Karine Bosch’s Blog

On SharePoint

What if your sandbox solution contains a Visual WebPart?

As already mentioned in previous post, Microsoft decided to stop support for coded sandbox solutions on SharePoint Online. Developers will have to do the necessary effort to migrate the impacted sandbox solutions into solutions that comply with the recommended coding practices. You can find a lot of code samples and recommendations on the PnP site.

If your sandbox solutions contains a Visual WebPart, you will have to take action, because a Visual WebPart generates an assembly.In general, a Visual Web Part consists of the following:

  • an .ascx file with UI elements like HTML controls and ASP.NET controls
  • an .ascx.cs code behind file with some business logic that interacts with the UI and other data in the SharePoint site
  • a .webpart file that contains the definition of the web part (i.e. title of the web part, description, full name of the assembly, etc)

You could try to replace all ASP.NET controls by HTML controls, and develop the business logic in JavaScript and JSOM or REST. Now you don’t need that DLL anymore, right? Now you can get rid of the the assembly by setting the Include Assembly in Package project property to false in Visual Studio, right?


Wrong! You can even try to remove the <%@Assembly> directives and the <%@Register> directives from the .ascx control:

visual webpart

Even if your Visual WebPart contains nothing else than HTML and JavaScript, that assembly is generated. This is because a Visual WebPart consists of an .ascx control. In SharePoint on premises environment, an .ascx control gets deployed to the /TEMPLATE/CONTROLTEMPLATES folder; but this is not the case with sandbox solutions. The .ascx control gets compiled into the DLL, meaning that all UI elements defined on the .ascx control are generated in code.

Trying to force things by removing the metadata element from the .webpart file, will result in an import error when the page loads:

webpart definition

In case of a complex Visual WebPart, your best option is to create a SharePoint Add-in Part:

  • A SharePoint hosted add-in consisting of HTML, JavaScript and using JSOM or REST to communicate with the host SharePoint site
  • A provider hosted add-in represented by an MVC application using CSOM to communicate with SharePoint

In case of less complex Visual WebPart, you could choose to replace it by embedding JavaScript

But all these approaches require you to go to each page that hosts the old Visual Web Part to replace it by the new development. This can be quite a challenge for large O365 tenants.

Another approach is to keep your .webpart file and change it to point to a JavaScript file. The advantage of this approach is that pages hosting your old Visual WebPart will now automatically host your new web part.

What are the steps you have to take to make this approach work?

  1. Write the JavaScript to replace the logic of your existing Visual Web Part
  2. Point the .webpart file to this JavaScript file
  3. Deactivate and remove the old sandbox solution
  4. Upload your JavaScript file into the Style Library of your site collection
  5. Upload the new .webpart file into the WebPart Gallery

Following image shows how you can modify the .webpart file:

  • Change the <metadata> tag to point to the ScriptEditorWebPart
      <type name="Microsoft.SharePoint.WebPartPages.ScriptEditorWebPart, Microsoft.SharePoint, Version=, Culture=neutral, PublicKeyToken=71e9bce111e9429c" />
      <importErrorMessage>Cannot import the migrated Demo Visual Web Part.</importErrorMessage>
  • Add a <property> element with name Content
  • Add a reference to your JavaScript file
  • Add a <div> element that will be the container of the UI that you have to develop in the JavaScript file

webpart definition pointing to js

In your JavaScript you will have to retrieve this <div> element, to be able to embed the rest of the UI.

You can even automate steps 4 and 5. You can upload the JavaScript into the Style Library using the following CSOM code:

        private static void UploadFileToStyleLibrary(ClientContext ctx, string sourcePath, string styleLibFolder, string jsFileName)
            // get the Style Library
            List styleLibrary = ctx.Site.RootWeb.GetList(ctx.Url + "/Style%20Library");

            // upload the js file
            string fullSourcePath = sourcePath + jsFileName;
            if (!styleLibFolder.EndsWith("/")) { styleLibFolder += "/"; }
            string fullTargetPath = "/Style%20Library" + styleLibFolder + jsFileName;
            var fileCreationInfo = new FileCreationInformation
                Content = System.IO.File.ReadAllBytes(fullSourcePath),
                Overwrite = true,
                Url = fullTargetPath

