3 ways to perform HTTP requests

Ever wonder how to perform HTTP requests from WarpScript? Depending on your needs, here are 3 ways to achieve it.

Often we need to perform HTTP requests in our scripts, for instance, to send a notification in a Slack channel or to a mailing API, to fetch data from a REST API or to send data to a tier server.

In Warp 10 2.x, there was 3 ways to perform HTTP requests in WarpScript or FLoWS.

Update: Since Warp 10 3.0, there is only one way : the HTTP function.

HTTP

HTTP allows you to send a GET or POST HTTP request (thanks Captain Obvious), waiting for a response.

Among the different options, you can even stream your data using the chunks.

Following Warp 10 3.0 trend, most of the configurations can be superseeded by capabilities.

Configuration parameters in /path/to/warp10/etc/conf.d/70--extensions.conf:

warpscript.extension.http=io.warp10.script.ext.http.HttpWarpScriptExtension: Activate the extension
warpscript.http.maxrequests: maximum parallel usage
warpscript.http.maxsize: maximum response size
warpscript.http.maxchunksize: maximum chunk size
warpscript.http.host.patterns: allow/disallow some hosts based on a comma separated list of regular expression like slack.com,.*!

WarpScript

First, you need to generate a token to be able to use HTTP: create a mc2 file with this content


{
  'id' 'TokenRead'
  'type' 'READ'
  'application' 'nonexistingapp' 
  'owner'  "c7636029-d7b5-4b7a-9aab-65d61419d719"
  'issuance' 0
  'expiry' [ 2200 01 01 ] TSELEMENTS->
  'labels' { }   
  'attributes' { 
    '.cap:http' 'localhost,127.0.0.1,api.mysite.fr'        // able to use HTTP
    '.cap:http.requests' '100000' // number of HTTP call allowed per script
    '.cap:http.size' '100000000' // raise HTTP download limit to 100MB
    '.cap:http.chunksize' '1000000' // raise HTTP size chunk to 1MB
  } 
  'applications' [ 'nonexistingapp' ]  
} TOKENGEN

Run it. By default, TOKENGEN is not accessible, so you need an access to the server to run /opt/warp10/bin/warp10.sh tokengen /path/to/this/file.mc2

Then you can use the generated token in your script with CAPADD, to enjoy HTTP!

'tokenWithHttpCapability' CAPADD
{
 'url' 'https://hooks.slack.com/services/T01xxxxNT1/B0xxxM/bzZxxx5yHdJJ'
 'method' 'POST'
 'headers' { 'Content-Type' 'application/json' }
 'body' {
   'channel' '#alertes'
   'username' 'Warp 10'
   'text' "Published in #alertes from user Warp 10."
   'icon_emoji' ':spock-hand:'
 } ->JSON
}
HTTP

WEBCALL (deprecated)

WEBCALL is historically the first one. It allows you to send an HTTP request without waiting for a response. This is a one-way action. WEBCALL is handy to notify an HTTP Server without slowing your code execution or introducing a bottleneck.

Configuration parameters in /path/to/warp10/etc/conf.d/10-webcall.conf:

webcall.host.patterns: allow/disallow some hosts based on regular expression like slack.com,!^127.0.0.1$,!^localhost$,^192.168.*
warpscript.maxwebcalls: maximum parallel usage
http.header.webcall.uuid: will appear in the X-Warp10-WebCall header
webcall.user.agent: "Warp10-WebCall" by default

WarpScript

'9M8DP97xxxxxxZwKm78zc' // WRITE token
'POST'
'https://hooks.slack.com/services/T01xxxxNT1/B0xxxM/bzZxxx5yHdJJ'
{ 'Content-Type' 'application/json' } // HTTP headers
{
  'channel' '#alertes'
  'username' 'Warp 10'
  'text' "Published in #alertes from user Warp 10."
  'icon_emoji' ':spock-hand:'
} ->JSON // Payload
WEBCALL

URLFETCH (deprecated)

URLFETCH allows you to perform a GET request and give you back the response, headers and status. It could be a bottleneck in your code, because it waits for the response.

For security reasons, you must be authenticated with a READ token.

By the way, this function will be deprecated in a future Warp 10 version because replaced by HTTP.

Configuration parameters in /path/to/warp10/etc/conf.d/70--extensions.conf:

warpscript.extension.urlfetch = io.warp10.script.ext.urlfetch.UrlFetchWarpScriptExtension: Activate the extension
warpscript.urlfetch.limit: maximum parallel usage a developer can bypass
warpscript.urlfetch.limit.hard: maximum parallel usage
warpscript.urlfetch.maxsize: maximum response size a developer can bypass
warpscript.urlfetch.maxsize.hard: maximum response size
warpscript.urlfetch.host.patterns: allow/disallow some hosts based on regular expression like slack.com,.*!

WarpScript

'oQ4t3FJSguxxxxxxxxxxUCXo3vvRGg0Q5.'  // READ token
AUTHENTICATE

'https://random-data-api.com/api/name/random_name?size=5' URLFETCH 0 GET 3 GET // get body
B64-> 'utf8' BYTES-> JSON-> // decode as JSON
<% 'name' GET %> F LMAP