fdc-api

Provides a REST server to query and retrieve USDA FoodData Central datasets. You can browse foods from different sources, perform simple searches, access nutrient data for individual foods and obtain lists of foods ordered by nutrient content. A rudimentary authentication/authorization framework is also included. A demonstration site for trying the usage examples below is available go.littlebunch.com. The site uses the Branded Food Products dataset. OpenAPI 3.0 documentation is available as HTML at https://go.littlebunch.com/doc. The specification can also be retrieved as JSON or YAML from the /docs endpoint described below.

What's in the repo

/api -- source for the REST web server
/docker -- files used for building docker images of the API server
/ds -- source for the data source interface. Implementations should also go here
/ds/cb -- couchbase implementation of the ds interface
/ds/cdb -- couchdb implementation of the ds interface (work-in-progress)
/model -- go types representing the data models

Quick word about datastores

I've done versions of this API in MySQL, Elasticsearch, CouchDB and Mongo but settled on Couchbase because of N1QL and the built-in full text search engine. I've heard it scales pretty good as well. :) It's also possible without a great deal of effort to implement a MongoDb, ElasticSearch or relational datastore by implementing the ds/DataSource interface for your preferred platform.

Building

The steps below outline how to go about building and running the applications using Couchbase. Additional endpoint documentation is provided by a swagger.yaml and a compiled apiDoc.html in the api/dist path. A docker image for the web server is also available and described below.

The build requires go version 12. If you are using Couchbase, then version 6 or greater is preferred. Both the community edition or licensed edition will work. Or, if you don't want to bother with the go stuff, you can launch a server from docker.io (see below under Running).

Step 1: Clone this repo

Clone this repo into any location other than your $GOPATH:

git clone git@github.com:littlebunch/fdc-api.git

and cd to the repo root, e.g.:

cd ~/fdc-api

Step 2: Build the binary

The repo contains go.mod and supporting files so a build will automatically install and version all needed libraries. If you don't want to use go mod then rm go.mod and go.sum and have at it the old-fashioned way. For the webserver:

go build -o $GOBIN/fdcapi api/main.go api/routes.go

You're free to choose different names for -o binary as you like.

You can also use the Docker file to create an image for the web server.

Step 3: Install and build a datastore

If you want to use Couchbase then use the ingest utility available at https://github.com/littlebunch/fdc-ingest.

Step 4. Start the web server (see below)

Configuration

Configuration is minimal and can be in a YAML file or envirnoment variables which override the config file.

couchdb:   
  url:  localhost   
  bucket: gnutdata   //default  bucket    
  fts: fd_food  // default full-text index   
  user: <your_user>    
  pwd: <your_password>    

Environment

COUCHBASE_URL=localhost   
COUCHBASE_BUCKET=gnutdata   
COUCHBASE_FTSINDEX=fd_food   
COUCHBASE_USER=user_name   
COUCHBASE_PWD=user_password   

Running

The instructions below assume you are deploying on a local workstation.

Start the web server:

$GOBIN/fdcapi -c /path/to/config.yml  
where    
  -d output debugging messages     
  -c configuration file to use (defaults to ./config.yml )      
  -p TCP port to run server (defaults to 8000)    
  -r root deployment context (v1)    
  -i username:password bootstrap an admin user and password
  -l send stdout/stderr to named file (defaults to /tmp/bfpd.out

Or, run from docker.io (you will need docker installed):

docker run --rm -it -p 8000:8000 --env-file=./docker.env littlebunch/fdcapi

You will need to pass in the Couchbase configuration as environment variables described above. The easiest way to do this is in a file of which a sample is provided in the repo's docker path.

Usage

A apiDoc.yaml OpenAPI 3.0 document which fully describes the API is included in the dist path. An html version is available to view at https://go.littlebunch.com/doc.

