-
Notifications
You must be signed in to change notification settings - Fork 0
JSON Type Definition
Peter edited this page Dec 1, 2017
·
6 revisions
Standard JSON-schemas used in the j
attribute of dataset.big
table.
name | Description | about |
---|---|---|
bag | JSON Bag (multiset), pairs of element-multiplicity. For optimized access and specific operations. | ref |
tab-aoa | JSON Tabular Array of Array-rows, one SQL field with all rows. Separated header. Best. | ref |
tab-1apr | Tabular, distributed in many SQL rows, one JSON-array per SQL-row. | ref |
objs-free | Objects, free-structure, all dataset in only one JSON set of objects. Non tabular (ex. free taxonomy) | |
json-ld | JSON for Linking Data, W3C standard v1.0 of 2014 | ref |
tab-aoo | JSON Tabular Array of Object-rows, one SQL field with all rows. Please avoid! prefer tab-aoa. | ref |
tab-1opr | Tabular, distributed in many SQL rows, each row with an Object. Please avoid! prefer tab-1apr. | |
tab-aoa-trans1 | A tab-aoa with Transactional-expenditure-data-v1 (see URL) semantics and its minimal structure. | ref |
The only valid JSON-schema for dataset.meta.info
is the FrictionLessData/spec/data-package v1.0.
Other types, valid only for output:
name | Description | about |
---|---|---|
tabh-aoa | tabh as "tab with header", for output. After header, same as tab-aoa. | ref |
All examples using near the same content, except non-tabular JTD.
A typical list of sample-frequency in statistical data, or element-multiplicity in generalized sets.
{
"maria": 909900,
"josé": 9221,
"carolina": 500
}
{
"@context": "http://schema.org/",
"@type": "Person",
"name": "Jane Doe",
"jobTitle": "Professor",
"telephone": "(425) 123-4567",
"url": "http://www.janedoe.com"
}
Any valid JSON-LD.
{
"id": 123,
"person": {
"name": "Jane Doe",
"jobTitle": "Professor",
"telephone": "(425) 123-4567"
},
"url": "http://www.janedoe.com"
}
Non-tabular, no limits to hierarchical levels.
At SQL, suppose 2 rows of dataset.big
, and its JSONb attribute j
,
row | j |
---|---|
1 | [ 1, 2, 3, "x" ] |
2 | [ 4, 5, 6, "y" ] |
At SQL, suppose 2 rows of dataset.big
, and its JSONb attribute j
,
row | j |
---|---|
1 | { "A":1, "B":2, "C":3, "S":"x" } |
2 | { "A":4, "B":5, "C":6, "S":"y" } |
[
[ 1, 2, 3, "x" ],
[ 4, 5, 6, "y" ]
]
The header stay at dataset.meta
, indirectly at info
or directly, cached as kx_fields
. To output as tab-aoa-header use SELECT dataset.f(j)
.
[
[ "Prefeitura de Campinas", "2016-11-29", "XPTO suprimentos Ltda.", "ctr234", "545665",
"ab3456", "CGM - Controladoria Geral do Município"],
[ "Prefeitura de Campinas", "2016-12-03", "Casas Bahia SA.", "ctr464", "736443", "bc1458", "CGM"]
]
[
{ "A": 1, "B": 2, "C": 3, "S":"x" },
{ "A": 4, "B": 5, "C": 6, "S":"y" }
]
[
{"administrator":"Prefeitura de Campinas", "date":"2016-11-29", "supplier":"XPTO suprimentos Ltda.",
"contract":"ctr234", "budgetLineItem":"545665", "invoiceID":"ab3456",
"procurer":"CGM - Controladoria Geral do Município"
},
{"administrator":"Prefeitura de Campinas", "date":"2016-12-03", "supplier":"Casas Bahia SA.",
"contract":"ctr464", "budgetLineItem":"736443", "invoiceID":"bc1458", "procurer":"CGM"
}
]
[
[ "A", "B", "C", "S" ],
[ 1, 2, 3, "x" ],
[ 4, 5, 6, "y" ]
]
To change this page, change also this spreadsheets.