README.md

September 19, 2025 · View on GitHub

DataStream R Package

This tool is useful for those who want to extract large volumes of data from DataStream. This R package allows users to call DataStream's Public API using R functions and specific search queries. The package includes several functions which accept a selection of filtering queries and returns a dataframe with the desired data from DataStream.

You might use this tool, for example, if you:

Cross-dataset downloads (i.e., to download all available pH data in Ontario on DataStream)
Want to count how many sites in New Brunswick have cesium data on DataStream

Request an API Token

To have full API permissions, users must request an API token which is required to call to the API

Installation

To install the most recent version in R:

# install.packages("remotes")
remotes::install_github("datastreamapp/datastreamr")

Attribution/Citation

Thank you ahead of time for using this data responsibly and providing the appropriate citations when necessary when presenting work to external parties. These dataset citations must be accompanied by a link to the DOI (https://doi.org/{value}). The dataset licence, citation, and DOI can be retrieved from the /Metadata endpoint.

Licence representations

The API returns the URL for a dataset's licence, these should be mapped to the full licence name with a link to the full licence details.

Attribution Licence:
- EN: Attribution Licence (ODC-By) v1.0
- FR: Licence d'attribution (ODC-By) v1.0
- URL: https://opendatacommons.org/licenses/by/1-0/
Public Domain Dedication and Licence:
- EN: Public Domain Dedication and Licence (ODC-PDDL) v1.0
- FR: Dédicace et licence du domaine public (ODC-PDDL) v1.0
- URL: https://opendatacommons.org/licenses/pddl/1-0/
Open Government Licence:
- EN: Open Government Licence (OGL)
- FR: Licence du gouvernement ouvert (OGL)
- URL: Dataset-dependent, entered by data provider (eg. https://open.canada.ca/en/open-government-licence-canada)

The Functions

The following functions are used to call DataStream's API and pull desired information.

setAPIKey():

Description
By default the environment variable "DATASTREAM_API_KEY" is used for setting the API key. Click here to request an api token

Usage

library(datastreamr)
# To set API Key for the current session, use:
setAPIKey('xxxxxxxxxx')

# Preferably, save the API key as an environmental variable
usethis::edit_r_environ()
# add DATASTREAM_API_KEY="xxxxxxxxxx" to the file, save, and restart R, then there is no need to include `setAPIKey()` within the script.

# Saving the API key as an environmental variable means it will remain private but be available when needed

metadata():

Description
Pulls only the dataset level metadata information including dataset name, citation, licence, abstract, etc.

Usage

metadata( 
  list(
    `$select` = NULL,
    `$filter` = NULL,
    `$top` = NULL,
    `$count` = "false"
  )
)

locations():

Description
Pulls only the location data including Location ID, Location Name, Latitude, and Longitude.

Usage

locations( 
  list(
    `$select` = NULL,
    `$filter` = NULL,
    `$top` = NULL,
    `$count` = "false"
  )
)

records():

Description
Pulls data formatted the same as the downloaded DataStream CSV’s including all columns listed in the DataStream schema .

Usage

This function will take longer than observations, but provides all available columns in one request.
Use this function if you aim to pull all location and parameter data in one call

records( 
  list(
    `$select` = NULL,
    `$filter` = NULL,
    `$top` = NULL,
    `$count` = "false"
  )
)

observations():

Description
Pulls data in a condensed format that must be joined with other endpoints to create a full dataset with all the DataStream columns. Specifically, location rows are not pulled, instead LocationId is pulled for each observation and then can be used in combination with locations().

Usage

This function will be quicker than records, but if location specifics are needed, needs to be paired with locations()
Use this function if you are uninterested in specific location coordinates, or in combination with locations() when you plan to pull >200,000 of rows of data

observations( 
  list(
    `$select` = NULL,
    `$filter` = NULL,
    `$top` = NULL,
    `$count` = "false"
  )
)

Function Inputs

All of the functions above accept query parameters. The ones supported are:

select: A list of allowable columns to return
- Fields to be selected are entered as a list.
- Example: select="DatasetName,Abstract"
- Default: All columns available.

