Oliver Kennedy

Currently the error aware CSV parser is a mod of the existing Spark CSV parser. 1. There's a lot of overhead in the CSV parser for dealing with things that...

Add Schema Documentation for the Mimir API

2

As reported by @heikomuller

Can't create missing-key lens over another missing-key lens

1

As reported by @heikomuller

Propagate Schema Changes From Adaptive Schemas through Models

2

To see this bug in action: ``` mimir> load 'test/data/temperature.csv'; mimir> create lens repaired as select * from temperature with domain('temperature > -30'); mimir> feedback temperature 0 is real; mimir>...

Pivot Lens

1

Another use for the adaptive schemas, creating pivot tables. Consider for example: ``` time,location,temperature 1505958603,den,24.5 1505958604,basement,21.3 1505959204,den,24.5 1505959204,basement,21.400000000000002 1505959803,den,24.6 1505959804,basement,21.3 1505960265,office,17.5 1505960403,den,24.5 1505960404,basement,21.40000000000000 1505961003,den,24.5 1505961005,basement,21.400000000000002 1505961603,den,24.400000000000002 1505961604,basement,21.3 1505962203,den,24.400000000000002 ``` It...

Extend ANALYZE to detect potential data errors

1

As of right now, ANALYZE only detects sources of uncertainty injected by Mimir. It would be helpful if Mimir had some facility to do syntactic analysis on a dataset being...

`UNNEST`

1

It would be nice to have an UNNEST operator, essentially an inverse aggregate (one row/cell to many). Examples: * Iterate over the elements of a JSON array * Regexp matches...

PDF Table Import

1

Might be interesting to see if we can incorporate stuff from this system for extracting tabular PDF data: https://github.com/WZBSocialScienceCenter/pdftabextract

See if we can plug macrobase into Mimir's explanation system

1

https://github.com/stanford-futuredata/macrobase