I work on some interesting projects as part of my job. And, sometimes you have to fill in some shoes that no one else wants to get into. But, as they say, variety is the spice that keeps things interesting in life.

During the second half of 2014 I was filling/fitting in the shoes of a HANA DBA. Mostly, it was run of the mill stuff, make sure everything is fine; trawl through the OOM (Out Of Memory) dumps and figure out who or what caused the dump and raise a case with SAP in case we failed to figure out why, apply some recommendations from SAP notes etc. In short, life was boring. One of the things we wanted to do was setup interim monitoring while Solution Manager was being setup. Anyways, back to the topic, when I say monitoring, it was nothing fancy; just get the backup statuses and some memory consumption related information had to be pulled in and put in the form of a dashboard. I was able to do it manually for two days before I decided to write some PowerShell to get some extra time on my hands.

In my case the problem statement is a familiar one, something we are used to in the DBA world. So, it was:

• Get backup data from a given SAP HANA instance.
  1. Create a connection to HANA instance of interest without using a DSN.
  2. Use this connection to run a query against the instance.
  3. Report the data.

And, I was reasonably sure that I was going to use it more than once and in more than one scenario.

Reading through various SAP blogs; I saw that most of the solutions published used the HANA studio (neat GUI), SAP GUI, Solution Manager, or required creating a DSN on the localhost. However there was one blog that describes how to make a connection to SAP Hana using Net Framework without a DSN. Since, I did not want a DSN this blog was perfect. The only prerequisite was that I install, ‘HANA Database client’ on the machine from where I was connecting.

In order to know, how to install the HANA Database Client, please refer Pages 10 through 13 in the PDF available at SAP HANA Database – Client installation and Update Guide.

Since we have broken down the solution into smaller steps, we have a pretty good idea what we need to tackle first i.e. “Create a connection to a given HANA instance, without using a DSN”. Because, once this is done we are good to go through a familiar process for getting the data.

<# .Synopsis    Get connection to a HANA database instance .DESCRIPTION    Create an OLEDb connection to a HANA database instance using a connection string.    HANA server port is defined by its instance number. When XX is the instance number,    the the port is 3XX15. .EXAMPLE    Get-HDBConnection -ServerName 'hana0001n' -InstanceNumber 61 #>

function Get-HDBConnection
{
    [CmdletBinding()]
    Param
    (
        # Name of the Linux host where HANA is running.
        [Parameter(Mandatory=$true,
                   ValueFromPipelineByPropertyName=$false,
                   Position=0)]
        [string]
        $ServerName,

        # Instance number of your HANA server.
        [Parameter(Mandatory=$true,
                   ValueFromPipelineByPropertyName=$false,
                   Position=1)]
        [int]
        $InstanceNumber
    )

    Begin
    {
        Write-Verbose -Message "Attempting to create connection string to $ServerName on port: " +
                                "3$($InstanceNumber.ToString())15"

        $hdbConnectionString = ""
        $hdbServerName = $("$ServerName" + ":" + "3$($InstanceNumber.ToString())15")

        if( [System.Intptr]::Size -ne 8){
		    #32-bit machine
		    $hdbConnectionString = [string]::Concat($hdbConnectionString, "Driver={HDBODBC32};")
        }else{
		    #64-bit machine.
		    $hdbConnectionString = [string]::Concat($hdbConnectionString, "Driver={HDBODBC32};")
        }
    }
    Process
    {
        $hdbCreds = $host.ui.PromptForCredential("HANA credentials", `
                                                  "Enter credentials for $($hdbServerName)", "", "")

        $hdbConnectionString = [string]::Concat($hdbConnectionString, "ServerNode=", $hdbServerName, ";")
	    $hdbConnectionString = [string]::Concat($hdbConnectionString, "UID=", $($hdbCreds.UserName), ";")
	    $hdbConnectionString = [string]::Concat($hdbConnectionString, "PWD=", `
                                                $($hdbCreds.GetNetworkCredential().Password), ";")

        Write-Verbose -Message "Connection string is: `n $hdbConnectionString"

        $hdbConn = New-Object System.Data.OleDb.OleDbConnection($hdbConnectionString)
        return $hdbConn
    }
    End
    {
    }
}

From here, all we need is to create a DB command that can be executed against the data source and a data adapter to fill the data table.

