Getting an SPWeb

The conventional way of obtaining a reference to an SPWeb is:

 get-spweb http://mybiglongservername:8080/path/to/web
 

Because this requires a fully-qualified URL, it gets when dealing with a lot of sites and webs.  Sometimes I’ll just throw the web in a variable, but that’s not exactly good practice because it can lead to memory leaks.  Instead, I wrote Get-SPWebRelative, which can be used like this:

 get-spwebrelative /path/to/web
 

You can even get the root web:

 get-spwebrelative /
 

Similarly, there is a Get-SPSiteRelative function that works the same way.

I was about to go on vacation, but I had just published my PowerShell extensions and thought it would be a good opportunity to use them.  So, in one of my prouder moments, I free-handed about 10 lines of PowerShell into an email and left town.  I was delighted to find out that the code actually worked (once you removed the incorrect characters that Outlook had injected).

So, here’s a quick little blog post explaining what I did.

First, I spooled up a standalone SharePoint 2010 VM and ran PowerShell ISE.

Then I did the following to initialize my extensions:

cd \VboxSvr\host_m\Development\wfPowerShell
. .\Boostrapper.ps1

Next I needed a page to play with, so I created a test web part page and added a couple of web parts:

Now, I could have manually created the dupes, but it was more fun to try it with a script:

</p><h1>This function has virtually no practical application :)</h1><p>filter Duplicate-WebParts(){ $Page = $_
$webparts = $Page | Export-WebParts
$Page | Import-WebParts -WebPartInfos $webparts
}

$page = Get-SPWebRelative / | Get-SPFile 'Shared Documents/TestDupes.aspx'
1..3 | % { $page | Duplicate-WebParts | out-null }

11:36:24 PM Imported web part: Web Part Page Title Bar    [Microsoft.SharePoint.WebPartPages.TitleBarWebPart]
11:36:27 PM Imported web part:     [Microsoft.SharePoint.WebPartPages.XsltListViewWebPart]
11:36:28 PM Imported web part: Content Editor    [Microsoft.SharePoint.WebPartPages.ContentEditorWebPart]
11:36:33 PM Imported web part: Web Part Page Title Bar    [Microsoft.SharePoint.WebPartPages.TitleBarWebPart]
11:36:33 PM Imported web part:     [Microsoft.SharePoint.WebPartPages.XsltListViewWebPart]
11:36:34 PM Imported web part: Content Editor    [Microsoft.SharePoint.WebPartPages.ContentEditorWebPart]
11:36:35 PM Imported web part: Web Part Page Title Bar    [Microsoft.SharePoint.WebPartPages.TitleBarWebPart]
11:36:36 PM Imported web part:     [Microsoft.SharePoint.WebPartPages.XsltListViewWebPart]
11:36:37 PM Imported web part: Content Editor    [Microsoft.SharePoint.WebPartPages.ContentEditorWebPart]

filter Remove-DuplicateWebParts([switch] $WhatIf, $DupeProperties = @('Title', { $_.GetType() })) {
    $Page = $_
    Log "processing WebPart duplicates in $($page.ServerRelativeUrl)"
    $webparts = $page | Get-WebParts
    $dupeGroups = $webparts | group $DupeProperties | where { $_.Count –gt 1 }
    $allDeletions = $dupeGroups | lwp -ActionLabel "Processing Dupe Group" -StatusReportBlock { $_.Name }  | % {
        $keep, $deletions = $_.Group
        return $deletions
    }
    if(!$WhatIf){
        $page | modify-webparts –delete –filter {
            ($allDeletions |> Id ) -contains $_.Id
        } | lwp -Action Deleted -Status { "$($_.Title) ($($_.Id))" } | log-item -prefix 'deleted '  -Process {"$($_.Title) ($($_.Id))"}  | out-null
    } else {
        $allDeletions | log-item -prefix "Delete " -Process {"$($_.Title) ($($_.Id))"} | out-null
    }
    return $Page
}

$page = Get-SPWebRelative / | Get-SPFile 'Shared Documents/TestDupes.aspx'
#1..2 | % { $page | Duplicate-WebParts | out-null }
$page | Remove-DuplicateWebParts | out-null