            Microsoft.SharePoint.Client.File uploadFile = styleLibrary.RootFolder.Files.Add(fileCreationInfo);

You can upload the new .webpart file into the WebPart Gallery using the following CSOM code:

        private static void UploadWebPartFile(ClientContext ctx, string sourcePath, string wepartFileName)
            // get the WebPart Gallery
            List wpGallery = ctx.Site.RootWeb.GetCatalog((int)ListTemplateType.WebPartCatalog);

            // check if there is a file with the same name
                Microsoft.SharePoint.Client.File file = wpGallery.RootFolder.Files.GetByUrl(wepartFileName);

                // delete the file if it already exists
            catch { }

            // updload the webpart file
            var fileCreationInfo = new FileCreationInformation
                Content = System.IO.File.ReadAllBytes(sourcePath + wepartFileName),
                Overwrite = true,
                Url = wepartFileName

            Microsoft.SharePoint.Client.File uploadFile = wpGallery.RootFolder.Files.Add(fileCreationInfo);

If you prefer to work with a SharePoint Add-in to provision your files, you could follow the PnP approach for App Script Parts.

Good luck!

August 6, 2016 Posted by | SharePoint 2010 | Leave a comment

Microsoft stops support for coded sandbox solutions on SharePoint Online

On July 29th Microsoft announced that it starts with the removal of coded sandbox solutions. You can read the official announcement here. This type of sandbox solutions was already deprecated since summer 2014.

A coded sandbox solution contains a DLL. Sandbox solutions containing declarative XML (f.e. the definition of site columns and content types), CSS, JS, images, etc. continue to be supported.

It is recommended to start migrating coded sandbox solutions as of today! You can find a lot of alternatives and sample code based on recommended coded practices on PnP. Additional migration guidance has been published here.

If you’re not sure if your O365 tenant is impacted by this decision, you can run a script that generates a list of sandbox solutions that are installed on your tenant. The report also displays a flag that indicates whether your sandbox solutions contains an assembly or not.

It is possible that your sandbox solutions contains nothing but CSS, JS, images, etc; but that your sandbox solution is listed as containing an assemby. It is possible that your sandbox solution contains an empty assemby. In that case you can get rid of that assembly by simply setting the Include Assembly in Package property to false. This is a property on the Visual studio project itself:


This is also documented in this KB.

August 4, 2016 Posted by | O365, SharePoint Online | Leave a comment

The SharePoint Framework

Microsoft just announced the new development framework to develop against SharePoint. Don’t get worried, don’t throw your Add-ins through the windows, they are there to stay. 🙂  The SharePoint framework is just another tool in your toolbox to to ease the client-side development of pages and (web) parts. This framework will be deployed to SharePoint Online. It is not sure yet if it will be released on SharePoint 2016 on premise, but anyway not on SharePoint 2013 on premise.

You can read more about this brand new SharePoint framework here.

The framework itself is not released yet, but you can start preparing to get started: start learning TypeScript and React, and become familiar with tools like node.js, Gulp, Yeoman, and Git. Waldek Mastykarz has just posted a nice writeup on all these tools and techniques. And more details can be found on the blog of Chris O’Brien.

As a side note, I had the chance to play with a “beta” version during the DevKitchen that was organized in Brussels, a few weeks ago. The SharePoint product group came over to Brussels to give us an introduction to this very new SharePoint framework. And together with the product group, the creme of the european SharePoint community came to Brussels to participate to this event. This was just a week before the explosions, AND in the neighbourhood of one of them. I still get goose bumps when I think about what could have happened to the SharePoint community.

Keep your eyes on the net, because you will see popping up a lot of posts the coming weeks and months 🙂

Have fun!

May 5, 2016 Posted by | SharePoint Online | Leave a comment

Do not use SPUtility.CreateNewDiscussions(SPListItemCollection, title) but use SPUtility.CreateNewDiscussions(SPList, title) method instead

Internally the method SPUtility.CreateNewDiscussions(SPListItemCollection, title) executes the SPListItemCollection.Add method to add a new discussion item to the list.


The SPListItemCollection.Add method is know for loading the whole list item collection into memory before adding a new Discussion list item. This can cause poor performance and even throttling for large lists.

It is recommended to use the SPUtility.CreateNewDiscussions(SPList, title) method instead, as this method internally executes the SPList.ItemAdd() method.