try
{
$hdbConnection = Get-HDBConnection $hdbServerName
$hdbCmdToRun = New-Object System.Data.OleDb.OleDbCommand($hdbSqlCmd,$hdbConnection)
$hdbCmdToRun.CommandTimeout = '300'
$hdbDataAdapter = New-Object system.Data.OleDb.OleDbDataAdapter($hdbCmdToRun)
$hdbDataTable = New-Object System.Data.DataTable

$hdbConnection.open()
[void]$hdbDataAdapter.fill($hdbDataTable)
$hdbConnection.close()

return $hdbDataTable
}
catch
{
$ex = $_.Exception
Write-Error $ex -Message "[Get-HDBData:Error-Querying-HANA]: " -ErrorAction Stop
}

finally
{
$hdbConnection.Dispose();
}

So, there you have it: TaDa! I know, who knew that connecting to a SAP HANA database was as simple as creating a connection string and using good’ol OLEDB to work the magic.

And here is a screenshot of this code in action.

Getting HANA backup details

Here’s an excerpt:

A New York City subway train holds 1,200 people. This blog was viewed about 5,700 times in 2015. If it were a NYC subway train, it would take about 5 trips to carry that many people.

Click here to see the complete report.

• Doug Finke’s – ImportExcel module.
• Warren F’s – PSExcel module

When I started writing the solution, I used Doug’s module because it is as easy as it can get, I was able to write and “test” the code in under 30 minutes. Basically, all I had to do was:

• Import the module
• Call ‘ConvertFrom-ExcelSheet’ with the proper parameters

Viola! you can now convert Excel Workbooks with multiple sheets in them.

As much as I hate reinventing the wheel, customers sometimes want you to do stuff that simply does not make any sense. So, I had to write a script that did not use the modules – sigh!.

However, while working on the script I encountered two small issues and thought, it might be useful for others to know about this as well.

• Figuring out the bottom most populated cell in the sheet.
  • Because the excel sheet that I had did not have any data in the first three rows, Dimension.Rows or Dimension.Columns was not working me. This I believe is because, If a range of cells is available Dimension returns the properties for the range. However, the End property does get the row and column of the bottom right cell. The original code with the cells to be changed highlighted and then follows the changed code.

$xl = New-Object OfficeOpenXml.ExcelPackage $FullName
$workbook = $xl.Workbook
$worksheet=$workbook.Worksheets[$Sheet]
$dimension=$worksheet.Dimension
$Rows=$dimension.Rows
$Columns=$dimension.Columns

Difference when there are empty cells in worksheet

$Rows=$worksheet.Dimension.End.Row
$Columns=$worksheet.Dimension.End.Column

• Figuring out the active worksheet
  • In order to export the data, the end user wanted to use the last active sheet and not the first sheet. Based on what I have read the below code, may not work if the excel workbooks were created programatically. It worked for me because the Excel workbooks were created manually so, there was always one and only one active sheet.

$activeTabIndex = [Int]$($xl.Workbook.WorkSheets | Where-Object {$_.View.TabSelected -eq $true} | Select Index).Index
$workSheet = $workBook.Worksheets[$activeTabIndex]

References:

• C#: Getting the number of rows/columns with ExcelPackage
• Determine last active sheet using EPPlus

The impetus for this post comes from the fact that I was looking for an way to get a list of all posts that I have written in the recent past which are tagged as PowerShell. Sure, I can browse by the tag and get the results; but I wanted to have a complete list at hand. That is where the WordPress API documentation came in handy.

If you have PowerShell V3, then you can the Invoke-RestMethod cmdlet to parse the results into objects and play around with them.

The structure of the URI for GETting posts is of the form:

https://public-api.wordpress.com/rest/v1/sites/$site/posts/

Where $site is the name of your site.

For my site the query is:

https://public-api.wordpress.com/rest/v1/sites/sqlchow.wordpress.com/posts/?number=50

Where 50 is the number of posts that I want to retrieve. How about the tag, you say? The query is only slightly different:

https://public-api.wordpress.com/rest/v1/sites/sqlchow.wordpress.com/posts/?number=50&tag=PowerShell

Since Invoke-RestMethod automatically converts the returned data into objects based on their type, we need not worry about the kind of response we get from the web page.

All we need to do is query the service and get our data.

