Andrus Adamchik

BQSubpocessTestRuntime: a way to run Bootique test as a subprocess

1

Sometimes it is not enough to run a unit test "app" in a separate thread via BQDaemonTestRuntime. Instead a full subprocess is required. E.g. we can't test https://github.com/nhl/bootique-logback/issues/10 properly (see...

``` %maven org.apache.avro:avro:1.10.1 ``` The `%maven` magic above doesn't add Apache Avro library to classpath properly, resulting in an exception like the one below when trying to access any Avro...

Error accessing transitive dependencies via %maven

9

So I am working on the DFLib Jupyter integration per https://github.com/nhl/dflib/issues/44 . It mostly works well. Found one problem though. I am including all the DFLib modules as "compile" dependencies...

Best way to bootstrap custom render functions

4

Thanks for the iJava kernel. I started using it with pandas-inspired [DFLib](https://github.com/nhl/dflib/). Everything works great, and now I can do data science in pure Java ! :) A question if...

Let's create a function similar to Pandas and Spark that would expand a DataFrame (or a Series) based on an iterable values in cells of a given column. This is...

Storing DataFrame in Excel

3

Already have ExcelLoader per #136. Let's also implemented ExcelSaver.

Let's research performance optimizations offered by [JEP 338: Vector API](https://openjdk.java.net/jeps/338). Cursory glance shows a lot of overlap with what we are doing with primitive (and object) arrays. It is already...

Memory- and creation-optimized numeric Series

1

Primitive Series (`IntSeries`, `LongSeries`, etc.) are much more efficient than their object counterparts (`Series`, `Series`). They take up to 5x less memory per cell and provide opportunities to implement faster...

It is about time we start supporting Parquet format. Parquet Java library (`parquet-avro`) is not particularly friendly for standalone use outside Hadoop, but looks like it is changing. Version 1.14.0...

```java SaveStats stats = connector .tableSaver("t1") .mergeByColumns("name", "id") .deleteUnmatchedRows() .save(df); ``` When "deleteUnmatchedRows" is in effect, "stats" does not include the info on deleted rows. We need to include that.