• My first public demo, by way of Brian Harry’s blog
• Keith Hill’s early explorations

And rightfully so: most of the operations you want to do locally are already possible without any special tools.  The “Powershell snap-in system” we know and love today was created to help IT admins manage their servers – even today, only a brave few developers have completely replaced cmd.exe at their daily workstation.

Nevertheless, we shouldn’t be satisfied to have our TFS cmdlets live in their own little world.  Luckily, all the info we could ever want about a TFS version controlled item – including its local state – comes packaged in the ExtendedItem object.  Even more fortuitously, if you revisit the exploration of QualifiedItemSpec with an eye for detail, you’ll notice that one overload of ToQualifiedItem() is not like the others.*  In the constructor where ExtendedItem decays into QualifiedItem, the properties saved into the resulting tuple include LocalItem and VersionLocal instead of more familiar ones like ServerItem.  This is not a coincidence :) 

Let’s sum up what we know so far.  Native TFS objects are funneled down the pipeline by stripping them to bare essentials about the items they represent (QIs); these serve as a common interface between cmdlets.  When an ExtendedItem is stripped down, the next cmdlet will use the local filename & local version info as its parameters.  Get-TfsItemProperty is the cmdlet that returns ExtendedItems.  Capiche?

Time to put our newfound knowledge into action.  If you’ve also been following the last several posts on file manipulation in powershell, these examples will hopefully inspire an “aha!” moment.

[Note that I have get-tfsitemproperty aliased to ‘tfprop’ rather than ‘tfproperties’ as it appears in a default install of the Power Tools.]

Are we having fun yet?!

So far we’ve only really exercised the filename part of the QI.  Equally interesting things happen when you take advantage of the special ability that Get-TfsItemProperty offers to manipulate the local version.

[Microsofties: think bbpack / jjpack]

Opens up a whole new world of TFVC scripting mania, wouldn’t ya say?

*I’ll admit the one for GettingEventArgs is a little funky as well.  Here, the choice of properties to expose came down to UX / gut feel.  TargetLocalItem is usually what you want to see; it tends to be the “final result” of mainline operations, surviving things like Rename and Undo fairly well.  But it’s not always the best choice.  Obviously when it’s null we need a backup; good old ServerItem will do.  It’s [almost always] guaranteed to be present, but it’s  not always perfect either – now you have the occasional strange appearance of server paths in your list of local files being touched by the Update.  Finding the right info gets really tricky really fast: was TargetLocalItem null because of a pending Delete, because it’s cloaked or otherwise excluded from our workspace, or because we’re undoing a pending Branch or Merge?  That’s just one example.  The logic behind which name tf.exe displays in which scenario is actually one of the more complex pieces of the app.  With powershell we can accept “good enough” since the object model lets you manipulate things to your heart’s content.

* note that the built-in documentation for Resolve-Path is wrong!  as of CTP3, it’s claiming the output is a string

Plus, you may be working with 3rd party tools that obey none of the above conventions. 

For example, with the TFS Power Tools, Hyung had the foresight to add string[] to the list of types we can convert to QualifiedItems behind the scenes using extra constructors, but we still don’t support the ValueFromPipelineByPropertyName idiom that’s needed to make non-trivial filesystem objects flow.  Furthermore, the objects we output are native TFS types, which were never designed with Powershell in mind.  Most of the time there’s little overlap between these objects and non-TFS cmdlets, but a big exception is the LocalItem property of the ExtendedItem object returned by “tfprop.”  In hindsight, we should have added a NoteProperty that aliased it to Path.  (…or FullName, or Name, depending on whether the PS guys make up their mind :) )

Then there are tools that have nothing at all to do with Powershell.  Some sort of wrapper script is going to be required regardless, but without consistency in the kind of input your wrapper can expect, it can get needlessly complex.

My solution to all of the above is outlined below.  Time to let the code do the talking:

Who knew that ValueFromPipelineByPropertyName worked on aliases?  I didn’t.  I thought it would be nice, but since I didn’t see it documented anywhere, my initial sketch for this function used a ton of ugly ParameterSets as a workaround.  Somewhere in there I decided to try Aliases; lo and behold, it works!  Yay.

This function is useful on its own – if nothing else I use it as a wholesale replacement for resolve-path, whose behavior on UNC paths is bizarre in my opinion – but its real value is as a helper.  For example, a couple snippets from my $profile:

I think the PS parser should be smart enough that double-parens are not required in the second example, but oh well.  Pretty simple overall, and super convenient.  Enjoy!