SPList.ItemAdd() does not load all list items in memory but executes a dummy CAML query to retrieve an empty SPListItemCollection to which a new item is added.

March 10, 2016 Posted by | SharePoint 2010 | 1 Comment

Office Dev PnP survey

The Office Dev PnP program has done a great job on building sample code and scenarios to guide SharePoint developers away from Full Trust code solutions to the add-in model. Through the past year the guidance started to evolve to other areas like ffice 365 APIs, Office Add-ins and unified APIs. PnP program has now evolved as open source community effort with both internal and external contributors.

This program is open source and driven by the community, with both internal and external contributors.

Every SharePoint developer gains from this initiative, and therefore I’m a big fan. and therefore I want to ask for your cooperation to fill out the survey http://aka.ms/officedevpnpsurvey.

Thanks for your cooperation!


March 9, 2016 Posted by | SharePoint 2010 | Leave a comment

Repair content type retention policies

Recently I was at a customer who implemented retention policies on content types. They have about 20 content types and a site collection with hundreds of sub sites with thousands of documents.

Becky Bertram wrote a nice detailed article on how to define retention policies for SharePoint 2010.

Problem description

When retention policies are applied on content types, there are 2 timer jobs that run (by default during the weekend):

  • Information management policy timer job: by default, runs on friday 11 PM. The job goes through libraries that have policies applied. It calculates the expiration date for every item.
  • Expiration policy timer job: by default, runs on saturday 11 PM. This job executes the action part of the retention policy. For example, if the action is to move expired documents to the recycle bin, expired documents will be deleted; if the action is set to move the documents to a send-to location, the expired documents will be moved.

My customer explained that when the  timer jobs ran for the first time, they ran for several hours and then just stopped running. As of then the jobs ran weekly, but with a lot of similar errors in the ULS logs:

"Error processing expiration of content in list <list name> in site <url to sp site>. 
 Error: Invalid field name. {b0227f1a-b179-4d45-855b-a18f03706bcb}".
"Error processing expiration of content in list <list name> in site <url to sp site>. 
 Error: Invalid field name. {acd16fdf-052f-40f7-bb7e-564c269c9fbc}".

From this post you can see that these guids refer to out of the box SharePoint fields:

Exempt from Policy b0227f1a-b179-4d45-855b-a18f03706bcb _dlc_Exempt
Expiration Date acd16fdf-052f-40f7-bb7e-564c269c9fbc _dlc_ExpireDate

When I tried to take a look at the Compliance Details of a document, I got the following error message: “column  ‘_dlc_exempt’ does not exist. It may have been deleted by another user”.

The Compliance Details menu option is only available on the context menu when retention policies are active for the specific document:

Compliance Details

After investigation I found out that part of the sites had the retention policies correctly applied, and part of the sites had not.

In the rest of the article I will use the term “active retention policy”. It means that retention policies are defined on content types, and that these content types are in use on a document library.


I found a very interesting article on retention policies on the net to get me started, so kudos to Mike Berryman. I started my investigation to repair the site collection based on this article.

When a SharePoint web has active retention policies, it should have the following properties on its property bag:

  • allowslistpolicy
  • dlc_sitehaspolicy
  • dlc_sitehasexpirationpolicy
  • dlc_webhasexpirationpolicy

As some SharePoint sites were working correctly and some not, I corrected each SPWeb as follows:

$web.AllowUnsafeUpdates = $true
if (!($web.Properties.ContainsKey("allowslistpolicy")))
    $web.Properties.Add("allowslistpolicy", $true)
if (!($web.Properties.ContainsKey("dlc_sitehasexpirationpolicy")))
    $web.Properties.Add("dlc_sitehasexpirationpolicy", $true)
if (!($web.Properties.ContainsKey("dlc_sitehaspolicy")))
    $web.Properties.Add("dlc_sitehaspolicy", $true)
if (!($web.Properties.ContainsKey("dlc_webhasexpirationpolicy")))
    $web.Properties.Add("dlc_webhasexpirationpolicy", $true)

A library with active retention policies should have the following hidden fields:

  • _dlc_Exempt
  • _dlc_ExpireDateSaved
  • _dlc_ExpireDate

I used the XML definition of these fields in order to create the missing fields:

