README.md

In chok68/havenondemand-r: HPE Haven OnDemand R Language Client Library - POST requests

HPE Haven OnDemand R Client Library

Overview

This library can be used to consume HPE Haven OnDemand - https://dev.havenondemand.com/apis from R language scripts.

What is HAVEN ONDEMAND?

Haven OnDemand is a set of over 70 APIs for handling all sorts of unstructured data. Here are just some of our APIs' capabilities:

• Speech to text
• OCR
• Text extraction
• Indexing documents
• Smart search
• Language identification
• Concept extraction
• Sentiment analysis
• Web crawlers
• Machine learning

For a full list of all the APIs and to try them out, check out https://www.havenondemand.com/developer/apis.

Installation

Development version from github:

install.packages("devtools") devtools::install_github("chok68/havenondemand-r")

Usage

Initialization

# include havenondemand library
library(havenondemand)

# with apikey and version. NOTE: replace your-api-key with one of your API keys (from your HPE OnDemand account keys) 
client <- HODClient(apikey = "your-api-key", version = "v1")

# with apikey (version will default to v1). NOTE: replace your-api-key with one of your API keys (from your HPE OnDemand account keys) 
client <- HODClient(apikey = "your-api-key")

Sample post sync call (INDEX_STATUS)

NOTE: in this sample, we check index status for an existing index called 'test'. If you don't have this index created call will fail. Use another index name if this is the case.

# include havenondemand library
library(havenondemand)

# initialize HOD Client. NOTE: replace your-api-key with one of your API keys (from your HPE OnDemand account keys) 
client <- HODClient(apikey = "your-api-key")

# Make api call
result <- tryCatch({
    client$postRequest(hodApp = HODApp$INDEX_STATUS, params = list(index = "test"), mode = HODClientConstants$REQUEST_MODE$SYNC)
}, warning = function(w) {
    print('Warning block called.')
    print(w)
}, error = function(e) {
    print('Error block called.')
    print(e)
}, finally = {
    print('Finally block called.')
})

# print result
print(result)

Sample post async call (list resources)

# include havenondemand library
library(havenondemand)

# initialize HOD Client. NOTE: replace your-api-key with one of your API keys (from your HPE OnDemand account keys) 
client <- HODClient(apikey = "your-api-key")

result <- tryCatch({

    # create client get job id
    r <- client$postRequest(hodApp = HODApp$LIST_RESOURCES, params = 
           list(flavor = 'standard', flavor = 'explorer', type = 'content', type = 'connector')
           , mode = HODClientConstants$REQUEST_MODE$ASYNC)

    # get job id
    jobID = r$jobID
    print (jobID)

    # check job status using the getJobStatus method call
    # If job status is finished then getJobStatus method itself will return job data as part of response. If job is not finished yet, it will return queued response.
    jobStatus <- client$getJobStatus(jobID)
    print (jobStatus)

    # get job data using the getJobResult method
    jobResult <- client$getJobResult(jobID = jobID)
    print(jobResult)

}, warning = function(w) {
    print('Warning block called.')
    print(w)
}, error = function(e) {
    print('Error block called.')
    print(e)
}, finally = {
    print('Finally block called.')
})

# print result
print(result)

Sample post async call with multiple file attachments (Text Extraction API)

# include havenondemand library
library(havenondemand)

# initialize HOD Client. NOTE: replace your-api-key with one of your API keys (from your HPE OnDemand account keys) 
client <- HODClient(apikey = "your-api-key")

result <- tryCatch({

    # create client get job id

    # upload a file named 'broch.pdf' in drive c: (windows os), adjust path to your os / file name.
    r <- client$postRequest(hodApp = HODApp$TEXT_EXTRACTION, params = list(
            file = httr::upload_file("c:/broch.pdf"),
            file = httr::upload_file("c:/HP_License_terms_may2012.doc"))
           , mode = HODClientConstants$REQUEST_MODE$ASYNC)

    # get job id
    jobID = r$jobID
    print (jobID)

    # check job status using the getJobStatus method call
    # If job status is finished then getJobStatus method itself will return job data as part of response. If job is not finished yet, it will return queued response.
    jobStatus <- client$getJobStatus(jobID)
    print (jobStatus)

    # get job data using the getJobResult method
    jobResult <- client$getJobResult(jobID = jobID)
    print(jobResult)

}, warning = function(w) {
    print('Warning block called.')
    print(w)
}, error = function(e) {
    print('Error block called.')
    print(e)
}, finally = {
    print('Finally block called.')
})

# print result
print(result)

Index management.