11:36:39 PM processing WebPart duplicates in /Shared Documents/TestDupes.aspx
11:36:42 PM deleted Web Part Page Title Bar (g<em>7571f1e5</em>1238<em>48d7</em>a029<em>66acf59d7a9c)
11:36:42 PM deleted Shared Documents (g</em>edb5d78a<em>c5ad</em>4f75<em>b134</em>01716c46f7eb)
11:36:43 PM deleted Content Editor (g<em>6f02ef6a</em>6c52<em>4c1c</em>864b<em>553dcddb27f8)
11:36:43 PM deleted Web Part Page Title Bar (g</em>19c9e065<em>4ac6</em>4329<em>9d3c</em>445b63cfac4e)
11:36:43 PM deleted Shared Documents (g<em>d1c7659a</em>6b26<em>43c2</em>88d7<em>b767349f689b)
11:36:43 PM deleted Content Editor (g</em>dc6cbcaf<em>0705</em>462e<em>ba39</em>a9fae18682fe)
11:36:43 PM deleted Web Part Page Title Bar (g<em>e1a26dc2</em>cc6f<em>49b6</em>bccd<em>ef825ab56dea)
11:36:43 PM deleted Shared Documents (g</em>20d52916<em>1bad</em>4c4c<em>9374</em>2fb27296b6cb)
11:36:43 PM deleted Content Editor (g<em>e1847600</em>d1f1<em>43c7</em>a83d_08f2273ac7ba)

Many of my colleagues seem to be unaware of any alternatives to the standard powershell.exe command prompt, which is a bit, um, “old school”.

If you weren’t aware of PowerShell ISE, the integrated development environment for PowerShell that you get for free, I’d recommend installing it and using it rather than the traditional PowerShell prompt.  The ISE is superior for several reasons:

• It has a tabbed interface for editing scripts
• It can host multiple PowerShell sessions at once
• It has superior auto-completion.  In fact, the auto-completion in the interpreter’s command-line is better than that of the normal PowerShell prompt, because you can re-position your cursor and get auto-completion without accidentally deleting everything after the cursor (as the command prompt-based one does)
• It supports breakpoints and debugging

For these reasons, I almost never use plain old powershell.exe.

I believe that it comes included with PowerShell for Windows 7, and to install it on Windows 2008 Server, all you have to do is add the PowerShell ISE feature using the Server Manager (or PowerShell).

I used it mostly for personal tasks and a handful of development tasks.  When I worked on a big SharePoint 2010 migration at the beginning of the year (which I will also be blogging about), I ended up upgrading the whole library to PowerShell 2.0 and also writing a ton of stuff to manage SharePoint 2010.  Since then, I’ve used it on an almost daily basis with every SharePoint 2010 project I encounter, and I will continue to enhance and refine it going forward.

Whereas most of the existing PowerShell API for SharePoint deals with administrative concerns (a la stsadm), I needed to manipulate a lot of SharePoint content.  Lists, List Items, Documents, Pages, WebParts, Groups, Permissions, PropertyBags, etc.  You get the gist.  

I was surprised to find that there weren’t (m)any cmdlets that addressed those concerns.  I’m still surprised, and maybe I missed something, but a deadline’s a deadline, so I wrote a bunch of my own cmdlets, which I am now sharing.  If you’ve ever tried to do any of these things in PowerShell, I think you’ll find these functions very helpful.  

I’ve just refactored everything into 2 major libraries (General PowerShell and SharePoint 2010) that are loaded using a dependency loader that is also part of the library.  I tried to modularize everything so that if you wanted to just use a couple of parts of it, you should be able to do so with minimal concern over dependencies on other script files.  I also just went through and added comment-based help and examples to many of the cmdlets.  In the future, I may refactor it to take advantage of PowerShell 2.0 modules, but for the time being, I like the drop-and-go simplicity of my current approach.

The new codeplex project is here:  wfpowershell.codeplex.com.

There’s no official release yet, but you can just download the latest revision from:

http://wfpowershell.codeplex.com/SourceControl/list/changesets.

The only thing you need to know is how to initialize the library once you’re in a PowerShell session.  All you need to do is dot-execute the Bootstrapper script.  Typically you would do something like this

 cd [library directory]
 . .\Bootstrapper.ps1
 

That’s about it.  You’ll have to browse through the scripts yourself to see what’s inside while I work on blogging and documenting how to use it.

Let me know what you think!

First of all, let me be clear that this is not a good practice.  Live filtering of tables doesn’t work very well, especially when you are using SP ListViews that are generating their own markup.  To make matters worse, the code was traversing the inner HTML of every TD on the page to examine its contents in order to determine whether to hide the page.  It was also 80 lines of JS.