$sqlchowBlogData = Invoke-RestMethod -uri "https://public-api.wordpress.com/rest/v1/sites/sqlchow.wordpress.com/posts/?number=50&tag=PowerShell"

$sqlchowBlogData | Get-Member
TypeName: System.Management.Automation.PSCustomObject

Name        MemberType   Definition
----        ----------   ----------
Equals      Method       bool Equals(System.Object obj)
GetHashCode Method       int GetHashCode()
GetType     Method       type GetType()
ToString    Method       string ToString()
found       NoteProperty System.Int32 found=28
posts       NoteProperty System.Object[] posts=System.Object[]

Let us check, how many posts we found:

$sqlchowBlogData.found
28

Now, let us see we can get all the post titles and the dates they were published. The data that we need is in the posts object. Let’s get the data.

$sqlchowBlogData.Posts | Select Title, @{Name="Published Date"; Expression={Get-Date ($_.Date) -Format "MMM-dd-yyyy"}} | Format-Table -AutoSize

title                                                                              Published Date
-----                                                                              --------------
Windows Azure setup for using with PowerShell                                      Mar-30-2014
Using verbal expressions to make regex easy in PowerShell                          Sep-07-2013
Down the rabbit hole with PowerShell and Windows Azure SQL Database                Sep-02-2013
Google search results using REST (custom search api) in PowerShell                 Aug-19-2013
Filtering data from SQL*Server Errorlogs                                           Aug-18-2013
.......
.......
Reading a tweeters timeline using PowerShell-Take2                                 Dec-27-2011
Reading a tweeters timeline using PowerShell                                       Dec-26-2011

Now that we have the details, we can use ConverTo-HTML to maintain an offline copy of certain posts that we need.

RESTful services are all around us, PowerShell gives us an easy way to interact with them. The possibilities are endless .

Cheers!!

First things first, we need a Windows Azure subscription. Presently, you can register for a free trail at www.windowsazure.com.

Note: If you would like to use the management portal for Windows Azure to create the database, you can use the following tutorial.

Pre-Requisites

We need a few tools and a little bit of setup on the management portal before we can start with the automation part.

1. We need to rename our Windows Azure subscription from the default name it has. This makes it easier to manage your subscriptions, if you have multiple services.
2. We need to upload a certificate to the windows azure subscription.
3. Download latest version of Windows Azure PowerShell, from here. The link points to the standalone installer.
4. Download the latest Windows Azure SDK for .NET from here.

1. Renaming the subscription

The Windows Azure services have two kinds of accounts.

1. Windows Azure Account, through which resource usage is reported and services are billed. Each account is identified by an email account, and is associated with at least one subscription. The account owner monitors usage and manages billings through the Windows Azure Account Center.
2. Subscriptions, which governs access to and use of Windows Azure subscribed service. The subscription holder uses the Management Portal to manage services.

The reason for this kind of setup is because in a corporate enrollment, an account owner might create multiple subscriptions to give IT folk’s access to services. Since resource usage within an account is reported for each subscription, an organization can use subscriptions to track expenses for projects, departments, regional offices, and so forth.

This is explained in detail here.

You change your subscription details by using the following steps:

1. Sign into the Windows Azure Management Portal.
2. Click on ‘Account’ → ‘View my bill’; you maybe prompted to login as you will be accessing sensitive details.
   

   How to: Change Subscription details

    

3. Click on ‘Edit Subscription Details’, to change the name of your subscription.

HowTo: Change Subscription Name – 1

HowTo: Change SubscriptionName-2

2. Creating and uploading a management certificate

Certificates used in Windows Azure are x.509 v3 certificates and can be signed by another trusted certificate or they can be self-signed. For the purposes of this blog we will be creating a self-signed certificate using ‘makecert.exe’. Before we go further, let us check how different certificates are used in Windows Azure.

Windows Azure has 3 types of certificates:

1. Management certificates – Stored at the subscription level, these certificates are used to enable Windows Azure using various tools. These certificates are independent of any hosted service or deployment.
2. Service certificates – Stored at the hosted service level, these certificates are used by your deployed services.
3. SSH Keys – Stored on the Linux virtual machine, SSH keys are used to authenticate remote connections to the virtual machine.

Windows Azure uses certificates to identify a trust relationship i.e. the party to be trusted has the private key. Here is how the certificates are used:

• Management certificates (.cer certificate files): the client connecting the service needs to be trusted and has the private key. These permit clients access to resources in your Windows Azure subscription.
• Service certificates (.pfx certificate files): the service needs to be trusted by the client connecting to the service. For example, in an SSL secured service scenario they provide secure interactions for users of your web application or service.