$`\color{blue}{\text{Note}}`$ : refer to Allowed Values section below for available select fields

filter: A list of conditions to filter by
- Available operators:
- eq: Used for exact matches.
- ne: Used for not equal to.
- gt: Used for greater than.
- lt: Used for less than.
- ge: Used for greater than or equal to.
- le: Used for less than or equal to.
- and: Used to combine multiple filters with an “and” condition.
- Grouping: filter="CharacteristicName eq 'Dissolved oxygen saturation' and DOI eq '10.25976/n02z-mm23'"
- Temporal (Dataset creation): filter="CreateTimestamp gt 2020-03-23"
- Temporal (Data date-range): filter="ActivityStartYear gt '2019'"
- Spatial: filter=RegionId eq 'hub.atlantic'
  - RegionId Values (these values are subject to change):
  - DataStream Hubs: hub.{atlantic,lakewinnipeg,mackenzie,greatlakes,pacific }
  - Countries: admin.2.{ca}
  - Provinces/Territories: admin.4.ca.{ab,bc,mb,nb,nl,ns,nt,nu,on,pe,qc,sk,yt}

$`\color{blue}{\text{Note}}`$ : refer to Allowed Values section below for available filter fields

top:
- Maximum: 10000
- Example: top=10

count: When TRUE, returns number of observations rather than the data itself
- Return only the count for the request. When the value is large enough it becomes an estimate (~0.0005% accurate)
- Example: count=true
- Default: false
Performance Tips
- Use select to request only the parameters you need. This will decrease the amount of data needed to process and transfer.

Allowed Values

The allowed select and filter options for each of the functions are listed HERE.

$`\color{green}{\text{Note:}}`$ When using the filter field, a useful resource is the "allowed values" tab of our upload template . This will give you available strings for:

MonitoringLocationType
ActivityMediaName
CharacteristicName

Full examples

Locations

Get Locations from a dataset

setAPIKey('xxxxxxxxxx')

qs <- list(
    `$select` = "Id,DOI,Name,Latitude,Longitude",
    `$filter` = "DOI eq '10.25976/xxxx-xx00'",
    `$top` = 10000
  )
result = locations(qs)

Get Locations from multiple datasets

qs <- list(
    `$select` = "Id,DOI,Name,Latitude,Longitude",
    `$filter` = "DOI in ('10.25976/xxxx-xx00', '10.25976/xxxx-xx11', '10.25976/xxxx-xx22')",
    `$top` = 10000)
result = locations(qs)

Observations

Get Temperature and pH observations from multiple datasets

qs <- list(
    `$select` = "DOI,ActivityType,ActivityMediaName,ActivityStartDate,ActivityStartTime,SampleCollectionEquipmentName,CharacteristicName,MethodSpeciation,ResultSampleFraction,ResultValue,ResultUnit,ResultValueType",
    `$filter` = "DOI in ('10.25976/xxxx-xx00', '10.25976/xxxx-xx11', '10.25976/xxxx-xx22') and CharacteristicName in ('Temperature, water', 'pH')",
    `$top` = 10000)
result = observations(qs)

Records

Get select fields from a dataset

qs <- list(
    `$select` = "DOI,ActivityType,ActivityMediaName,ActivityStartDate,ActivityStartTime,SampleCollectionEquipmentName,CharacteristicName,MethodSpeciation,ResultSampleFraction,ResultValue,ResultUnit,ResultValueType",
    `$filter` = "DOI eq '10.25976/xxxx-xx00'",
    `$top` = 10000)
result = records(qs)

Metadata

Get the DOI, Version, and DatasetName for a dataset

qs <- list(
    `$select` = "DOI,Version,DatasetName",
    `$filter` = "DOI eq '10.25976/xxxx-xx00'",
    `$top` = 10000)
result = metadata(qs)

Get Result Count

qs <- list(
    `$filter` = "DOI eq '10.25976/xxxx-xx00'",
    `$count` = "true")
count = observations(qs)

Tests

Dockerfile is provided to run the unit tests and the integration tests. To build the docker image for running tests and other debugging purposes you can run:

docker build -t datastreamr .

To run the unit tests:

docker run --rm -e DATASTREAM_API_KEY=$(cat api_key.txt) datastreamr R -e "library(testthat); test_file('tests/testthat/test_unit.R')"

To run the integration tests:

docker run --rm -e DATASTREAM_API_KEY=$(cat api_key.txt) datastreamr R -e "library(testthat); test_file('tests/testthat/test_integration.R')"