RaygunML

Raygun.io provider for MarkLogic.

Installation

• Copy RaygunML folder into your modules database, file system, or into MarkLogic/Modules/MarkLogic.
• If you placed RaygunML folder in the filesystem or modules database of one of your app servers, you can execute HTTP GET request on raygun-init.xqy (the script is meant to be executed by admin MarkLogic user) with the following parameters:
  • isconfig: boolean - If set to false, arguments api-key, mode, database, and user have to be provided; otherwise argument config has to be provided.
  • api-key: Your Raygun api key can be found in Raygun.io web application under Plan Settings --> Application Settings.
• mode: Valid modes are development/qa/production - in development mode reporting to raygun is turned off. - database: One or more database names. RaygunML init script will insert into the listed databases an xml config file in the following format:

  <raygun>
    <api-key>raygun-api-key</api-key>
    <mode>raygun-mode</mode>
  </raygun>
  

- **user**: One or more MarkLogic users. RaygunML creates two
  MarkLogic roles: *raygun-read* and *raygun-update*. By default, 
  listed database users get only *raygun-read* role assigned to them.
  Therefore, listed users will not be able to change raygun api key or
  raygun mode. If you want the specified database users to be able to
  change raygun api key or raygun mode in the configuration file
  inserted into the listed databases, you have to assign them 
  *raygun-update* role in MarkLogic admin interface or
  programmatically.
- **config**: raygun initialization config xml file in the following
  format:

 <raygun>
	    <api-key>raygun-api-key</api-key>
        <mode>raygun-mode</mode>
        <usernames>
		  <user>user1</user>
		  <user>user2</user>
		        ....
	    </usernames>
	    <databases>
		  <database>database1</database>
		  <database>database2</database>
	                  ....
	    </databases>
</raygun>

Installation Examples
Say I have an app server at port 8080 with modules in the file system and I placed RaygunML folder in the root ("/") directory of that server. Here's an example of RaygunML installation without config.xml:

http://localhost:8080/RaygunML/raygun-init.xqy?&api-key=apikey&mode=development&database=database1&database=database2&user=user1&user=user2&isconfig=false

Installation with config.xml.

http://localhost:8080/RaygunML/raygun-init.xqy?config=<raygun><api-key>apikey</api-key><mode>development</mode><usernames><user>user1</user><user>user2</user></usernames><databases><database>database1</database><database>database2</database></databases></raygun>&isconfig=true

Powershell script to initialize RaygunML. Make sure to place this Powershell script in the same directory as config.xml (see config).

$file = Get-Content config.xml
$file = $file -replace '\s',''
$req = "http://localhost:8080/RaygunML/raygun-init.xqy?config=" + $file +
        "&isconfig=true"
$cred = Get-Credential

Write-Output $req
Invoke-WebRequest -Uri $req -Credential $cred

You can always initialize RaygunML in QConsole. Say you placed RaygunML folder in .../MarkLogic/Modules/MarkLogic. You can execute the following code in QConsole to initialize RaygunML:

xquery version "1.0-ml";

import module namespace raygun-deploy = "http://raygun.io" at
  "/MarkLogic/RaygunML/raygun-deploy.xqm";

raygun-deploy:init(
  <raygun>
    <api-key>raygun-api-key</api-key>
    <mode>raygun-mode</mode>
    <usernames>
      <user>user1</user>
      <user>user2</user>
    </usernames>
    <databases>
      <database>database1</database>
      <database>database2</database>
    </databases>
  </raygun>)

Usage

In order to report an exception to Raygun, you can use the following code:

import module namespace raygun = "http://raygun.io" at
  "/RaygunML/raygun.xqm";

try
{ 
  error((), "Error", ("errorval1", "errorval2"))
catch($error) 
{ 
  raygun:report-error($error)
}

Version

Alpha.