$displayName_exempt = "Exempt from Policy"
$schemaXml_exempt = "<Field ID='{B0227F1A-B179-4D45-855B-A18F03706BCB}' 
    Name='_dlc_Exempt' StaticName='_dlc_Exempt' DisplayName='_dlc_Exempt' 
    SourceID='http://schemas.microsoft.com/sharepoint/v3' Group='Document and Record Management Columns' 
    Type='ExemptField' Indexed='FALSE' Hidden='TRUE' CanToggleHidden='TRUE' 
    ShowInNewForm='FALSE' ShowInEditForm='FALSE' ShowInFileDlg='FALSE' ShowInDisplayForm='FALSE' Required='FALSE' 
    Sealed='TRUE' ReadOnly='TRUE' OverwriteInChildScopes='TRUE'/>"
$displayName_expireDateSaved = "Original Expiration Date"
$schemaXml_expireDateSaved = "<Field ID='{74E6AE8A-0E3E-4DCB-BBFF-B5A016D74D64}' 
    Name='_dlc_ExpireDateSaved' StaticName='_dlc_ExpireDateSaved' DisplayName='_dlc_ExpireDateSaved' 
    SourceID='http://schemas.microsoft.com/sharepoint/v3' Group='Document and Record Management Columns' 
    Type='DateTime' Indexed='FALSE' Hidden='TRUE' CanToggleHidden='TRUE' ShowInNewForm='FALSE' 
    ShowInEditForm='FALSE' ShowInFileDlg='FALSE' ShowInDisplayForm='FALSE' Required='FALSE' 
    Sealed='TRUE' ReadOnly='TRUE' OverwriteInChildScopes='TRUE' />"

$displayName_expireDate = "Expiration Date"
$schemaXml_expireDate = "<Field ID='{ACD16FDF-052F-40F7-BB7E-564C269C9FBC}' 
   Name='_dlc_ExpireDate' StaticName='_dlc_ExpireDate' DisplayName='_dlc_ExpireDate' 
   SourceID='http://schemas.microsoft.com/sharepoint/v3' Group='Document and Record Management Columns' 
   Type='DateTime' Indexed='TRUE' Hidden='TRUE' CanToggleHidden='TRUE' 
   ShowInNewForm='FALSE' ShowInEditForm='FALSE' ShowInFileDlg='FALSE' ShowInDisplayForm='FALSE' Required='FALSE' 
   Sealed='TRUE' ReadOnly='TRUE' OverwriteInChildScopes='TRUE' />"

As some SharePoint libraries were working correctly and some not, I corrected each SPList as follows:

function RepairField($lib, $fieldInternalName $fieldDisplayName, 
       $schemaXml) { 
   $field = $null
      $field = $lib.Fields.GetFieldByInternalName($fieldInternalName)
   catch {}
   if ($field -eq $null)
       $field = $lib.Fields[$fieldInternalName]
       if ($field -ne $null)
            $field.Title = $fieldDisplayName
RepairField($lib, "_dlc_exempt", 
RepairField($lib, "_dlc_ExpireDateSaved", 
RepairField($lib, "_dlc_ExpireDate", 

Each document on which a retention policy applies, has the following properties in its property bag:

  • ItemRetentionFormula
  • _dlc_ItemStageId
  • _dlc_ItemScheduleId

I corrected the items as follows:

$files = $lib.RootFolder.Files
foreach ($file in $files)
    $item = $file.Item
    # add the property ItemRetentionFormula to the property bag
       write-host ("set the retention properties  for item ID " + $item["ID"])
    $item["Modified"] = $item["Created"]

    if ($item.Properties.ContainsKey("ItemRetentionFormula") -eq $false)
        $item.Properties.Add("ItemRetentionFormula", $true)            
    if ($item.Properties.ContainsKey("_dlc_ItemStageId") -eq $false)
        $item.Properties.Add("_dlc_ItemStageId", $true)
    if ($item.Properties.ContainsKey("_dlc_ItemScheduleId") -eq $false)
        $item.Properties.Add("_dlc_ItemScheduleId", $true)
    # clear the item stage id
    $item.Properties["_dlc_ItemStageId"] = ""
    # set the schedule type
    $item.Properties["_dlc_ItemScheduleId"] = $null
    # set the property to contain the formula
    $item.Properties["ItemRetentionFormula"] = $null

For completeness, at the end I disposed the SPWeb object:

$web.AllowUnsafeUpdates = $false


  • I used the server-side object model.
  • As I had to work on the production environment, I had to write the code in PowerShell, but you can also do it in C#.





