R package for accessing google datastore
R
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
R
man
tests
.Rbuildignore
.gitignore
.travis.yml
DESCRIPTION
LICENSE
NAMESPACE
README.md
Rdatastore.Rproj
client-secret.json.enc

README.md

Build Status

rdatastore

R package for accessing google datastore

To Do

  • Add tests for autoallocation of ids when committing.
  • Add tests for keep_existing T/F.
  • Add test for datetime consistancy.
  • When committing, return data and transaction id.
  • Add further Blob support.
  • Prohibit list, vector, and matrix storage; allow for object storage.

Functions

  • lookup - Lookup single entities. Returns a data frame.
  • commit - Update a single entity.
  • commit_df - Save a dataframe to the google datastore with update/upsert/insert/delete (consistant data).
  • commit_list - Save a list (heterogeneous data).
  • runQuery - Function for querying. Possibly using dplyr paradigms.
  • datasets - A centralized function for tracking/accessing datasets. Read only store.
  • [/] gql - Add ability to return all results (batches).

Features

  • keep_existing When updating - keep existing values, add new ones, and overwrite where specified.
  • Blob Store - limited.

commit_df()

User specifies a key (unique value) for a given dataframe

Planned Arguments:

  • use row numbers/names - Use row names for 'names' column and/or numbers; Retreived data will be slightly modified.
  • keep_existing - retain existing data; only update values/insert new ones.

datasets()

The aim of the datasets function is to centralize storage of datasets and make them easily searchable.

gql()

Used for running queries.