Fetch a single food by FoodData Central id=389714:

curl -X GET https://go.littlebunch.com/v1/food/389714 

Fetch a single food by GTIN/UPC 042222850325

curl -X GET https://go.littlebunch.com/v1/food/042222850325

Fetch all nutrient data for a food

curl https://go.littlebunch.com/v1/nutrients/food/389714  

Fetch nutrient data for a single nutrient for a food identified by GTIN/UPC

curl https://go.littlebunch.com/v1/nutrients/food/042222850325?n=208 

Fetch food data for a list of FoodData Central ids:

Returns list of foods identified by an exploded array of up to a maximum 24 id's. The array may contain a mix of GTIN/UPC codes and FDC IDs.

curl 'https://go.littlebunch.com/v1/foods?id=344604&id=042222850325&id=344606'  

Fetch nutrient data for a list of FoodData Central ids

curl 'https://go.littlebunch.com/v1/nutrients/foods?id=344604&id=042222850325&id=344606'  

Fetch nutrient data for a single nutrient for a list of FoodData Central ids

curl 'https://go.littlebunch.com/v1/nutrients/foods?id=344604&id=042222850325&id=344606?n=208'  

Browse foods:

curl 'https://go.littlebunch.com/v1/foods/browse?page=1&max=50&sort=foodDescription'
curl 'https://go.littlebunch.com/v1/foods/browse?page=1&max=50&sort=company&order=desc'    

Search foods (GET):

Perform a simple keyword search of the index. Include quotes to search phrases, e.g. ?q='"bubbies homemade"'. For more complicated and/or precise searches, use the POST method.

curl 'https://go.littlebunch.com/v1/foods/search?q=bread&page=1&max=100'    
curl 'https://go.littlebunch.com/v1/foods/search?q=bread&f=foodDescription&page=1&max=100'   

Search foods (POST):

Perform a string search for 'raw broccoli' in the foodDescription field:

curl -XPOST https://go.littlebunch.com/v1/foods/search -d '{"q":"broccoli raw","searchfield":"foodDescription","max":50,"page":0}'

Perform a string search for 'corn flakes' in the foodDescription field and limit results to the cereal foodgroup:

curl -XPOST https://go.littlebunch.com/v1/foods/search -d '{"q":"corn flakes","searchfield":"foodDescription","foodgroup":"cereal","max":50,"page":0}'

Perform a WILDCARD search for company names that match rond:

curl -XPOST https://go.littlebunch.com/v1/foods/search -d '{"q":"ro*nd*","searchfield":"company","searchtype":"WILDCARD","max":50,"page":0}'

Perform a PHRASE search for an exact match on "broccoli rabe" in the "ingredients field:

curl -XPOST https://go.littlebunch.com/v1/foods/search -d '{"q":"broccoli rabe","searchfield":"ingredients","searchtype":"PHRASE","max":50,"page":0}'

Perform a REGEX (regular expression) search to find all foods that begin with "Olive" in the foodDescription field:

curl -XPOST https://go.littlebunch.com/v1/foods/search -d '{"q":"^OLIVE+(.*)","searchfield":"foodDescription","searchtype":"REGEX","max":50,"page":0}'

Peform a REGEX search to find all foods that have UPC's that begin with "01111" and end with "684"

curl -XPOST https://go.littlebunch.com/v1/foods/search -d '{ "q":"^01111\\d{2,4}684","searchtype":"REGEX","searchfield":"upc"}'

Fetch documentation

Download OpenAPI 3.0 specification rendered as JSON or YAML

curl https://go.littlebunch.com/v1/docs/json

curl https://go.littlebunch.com/v1/docs/yaml

Fetch dictionary lists

Nutrient List

curl https://go.littlebunch.com/v1/dictionary/NUT

Food Group list

curl https://go.littlebunch.com/v1/dictionary/FGGPC

Derivation codes

curl 'https://go.littlebunch.com/v1/dictionary/DERV'

Run a nutrient report sorted in descending order by nutrient value per 100 units of measure

Find foods which have a value for nutrient 208 (Energy KCAL) between 100 and 250 per 100 grams

curl -X POST https://go.littlebunch.com/v1/nutrients/report -d '{"nutrientno":207,"valueGTE":10,"valueLTE":50}'