January 4, 2016 Posted by | SharePoint 2010, SharePoint 2013 | 7 Comments

SharePoint 2007 out of support

Last week I received the question “Is SharePoint 2007 still supported?”

My first reaction was “euh, what??” SharePoint 2010 mainstream support has just ended, so SharePoint 2007 is surely not supported anymore. But the customer referred to this page, saying that SharePoint 2007 was under support till October 10th 2017: https://support.microsoft.com/nl-be/lifecycle/search?sort=PN&alpha=SharePoint%20Server%202007&Filter=FilterNO

My colleague Peter Loete and I took a closer look to the support page, and this is the table from which the customer deducted that SharePoint 2007 was still under support:

SP2007 support

So, no mainstream support anymore for SP2007 SP3: it ended on September 10th 2012. Extended support, at the other side, is still available till October 10th 2017.

What’s included in extended support? You can read more about it here.

October 18, 2015 Posted by | SharePoint 2010 | Leave a comment

SharePoint 2010 mainstream support ends on October 13th 2015

Although a lot of my customers still are on SharePoint 2010, mainstream support ends for all SharePoint 2010 installations with SP2 on October 13th 2015. If you’re not yet on SP2, you can still download it from here.

If you are a Microsoft Premier customer, you get extended support. You can read the details here.

You can find official information on the product lifecycle of SharePoint 2010 here.

October 18, 2015 Posted by | SharePoint 2010 | Leave a comment

Add or Modify SharePoint 2013 Search Topology using a PowerShell built User Interface

In SharePoint 2013, there is no real user interface to modify the search topology. Well, there is, but you can only use for a single server farm. If you have more servers in your SharePoint farm, you have to do this through PowerShell.

One of my South-African Premier Field Engineer colleagues Scott Stewart developed a tool on top of PowerShell WITH UI to create or modify a search topology.

Read more about it here: “Add or Modify SharePoint 2013 Search Topology using a PowerShell built User Interface

A few screenshots to get you curious 🙂

Search topology tool

Search topology

It looks like a very promising tool! Have fun with it!

May 22, 2015 Posted by | Search, SharePoint 2013 | 1 Comment

Creating fields using CSOM in SharePoint 2013

Recently, I was working on a project where we wanted to create SharePoint fields using the .NET client object model (CSOM). You can easily find examples on the internet on how to create a text field, but we had a lot of trouble to find out how to create a calculated field.

Calculated fields can be created in a number of flavors. The following code snippet generates a calculated field that will show employee data as follows:

Karine Bosch (id 82176)

string formula = "<Formula>=FirstName&amp; \" \" &amp;LastName&amp; \" (id: \" &amp;EmployeeID&amp; \" \"</Formula>"
      + "<FieldRefs>"
      + "<FieldRef Name='FirstName' />"
      + "<FieldRef Name='LastName' />"
      + "<FieldRef Name='EmployeeID' />"
      + "</FieldRefs>";

string schemaCalculatedField = "<Field ID='<GUID>' Type='Calculated' Name='FullName' StaticName='FullName' 
   DisplayName='Full Name' ResultType='Text' Required='TRUE' ReadOnly='TRUE'>" + formula + "</Field>";
Field fullNameField = demoList.Fields.AddFieldAsXml(schemaCalculatedField, true, AddFieldOptions.AddInternalNameHint);

You can read more on how to create other type of fields using CSOM here.


To learn how to provision a calculated field, I created this type of field in a blank site through the UI of SharePoint. Then I saved the site as template. This saves the site as a .wsp and stores it in the Solution gallery. From there you can download it and import it in Visual Studio.

Import wsp

Once the .wsp imported, you will notice that all fields, content types, list instances and other elements that make up your site are translated into CAML and features:

Import wsp 2

Click on the elements.xml in the Fields folder to see the definition of all fields.


May 14, 2015 Posted by | CSOM, SharePoint 2013 | Leave a comment