Enough talk, let us start by creating a self-signed certificate. To create your own self-signed management certificates, open a Visual Studio command prompt as an administrator, and then run the following command:

makecert -sky exchange -r -n "CN=" -pe -a sha1 -len 2048 -ss My ".cer"

If you are like me, you would get a doubt about what the parameters mean. Especially what ‘pe’ and ‘sky’ mean?

To put simply, unless we specify the -pe command line option to MakeCert, the certificate created can only be used to decrypt data on that machine. This is not an ideal scenario because, this would prevent any other machine from being able to decrypt the payload encrypted using the public key.

The option ‘sky’ specifies the type of the subject key. It accepts two arguments (signature|exchange). What these arguments mean is:

• ‘signature’ the certificate can only be used to sign a payload.
• ‘exchange’ the certificate can be used to sign AND/OR encrypt a payload.

Now that we have created our certificate, we need to export it in order to be able to upload the file to Azure. We can do this in a few easy steps:

• Start / run / certmgr.msc
• Navigate the tree hierarchy until you find the certificate you just created. (usually under Personal→Certificates).
• On the left pane right click on the certificate and select / all tasks / export.
• Select “No, do not export the private key”
• Click on next until you are asked for the Export path. Enter the desired export path.
• Follow the prompts and finish the export.

Uploading the certificate:

Management and service certificates can be uploaded through the Windows Azure Management Portal.

HowTo: Upload Certificate

3. Download latest version of Windows Azure PowerShell, from here. The link points to the standalone installer.

    Please download and install from the link given above.

4. Download the latest Windows Azure SDK for .NET from here.

    Please download and install from the link given above.

This is it. We are now setup to work with our azure subscription from within PowerShell.

References:

1. SQL Database – Documentation.
2. Manage Certificates in Windows Azure.
3. Create and upload a certificate for Windows Azure.
4. makecert option -pe.
5. makecert option -sky.

TSQL-Tuesday#050: The automation story

What is T-SQL Tuesday?
T-SQL Tuesday is a monthly blog party hosted by a different blogger each month. This blog party was started by Adam Machanic (blog|twitter). You can take part by posting your own participating post that fits the topic of the month and follows the requirements (click adjacent image to know more). Additionally, if you are interested in hosting a future T-SQL Tuesday, contact Adam Machanic on his blog.

The story

A few years ago, when I started of in a large project, the sheer number of instances we were supporting boggled my mind. What was even more interesting was that the installation of SQL*Server was semi-automated (it involved a configuration file and a Perl script andddd..we had some interesting calls with the good folks at MS support because of this ). This whole setup really got me thinking about how and what we could automate. Until then, I was just writing some scripts and saving them in a folder to use later on but, nothing at this scale. While I was still in awe at the setup and architecture of the solution I started to realize few things:

1. You may not be able to automate somethings.
2. You need to have a broad understanding of the shop you are running.
3. A deeper understanding of our own technical skills.
4. There should be an Opportunity. But, this is again tied to the above two points. If look closely enough you will find Opportunities to automate.

So, I started looking at Perl and got so engrossed in it that I started looking at automating summarizing perfmon logs using couple of modules from CPAN. After spending a week of sleepless days and nights on it, I had a working version which found only two interested people in our team who would even use it. That was my second set of lessons:

1. Draw the line when you think the effort you put in outweigh the benefits (for you or the team).
2. Packaging and Presentation of your solution makes a world of difference.

Since, I did not see any point in developing the module further so left it as is. But, I was not done with the summarizing CSV files business, I started looking at what other tools were available to do this kind of work and that is when I found PowerShell. The very first version was a little uncomfortable to work with but, I found what I was looking for, “*Import-Csv” .

I went through the first 4 steps, or did a feasibility analysis and then I went through the last 2 steps of implementation and introduction. Life was right again.

I was chatting with a colleague about why it worked the second time around and that too with a scripting language hardly anyone knew a that time. His response was that the second solution was more innovative. That was another important lesson I learnt about automation, Innovation.

Innovation

Innovation, is all about doing “stuff” in a different way than the usual. It could be as simple as taking a new route to work or as complex as rethinking the way you commute to work. Well, these maybe bad examples but, the gist is that all innovations need not be discoveries.

