Larefly

Compressed Parquet group by # 2: Note: Data was taken from: #112 (comment) ```r res

Branch `data-types` attempt to parse AWS Athena arrays, maps and row data types. ```r library(DBI) con

What the above AWS Athena Queries return from AWS Athena: ``` query1: +-------+ | items | +-------+ | [4,5] | +-------+ query2: +--------------------------------------------------+ | items | +--------------------------------------------------+ | [{"a1":1,"a2":2,"a3":3}, {"b1":4,"b2":5,"b3":6}]...

Returning R formats when `convert_array = TRUE` ```r df1: # items # 1: df1$items[[1]] # [[1]] # [1] 4 # # [[2]] # [1] 5 df2: # items # 1:...

As Json can be parsed different ways, it might be worth letting the user provide custom parsers if they don't like the default. For example: ```r library(DBI) # use default...

Another question: Should `convert_array` parameter be in `dbConnect`? Would it be better to have in `noctua_options` so that arrays can be parsed in multiple ways .... or even both?

Added support to more AWS Athena Data types: varbinary, ipaddress and json. I am pretty happy with the binary method currently proposed in branch: `data-types` ```r library(DBI) library(data.table) # default...

For completeness here is the dplyr method for above data.table: ```r library(DBI) library(dplyr) noctua::noctua_options("vroom") # default binary conversion connection con1

@OssiLehtinen been pondering on it all last night thinking it over :P I am guessing you would say the same for the `binary` and `bigint` parameters? I am more than...

Added parameters for `bigint`, `binary` and `json` conversion within `noctua_options`. `noctua_options` allows for conversion methods to be changed with the bonus of not affecting other conversion methods if a user...