I replaced it by adding a script link to JQuery and adding this function:

function HideColumn(i, el){

$("th:nth-child(" + i + "), td:nth-child(" + i + ")", el).hide();

}

Great talk on user interaction anti-patterns by Bill Scott of Yahoo! and now Netflix.

function Html-ToText {
 param([System.String] $html)

 # remove line breaks, replace with spaces
 $html = $html -replace "(`r|`n|`t)", " "
 # write-verbose "removed line breaks: `n`n$html`n"

 # remove invisible content
 @('head', 'style', 'script', 'object', 'embed', 'applet', 'noframes', 'noscript', 'noembed') | % {
  $html = $html -replace "<$_[^>]*?>.*?</$_>", ""
 }
 # write-verbose "removed invisible blocks: `n`n$html`n"

 # Condense extra whitespace
 $html = $html -replace "( )+", " "
 # write-verbose "condensed whitespace: `n`n$html`n"

 # Add line breaks
 @('div','p','blockquote','h[1-9]') | % { $html = $html -replace "</?$_[^>]*?>.*?</$_>", ("`n" + '$0' )} 
 # Add line breaks for self-closing tags
 @('div','p','blockquote','h[1-9]','br') | % { $html = $html -replace "<$_[^>]*?/>", ('$0' + "`n")} 
 # write-verbose "added line breaks: `n`n$html`n"

 #strip tags 
 $html = $html -replace "<[^>]*?>", ""
 # write-verbose "removed tags: `n`n$html`n"
  
 # replace common entities
 @( 
  @("&amp;bull;", " * "),
  @("&amp;lsaquo;", "<"),
  @("&amp;rsaquo;", ">"),
  @("&amp;(rsquo|lsquo);", "'"),
  @("&amp;(quot|ldquo|rdquo);", '"'),
  @("&amp;trade;", "(tm)"),
  @("&amp;frasl;", "/"),
  @("&amp;(quot|#34|#034|#x22);", '"'),
  @('&amp;(amp|#38|#038|#x26);', "&amp;"),
  @("&amp;(lt|#60|#060|#x3c);", "<"),
  @("&amp;(gt|#62|#062|#x3e);", ">"),
  @('&amp;(copy|#169);', "(c)"),
  @("&amp;(reg|#174);", "(r)"),
  @("&amp;nbsp;", " "),
  @("&amp;(.{2,6});", "")
 ) | % { $html = $html -replace $_[0], $_[1] }
 # write-verbose "replaced entities: `n`n$html`n"

 return $html

}

And you can run it like this:

Html-ToText (new-object net.webclient).DownloadString("<a href="http://winstonfassett.com">http://winstonfassett.com</a>")

But unfortunately, the Jad website has vanished from the internet.  Can you believe that?  In this day and age, when all you need is a place to host a couple of pages and a couple of small zip files, it’s just gone, and in its place is a dumb generic domain landing page.  

It’s crazy that there is no where else to get Jad.  Crazy because there are tons of tools that sit on top of Jad.  JadClipse, cavaj, JD Decompiler, etc.  But the heart of the thing is Jad, and every one of those tools requires you to download Jad separately, but oops:  Jad has vanished from the internet.  

I managed to find an old version of the Jad exe on an old machine, but then today I tried looking around and found that Tomas Varaneckas set up a mirror.  It was hard to find however, so I’m posting the link here for posterity.  Thanks Tomas!

Fortunately, someone thought to set up a mirror.

Then today I got fed up, tracked down the url for manually downloading it (this forum thread is a good source).  But it still didn’t work!  It kept hanging after a couple of minutes.

Eventually I realized that it might have something to do with my router settings.  I had set connections to time out after 120 seconds in order to better handle lots of bit torrent traffic.  It’s usually not a problem, as my browsers are usually able to resume downloads when connections are broken, even when I lose internet connectivity completely.  But perhaps there’s something about edge / content delivery network downloads that requires a persistent connection.  Anyway, I raised my connection timeout to 600 seconds and now all my Apple downloads are working fine.

Update: scratch that.  Didn’t work totally fine.  The iPhone update still keeps hanging and eventually dying.  Now I’m trying to nurse it along by occasionally “pausing” and “resuming” the download in Google Chrome.  Looks like it might work, if I’m attentive.  What a beating.

Update 2: So nursing it along proved to be too tedious.  After 45min of starting and stopping, it failed at 240/246M!  Arrgggh!  So now I’m thinking maybe it has something to do with the wifi.  So this time I SSH’d into my Linux server (which is wired to the net), used wget to download, and it took 5 minutes.

Update 3: Scratch that.  It turns out I downloaded the update for the 3G iPhone instead of my 1st generation iPhone.  The first generation firmware update keeps failing over and over and over.  It’s killing me.  Somehow I got the 3G version, 246M, in an amazing 5 minutes, but the download for my phone is coming in at 50k/s and stalling out at around 2% complete.  This is ridiculous.