I mostly work on SAP on SQL systems. Recently, we had a requirement at one of our clients to track the database growth on a daily basis. The initial thought was to use one of the canned scripts that I had written a while back. Something as elementary as the below would’ve worked just fine.

DECLARE @dbDataSize INT,
	@dbDataUsed INT,
	@dbLogSize  INT,
	@dbLogUsed  INT
SELECT @dbDataSize = SUM(size) FROM sys.database_files WHERE [type] = 0
SELECT @dbDataUsed = SUM(FILEPROPERTY(name,'SpaceUsed')) FROM sys.database_files WHERE [type]= 0
SELECT @dbLogSize = SUM(size) FROM sys.database_files WHERE [type] = 1
SELECT @dbLogUsed = SUM(FILEPROPERTY(name,'SpaceUsed')) FROM sys.database_files WHERE [type]= 1

SELECT
 DBName = DB_NAME(),
 DataSize_MB = @dbDataSize/128,
 DataFree_MB = (@dbDataSize - @dbDataUsed)/128,
 DataFreePct = CAST((CAST((@dbDataSize - @dbDataUsed) AS DECIMAL(15,2))/CAST(@dbDataSize AS DECIMAL(15,2)))*100 AS DECIMAL(4,2)),
 LogSize_MB = @dbLogSize/128,
 LogFree_MB = (@dbLogSize - @dbLogUsed)/128,
 LogFreePct = CAST((CAST((@dbLogSize - @dbLogUsed) AS DECIMAL(15,2))/CAST(@dbLogSize AS DECIMAL(15,2)))*100 AS DECIMAL(4,2))

But, if you had worked with SAP systems long enough you know the BASIS admins get a lot of cool reports in DB02, one of which incidentally is a DB growth report. So, I started looking through the stored procedures in the database and found this beauty called ‘sap_dbszhist‘. Now this procedure takes four parameters ‘db_name’, ‘object_scheme’, ‘from_dt’ and ‘to_dt’. But, we do not need to pass any of these parameters. The default behavior is:

• If db_name is null then current DB context is assumed.
• If obj_schema is null schema details are gotten from sys.objects using @@PROCID
• If the from_dt parameter is null it assigns a date in the past 3yrs from today, and to_dt is today’s date.

So, without a lot of effort I now have a historical record of DB growth by day. All we need now is a process to put this in a central repository for some SSRS fun.

Moral…

What I love about this is that the approach combined all the lessons learned over a period time and gave me a chance to think outside the normal.That is the final lesson, I learnt:

Automation, it’s all in the mind.

TSQL-Tuesday#050: The automation story

Automation plays a huge part in our lives and the DBA profession is no so different. A couple of years ago this topic was shared by Pat Wright. I would like to revisit this again and see how much of our approach has changed.

You could write about, what options you would consider when automating something? Where do you draw the line? What are our preferred tools for automation? T-SQL, PowerShell, VBScript or Batch files(?) or maybe just share something that you automated in the last couple of years.

What is T-SQL Tuesday?
T-SQL Tuesday is a monthly blog party hosted by a different blogger each month. This blog party was started by Adam Machanic (blog|twitter). You can take part by posting your own participating post that fits the topic of the month and follows the requirements below. Additionally, if you are interested in hosting a future T-SQL Tuesday, contact Adam Machanic on his blog.

How to Participate

• Your post must be published between 00:00 GMT Tuesday, January 14th, 2014, and 00:00 GMT Wednesday January 15th, 2013.
• Your post must contain the T-SQL Tuesday logo from above and the image should link back to this blog post.
• Trackbacks should work. But, please do add a link to your post in the comments section below so everyone can see your work.
• Tweet about your post using the hash tag #TSQL2sDay.

Here’s an excerpt:

A New York City subway train holds 1,200 people. This blog was viewed about 7,100 times in 2013. If it were a NYC subway train, it would take about 6 trips to carry that many people.

Click here to see the complete report.

T-SQL Tuesday #046: hosted by Rick Krueger

This month the T-SQL Tuesday invitation is from Rick Krueger (B|T), who want us to dust out the closet and tell how how we escaped from the jungles of Borneo, swam across the vast Indian ocean and saved the tiger with nothing but a few clips and rubber band a.k.a Rube Goldberg Machine.