First, let’s look back at the legacy tf.exe command line.  In the effort to be both user friendly and flexible, its behavior has actually become quite complicated.  As of TFS 2008, the logic appears to be evaluated in this order of precedence:

1. If the /prompt flag is present for this command, always use Prompt mode, otherwise
2. If the /noprompt flag is present for this command, always use Noprompt mode, otherwise
3. If the tf.exe process’ stdout stream is redirected, use Noprompt mode (unless the environment variable TFS_IGNORESTDOUTREDIRECT is set), otherwise
4. If executing in script mode with a @command file, use Noprompt mode (unless the most recent invocation of the setnoprompt command was “setnoprompt false”), otherwise
5. Default to Prompt mode

Yikes.  I’d actually forgotten just how many variables there were until I started running tests for this blog post.  Suffice to say, the Powershell cmdlets approach things in a much simpler way:

1. If the –prompt switch is present for this cmdlet, always use Prompt mode, otherwise
2. Default to Noprompt mode

Simple enough.  I know a lot of power users will rejoice at this, having long since resigned themselves to /i muscle memory lest random WinForms interrupt their console workflow.  However, there are ramifications to this behavior that you should be aware of.  “Prompt mode” means more than simply popping up dialogs.  First of all, there are some UI features that are still not accessible in Prompt mode without additional parameters, such as tf diff /configure.  More importantly, leaving Prompt mode turns off a number of safeguards:

• Undo will permanently discard your local changes.  (I think TFS should use the Recycle Bin, but that’s another show…)
• Shelve /delete will permanently delete shelvesets.  (The server used to keep them around until a cleanup job ran, but again that’s the way things work now.)
• Destroy, well, destroys.

These aren’t the only examples; you get the idea.  None of this behavior is new to TFS in general, but with the default settings of the Powershell tool, it’s a lot easier to shoot yourself in the foot.  Even good old Checkin is likely to surprise you the first time you see it simply go without any further interaction.  [a “no blank comments” checkin policy might help here, nudge nudge] 

So have fun, just be careful.  If you’re using the power tools then you are ready to be a power user, right? :)

Like everyone else, I wish the TFS team had supported x64 much earlier on.  You can see in Microsoft.TeamFoundation.Common.dll that while there are a decent number of routines under the NativeMethods static class that need to be ported, it’s not an enormous task.  Turns out that supporting setup and deployment on x64 servers (with all the requisite expansion of the test matrix, etc.) is by far the higher cost.  Good news is that x64 is supposed to be on the Rosario feature list.

In the meantime, I can’t have John Robbins’ impression of our power tool be negative!  I consider my trip to attend his “Native Windows Debugging” class one of the highlights of my MS career.  As a TFS power user I frequently live in x86 shells, so here’s a tidbit from my $profile that helps keep me sane:

$isX86Process = ( $env:Processor_Architecture -eq "x86" )

...

$Global:PscxHostTitlePreference = { $(if ($IsAdmin) { "Admin: " }) +
                                    $PscxWindowTitlePrefix +
                                    $(get-location).Path +
                                    $(if ($IsX86Process) {" (x86)"})
                                  }

Naturally, I like Vivek’s auto-Fork example too – may need to incorporate somewhere into my library of stuff…

Unfortunately, this syntax is not supported.  (Neither is the /workspace parameter, in case you were wondering.)  It’s an oversight I hope will be corrected in the future.  Until then, here’s a quick fix:

tf view $/project/file.txt > temp.txt
tf diff c:\file.txt temp.txt
del temp.txt

This technique should be easy to put into your scripting language of choice.

However, there are some key differences:

• ConvertTo-FixedByte lets you specify any width between 6 and 22, while Format-Byte is fixed at 10 characters.
• ConvertTo-FixedByte allows the entire UInt64 range (~10^20), while Format-Byte only goes up to TB (~10^15).
• ConvertTo-FixedByte does not allow pipeline input, while Format-Byte does.
• ConvertTo-FixedByte uses the SI standard prefixes, as recommended by IEC, IEEE, EU, and NIST.  Format-Byte uses binary-based prefixes (although it does not display them as KiB, MiB, etc).
  PS> convertto-fixedbyte 2000 -width 10
        2 KB
  PS> format-byte 2000
    1.953 KB
• ConvertTo-FixedByte uses the biggest possible unit so long as the value is at least one unit.  Format-Byte waits until you’ve exceeded 2^n to use the next biggest unit.
  PS> convertto-fixedbyte 1000 -width 10
        1 KB
  PS> format-byte 1024
     1024 B
• Both cmdlets attempt to output a fixed width by rounding, but Format-Byte’s approach fails to handle some edge cases correctly:
  PS> convertto-fixedbyte (1000*999) -width 10
      999 KB
  PS> convertto-fixedbyte (1000*999+1) -width 10
  999.001 KB
  PS> convertto-fixedbyte (1000*999+499) -width 10
  999.499 KB
  PS> convertto-fixedbyte (1000*999+500) -width 10
        1 MB
  PS> convertto-fixedbyte (1000*999+501) -width 10
        1 MB
  PS> convertto-fixedbyte (1000*1000-1) -width 10
        1 MB
  PS> convertto-fixedbyte (1000*1000) -width 10
        1 MB
  PS> convertto-fixedbyte (1000*1000+1) -width 10
        1 MB
  PS> format-byte (1024*1023)
     1023 KB
  PS> format-byte (1024*1023+1)
  1023.001 KB
  PS> format-byte (1024*1023+511)
  1023.499 KB
  PS> format-byte (1024*1023+512)
  1023.5 KB
  PS> format-byte (1024*1023+513)
  1023.501 KB
  PS> format-byte (1024*1024-1)
  1023.999 KB
  PS> format-byte (1024*1024)
     1024 KB
  PS> format-byte (1024*1024+1)
        1 MB
• Both cmdlets omit trailing 0’s, instead padding the front with extra spaces.  [thought there was a slight difference here, but now I can’t repro it!]
• And of course: PSCX is open-source, while the Power Tools are not.  So I can’t read the PSCX source and Keith can’t read mine.  Even so, I hope some of the ideas can cross-pollinate :)