This code illustrates how to create, add to, and query index. This code is also explained in the Wiki: https://github.com/chok68/havenondemand-r/wiki/Hello-World-R-Script

# include havenondemand library
library(havenondemand)

# initialize HOD Client. NOTE: replace your-api-key with one of your API keys (from your HPE OnDemand account keys) 
client <- HODClient(apikey = "your-api-key")

result <- tryCatch({

    # STEP 1: call create text index (this call is required
    #    one time only per index life)
    client$postRequest(hodApp = HODApp$CREATE_TEXT_INDEX, params = list(index = "myindex", flavor = "explorer"),
       mode = HODClientConstants$REQUEST_MODE$SYNC)

    # STEP 2: call add to text index (we're indexing this entry:
    #    https://en.wikipedia.org/wiki/Hewlett_Packard_Enterprise_Software
    client$postRequest(hodApp = HODApp$ADD_TO_TEXT_INDEX, params = list(url = "https://en.wikipedia.org/wiki/Hewlett_Packard_Enterprise_Software",
       index = "myindex"), mode = HODClientConstants$REQUEST_MODE$SYNC)

    # STEP 3: call add to text index (we're indexing this entry:
    #    https://en.wikipedia.org/wiki/HP_Information_Management_Software
    client$postRequest(hodApp = HODApp$ADD_TO_TEXT_INDEX, params = list(url = "https://en.wikipedia.org/wiki/HP_Information_Management_Software",
        index = "myindex"), mode = HODClientConstants$REQUEST_MODE$SYNC)

    # STEP 4: query text index
    r <- client$postRequest(hodApp = HODApp$QUERY_TEXT_INDEX, params = list(text = "their applications and databases are growing", 
        indexes = "myindex"), mode = HODClientConstants$REQUEST_MODE$SYNC)
    d = r$documents

    # STEP 5: show the reference (for document index=1) that matched our search
    print(d[[1]]$reference)

}, warning = function(w) {
    print('Warning block called.')
    print(w)
}, error = function(e) {
    print('Error block called.')
    print(e)
}, finally = {
    print('Finally block called.')
})

Error Handling

All the calls call stop(message) if an error is detected:

# include havenondemand library
library(havenondemand)

# initialize HOD Client. NOTE: replace your-api-key with one of your API keys (from your HPE OnDemand account keys) 
client <- HODClient(apikey = "your-api-key")

# call that result in error ('ur' parameter is wrong, it should be 'url')

result <- tryCatch({
    client$postRequest(hodApp = HODApp$SPEECH_RECOGNITION, params = list(ur = "https://www.havenondemand.com/sample-content/videos/hpnext.mp4"), HODClientConstants$REQUEST_MODE$ASYNC)
}, warning = function(w) {
    print('Warning block called.')
    print(w)
}, error = function(e) {
    print('Error block called.')
    print(e)
}, finally = {
    print('Finally block called.')
})

HODClient Instance Methods

    # calls POST Request
    #
    # @param params params to be passed
    # @param hodApp end point to be called
    # @param mode sync/async
    # @return json response
    postRequest = function(hodApp, params, mode)

    # Get status of the job submitted
    # @param jobId id of the job submitted
    getJobStatus = function(jobID)

    # Get result of the job submitted
    # @param jobId id of the job submitted
    getJobResult = function(jobID)

Contributing

TODO: Pending from HPE Haven OnDemand Team

chok68/havenondemand-r documentation built on May 13, 2019, 5:32 p.m.