While working for a Big blue box electronic retailer, we had to support a few SQL 2000 instances. And we had to figure out a way to accurately determine if certain jobs had finished or were still executing. There were other jobs that needed to be put on hold until the job of interest was executing.

The constraints placed were:

1. Conclusively determine if the job has started execution.
2. Without the use of undocumented features we should be able to determine the current execution state of the job.
3. Once a job’s current state is determined, we should wait until the state changes.
4. Depending on the execution state, we should decide either to wait for a further delta minutes and check the state or determine if the job finished successfully.
5. Return the final result to the calling code block.

This is how I went about it:

1. Wait for a delay of 1 minute before even checking.
2. Checking if we actually had a running job, was done in a roundabout way. The last step_name of the previous execution of the job always contains ‘(Job outcome)’ in sysjobhistory table. Because of this, we can pick an instance_id of this job that has an instance_id greater than the MAX(instance_id) when the step_name was ‘(Job outcome)’.
3. I had to BCP the results of sp_help_job to a file because I kept getting ‘Error Message:Msg 8164, Level 16, State 1, Procedure InsertIntoSp, Line 10’ – “An INSERT EXEC statement cannot be nested.”
4. Use BULK INSERT to load data into a temp table.
5. Check if the execution status of the job was 4 i.e. waiting for next scheduled execution.
   1. If the status was not 4, then wait for a minute and then repeat steps 1-5.
   2. If the status was 4 then move to next stage.
6. If we find that the job completed successfully then validate with sysjobhistory and check if the job really succeeded. Didn’t want false positivies.
7. If we did not find a job executing on Step 2., return a non-zero value.

This whole thing, was put in a store procedure and then called from a job step for a long running job. For example, like a database backup job. And additional logic was written around the return value to write proper error numbers to application log. These event-ids where then picked up by monitoring applications and alerts where raised.

I have mixed feelings about this solution. Although, it worked without a problem (I am secretly proud that I was able to write something like this); I would have loved to just upgrade the servers to SQL 2008 and be done with it.

‘Regular Expressions (regex) are my favorite programming tool’

said no one ever.

When I first came across them, I was a bit intimidated as many of us would be. They looked like complete gibberish. But, due to the nature of the code I was writing, I had to learn (or at least understand) them. Parsing SQL*Server error log files or netbackup logs, is not easy in Perl without regular expressions.

After using them a few times you realize that they are basically a tool to find specific patterns within a given text. But, you hit barriers when you try to use them to capture complex patterns. For example, consider this regex pattern:

'^(?:http)(?:s)?(?:://)(?:www\.)?(?:[^\ ]*)$'

Well what does it do? Not very easy to figure out right? But, it is the regex pattern for a valid URL; true story.

This is where verbal expressions steps in. Verbal Expressions is an open-source project that aims to simplify the task of writing difficult, complex regular expressions by providing a means of specifying regular expression patterns using easily readable text.

What it means is that, we write ‘Four times X plus 3 equals seven’, and get ‘4x+3=7’. That example, could be a little misleading so let try it again, a valid url would consist of http at the begining of the line, maybe s if you are accessing a secure site, then it will have ://, maybe www. if you are so inclined to type and anything but a space. The aim of verbal expressions is to let you type the above and get a regular expression that matches a URL which is this: this:‘^(?:http)(?:s)?(?:://)(?:www.)?(?:[^\ ])$’.

Now, we need not be afraid of regular expressions.

How to start

It’s easy, just import the module.

Import-Module -Name Poverb

Note:This is currently a partial implementation only.

How To use it

Currently the output is only a string that has the regular expression you desire.

	Get-RegexFromVerbex -Verbex {
		startofline
		then "http"
		maybe "s"
		then "://"
		maybe "www."
		anythingbut " "
		endofline
	}

the output is:

	'^(?:http)(?:s)?(?:://)(?:www\.)?(?:[^\ ]*)$'

How do I get it

You can get it from github, PowerShellVerbalExpressions

Other implementations

You can view all implementations on VerbalExpressions.github.io

References

1. Jim Hollenhorst – The 30 Minute Regex Tutorial
2. VerbalExpressions.github.io
3. Peder Søholt
4. Karl Prosser for writing Invoke-Ternary filter
5. Doug Finke
   • For tweeting about Verbal expressions
   • For writing about DSLs in his book