Google apps script database abstraction exposed version

Database abstraction – GoogleApps Script
Desktop liberation

Google Apps Script libraries providing a
noSQL API for a variety of backends
Provides reusability and transportability of
code between databases, and simplified
migration between them
SimpleAPI to learn, with the most common
capabilities exposed and standardized
A consistent results structure
Locking, backing off , caching, authentication
and limits handled transparently

With dbAbstraction, noSQL queries are the same for all
supported backend stores. So far this includes
scriptdb
Parse.com
Fusion
Orchestrate.io
Fusion
Sheets
Import.io (readonly)
Drive
Properties
DataStore
MongoDB

A web app that exposes the GAS dbab as a
JSON rest API
Handles POST and GET queries, updates,
inserts, gets, count and removes
Exposes all dbab capabilities
Oauth2 is mandatory to authenticate to the
webapp
Authentication credentials for backend
databases are securely stored in your copy of
its property store

cDataHandler
library
cEzyOauth2
library
cCacheHandler
library
Dbab webapp
App template
Your property
store
Backend driver
Backend
database
cFlattener
library
noSQL queries
Access token
Google oauth2
infrastructure
Query caching
cNamedLock
library
Transaction locking
JSON API
Query translation
authentication
Cloud
credentials
Access & refresh tokens
(un)Flatten objects
Your client
Oauth2
authentication
JSON REST API

You authenticate to dbab via oAuth2 using
Drive scope
Your published copy of dbab webapp already
has your various authentication properties
built in for your supported databases
If the backend database needs oauth2, that’s
also handled by the dbab JSON api

Same format as the GAS API
{
handleCode: 0, // +ve number= good, -ve bad
handleError:’’, //’some text about an error’
data:[], // the results
driverIds:[], // if asked for, any keys the driver generates automatically
handlerIds:[] // if asked for, the key the driver considers to be the unique
key for each data item
+ various other informational properties about handler version etc…
}

Now it’s simple to write an API for another
language that looks like the standard GASAPI
GAS dbab handles the translation to various
backends, with new features and new supported
drivers immediately available
Your API just needs to know how to talk to the
GAS dbab JSONAPI
You can centralize your various credentials in the
cloud, protected by oauth2
Your code can look very similar to the same code
using the GAS API directly (language syntax
constraints aside)

• Gets handler, creates nosql directives, deals with
responsesVBA application
• Deals with oauth2 authentication to your web app.
Translates directives to JSON API requests and handles
responses
VBA API
• Deals with requests and responses
Dbab GAS REST API
• deals with authentication credentials, caching. Gets
handler and interprets nosql queriesDbab GAS API
• Translates nosql queries to backend syntax and handles,
authentication, locking, responses and requestsDbab GAS driver
• Handles queries and returns data and responses to its
driverVarious backends

Oauth2 integration withVBA is covered
elsewhere in detail
Create an application in Google Cloud
Console
If usingVBA
 Do a one off registration to yourWindows registry
 From now on you can simply do this to get
authorized
Set oauth2 = getGoogled("drive")

Copy from this template
Add any credentials needed for any backend
as described here
Save a version and publish as a webapp – (set
to run as “user accessing the webapp”)
Take a note of the webapp URL
Run doGet to authorize it
If needed, run the web app once with
?driver=datastore to provoke setting up of
the refresh token structure

Do any setup needed to create your
databases or sheets you are using as a
backend
You can now access your webapp as a REST
API to talk to the supported backends
If you want to use theVBA API, continue to
step 4

Get emptycdataset.xlsm from desktop
liberation downloads
Set up your handler and web app URL
Set handler = New cDbAb
handler.setOauth2(oauth2).setEndPoint (_
"https://script.google.com/macros/s/xxxxxx/exec")
Set up your database type name, database
and data silo
handler.setDbName("sheet") _
.setDbId(“xxxx") _
.setSiloId(“yyyy”)
You can start…

I’ve tried to keep theVBA syntax is as close to original
Google Apps Script as possible within the limitations
of the language.
VBA has no native JavaScript like objects or JSON, but
I use cJobject throughout
The cJobject JSON parser allows both single and
double quotes to be used in JSON eg.
{‘name’:’ethel’}
can be used so it can more be easily represented as aVBA string
“{‘name’:’ethel’}”
cDataset can be used to write and read excel
workbooks to and from JSON

