data slips through Polars unnoticedTrouble with Polars structTrouble with Polars rollingTrouble with Polars dateTrouble with Polars mismatch
History
steady to fading
Polars costs outgrow the value it gives (2 posts)