[Discussion][Document] The content of libraries implementation status page

[acezen]

as issue #352 #362 describe, we need to provide an implementation status to indicated the format implementation status of GraphAr libraries.
I draft a page about the implementation status and maybe I'm missing something important. So I want to discuss first.

[acezen]

Implementation Status

The following tables summarize the features available in the various official GraphAr libraries. All libraries currently follow version 1.0.0 of the GraphAr format.

Property

Property feature	C++	Java	Scala	Python
primary key	✓	✓	✓	✓
nullable	✓		✓	✓ (2)

Notes:

• (1) Through py4j bindings to the GraphAr spark library.

Property Group

Property Group	C++	Java	Scala	Python
create	✓	✓	✓	✓
add	✓	✓	✓	✓
delete

Adjacency List

Adjacency List	C++	Java	Scala	Python
CSR	✓	✓	✓	✓
CSC	✓	✓	✓	✓
COO	✓	✓	✓	✓

Data Types

Primitive

Data type	C++	Java	Scala	Python
Boolean	✓	✓	✓	✓
Int32	✓	✓	✓	✓
Int64	✓	✓	✓	✓
Float	✓	✓	✓	✓
Double	✓	✓	✓	✓
String	✓	✓	✓	✓
Date
Time
Timestamp

Nested

Data type	C++	Java	Scala	Python
List	✓ (1)

Notes:

• (1) The list data type is not supported by CSV payloads.

Payload Data File Formats

Format	C++	Java	scala	python
Avro
CSV	R/W	R (1)	R/W	R/W (2)
ORC	R/W	R (1)	R/W	R/W (2)
Parquet	R/W	R (1)	R/W	R/W (2)

Notes:

• R = Read supported
• W = Write supported
• (1) Through FastFFI bindings to the GraphAr C++ library.
• (2) Through py4j bindings to the GraphAr spark library.

Graph Metadata

Graph Metadata	C++	Java	Scala	Python
Graph name	✓	✓	✓	✓
Version	✓	✓	✓	✓
Extra info	✓

[acezen]

cc\ @lixueclaire @Thespica @SemyonSinchenko Please let me know your opinion about the draft.

[Thespica]

LGTM

Btw, Should we add diff of schema in Java SDK as we had discussed here #346 ?

[acezen]

I think the JAVA SDK should update to align the schema, the diff is just a ad-hot solution for passing the CI

[lixueclaire]

cc\ @lixueclaire @Thespica @SemyonSinchenko Please let me know your opinion about the draft.

Hi, @acezen , this proposal looks promising. I recommend adding brief details on the following: Compression method and Supported Reader/Writer Levels (low-level, mid-level, high-level).

[acezen]

this is a good advice!