Code What it does
Set result = handler.remove() Removes everything from the handler’s silo
Set result = handler.save(testData) Saves data in the given cJObject.You can
easily create a cJobject from an Excel sheet
like this
Set ds = New cDataSet
Set testData = ds.load("dbab").jObject(, , , ,
"data")
Set result = handler.query() Get everything in the handler’s silo
Set result = handler.query(, "{'limit':2}") Get the first 2 objects. Note that both single
quotes and double quotes are allowed by
this JSONParser to make it easier to enter
JSON as aVBA string
Set result = handler.query(, "{'sort':'-name'}") Get everything in the handler’s silo, sorted in
reverse order using the property ‘name’

Code What it does
Set result = handler.query(, "{'sort':'-
name','skip':3}")
Sort by name and skip the first 3
Set result = handler.query("{'name':'ethel'}") Select all objects whose name is ‘ethel’
Set result = handler.query(
("{'stuff':{'sex':'female'}}")
Select all objects where stuff.sex = ‘female’
Set result = handler.query(
("{'stuff.sex':'female'}")
Same as above. Note that you can use either
deep or flattened objects for both data and
queries. Regardless of how the driver has to
store the data (for example two dimensional
backends like sheets need to store flattened
data), the result of the query will be natural,
unflattened objects

Code What it does
Set result =
handler.query("{'stuff.sex':'male'}", , 1, 1)
Select all objects where name is john or
mary and stuff.sex is male, don’t use cache,
and return the unique keys recognized by
the driver for each object as well
handler.update(result.handleKeys, data) Replace all the objects with the keys
returned from the previous query with new
data
handler.remove("{'stuff.sex':'male'}") Remove all the objects where stuff.sex is
male
Set r2 =
handler.getObjects(result.handleKeys)
Get all the objects whose keys were
returned by a previous query

Code What it does
Set result = handler.query("{'name':" &
handler.constraints("[['IN',['ethel','fred']]]")
& "}"
Note that queries other than equality, need
constraints. Unfortunately the syntax is a bit
fiddly to construct as aVBA string.The
function handler.constraints() converts your
query into consumable syntax for the API.
This query selects all objects whose name is
in the given list
handler.query("{'stuff.age':" &
handler.constraints("[['GT',25],['LTE',60]]") &
"}")
Select all objects where stuff.age > 25 and
<= 60
handler.query("{'stuff':{'age':" &
handler.constraints("[['GT',25]]") & "}}")
Select all objects where stuff.age > 25
"{'stuff':{'sex':'male', 'age':" &
handler.constraints("[['GTE',25],['LT',60]]") &
"}}"
Select all objects where stuff.sex = ‘male’
and age is >= 25 and < 60

Code What it does
handler.query( "{'name':" &
handler.constraints("[['IN',['john','mary']]]")
& ",'stuff.sex':'male','stuff.age':" &
handler.constraints("[['GT',25]]") & "}"
Select all objects where name is john or
mary and stuff.sex is male and stuff,age > 25
handler.query("{'stuff.age':" &
handler.constraints("[['GT',25],['LTE',60]]") &
"}")
Select all objects where stuff.age > 25 and
<= 60
handler.query( "{'stuff.age':" &
handler.constraints("[['GT',25]]") & "}",
"{'limit':1}“)
Select the first object where stuff.age > 25
handler.query( "{'stuff':{'sex':'male', 'age':" &
handler.constraints("[['GTE',25],['LT',60]]") &
"}}“)
Select all objects where stuff.sex = ‘male’
and age is >= 25 and < 60

It’s simple to copy the database contents to a
worksheet
Do a query
Construct a sheet
Set result = handler.query()
Set ds = makeSheetFromJob(result.data, “yoursheetname”)
Since an excel sheet is two dimensional, any
deep objects will be flattened to dot syntax
so they can be represented as an excel table

It’s simple to copy the worksheet contents to a
database
Load the sheet and convert it to JSON
Save to your selected back end
Set testData = ds.load(“yoursheetname”).jObject(, , , , "data")
Set result = handler.save(testData)
Any flattened objects will be automatically
converted back to deep objects

Read about Google Apps Script data
abstraction here.
Read about the dbab JSONAPI here
Read about theVBA API client here.
Read about ezyOauth2 here.
Contact me on google plus or at my forum
and share with others how you are using this
API
Join the Google Apps Script Community

Google apps script database abstraction exposed version

More Related Content

What's hot (20)

Similar to Google apps script database abstraction exposed version (20)

Recently uploaded (20)

Google apps script database abstraction exposed version