Data Publication Requirements & Rules

Owned by Steven Revill, created
Last updated: Mar 19, 2021 by Katy Aird
2 min read

The data you publish into uSmart to enable reuse can be in a number of formats from pdf and images to shapefiles. However, if you are adding data for the specific purposes of integrated reuse and manipulation then it is most likely to be an API or a bulk csv upload. For both there are some key rules to be cognisant of:

Seeding a dataset

Seeding a dataset (creating a single sample row with complete data included) is not always necessary but it can have the following benefits. It ensures the dataset knows the type of data to be included into each column for example date, integer, string.

This is relevant as if a column is left blank in the first row it will be assumed to be an integer.

Data Type

Whether seeding or not it is important to consider that data added will be treated and assumed to be a certain type of data depending on the payload content. For example, if a number is a phone number then it is recommended to write it as an integer such as +44(0)7921… if the data is added as 07921… it will assume it is a number and will remove the initial zero storing as 7921…

Blank rows

Rows of data with no data payload (even if this includes column separators) will cause the system to end the data ingest process. So for example data such as this will not be ingested into the system:

column A, column B, column C, column D

1, 2, 3, 4,

1, 2, 3, 4,

, , , ,

1, 2, 3, 4,

Please ensure therefore that no blank rows are added.

Date format

uSmart follows the ISO 8601 format for date and time. If your data is uploaded in this format uSmart will detect this as a date/ time format and enable visualisations and graphing based on time series. uSmart may also pick up a date/ time format if it is in a different format however it may end up assuming it is an integer or string.

Decimal points

Data should not be uploaded with a decimal point at the beginning. E.g. “.8” should be written as “0.8”.