<View>
  <Name>Item</Name>
  <ViewSelectedBy>
    <TypeName>Microsoft.TeamFoundation.VersionControl.Client.Item</TypeName>
  </ViewSelectedBy>
  <TableControl>
    <TableHeaders>
      <TableColumnHeader>
        <Width>7</Width>
        <Alignment>Right</Alignment>
      </TableColumnHeader>
      <TableColumnHeader>
        <Label>CheckinDate</Label>
        <Width>10</Width>
        <Alignment>Right</Alignment>
      </TableColumnHeader>
      <TableColumnHeader>
        <Label>ContentLength</Label>
        <Width>7</Width>
        <Alignment>Right</Alignment>
      </TableColumnHeader>
      <TableColumnHeader>
        <Label>ServerItem</Label>
        <Alignment>Left</Alignment>
      </TableColumnHeader>
    </TableHeaders>
    <TableRowEntries>
      <TableRowEntry>
        <TableColumnItems>
          <TableColumnItem>
            <PropertyName>ChangesetId</PropertyName>
          </TableColumnItem>
          <TableColumnItem>
            <ScriptBlock>$_.CheckinDate.ToString('d')</ScriptBlock>
          </TableColumnItem>
          <TableColumnItem>
            <ScriptBlock>
              if ($_.ItemType -eq [Microsoft.TeamFoundation.VersionControl.Client.ItemType]::Folder)
              { '&lt;DIR&gt;' }
              else
              { ConvertTo-FixedByte $_.ContentLength 7 }
            </ScriptBlock>
          </TableColumnItem>
          <TableColumnItem>
            <ScriptBlock>
              ConvertTo-FixedPath $_.ServerItem 7,10,6
            </ScriptBlock>
          </TableColumnItem>
        </TableColumnItems>
      </TableRowEntry>
    </TableRowEntries>
  </TableControl>
</View>

And a few examples of what it looks like, taken from an 80- and a 112-character-wide console:

PS> tfdir 

Changes CheckinDat Content ServerItem
   etId          e  Length
------- ---------- ------- ----------
   8224  2/25/2009   <DIR> $/Test-ConchangoV2/super/u...g/directorynamefromhell
   8224  2/25/2009   11 KB $/Test-ConchangoV2/super/u...efromhell/devilsurf.jpg
   8224  2/25/2009   12 B  $/Test-ConchangoV2/super/u...orynamefromhell/foo.txt
   8224  2/25/2009    5 MB $/Test-ConchangoV2/super/u.../PowerGUI.1.6.1.639.msi

PS> tfdir 

Changes CheckinDat Content ServerItem
   etId          e  Length
------- ---------- ------- ----------
   8224  2/25/2009   <DIR> $/Test-ConchangoV2/super/ultra/very/very/mega/long/directorynamefromhell
   8224  2/25/2009   11 KB $/Test-ConchangoV2/super/ultra/very/very/m...ong/directorynamefromhell/devilsurf.jpg
   8224  2/25/2009   12 B  $/Test-ConchangoV2/super/ultra/very/very/mega/long/directorynamefromhell/foo.txt
   8224  2/25/2009    5 MB $/Test-ConchangoV2/super/ultra/very/very/m...torynamefromhell/PowerGUI.1.6.1.639.msi

You can see that the Powershell team’s “expand toward the right to minimize visual variance” concept is used to great effect.  However, we violated their “never provide a label” maxim in 3 out of 4 columns.  Oops!  Seriously though, I don’t think anyone would want to use a ScriptProperty with the exact parameters found in the formatter XML.  What we provided instead is far more useful: a couple of fully generic cmdlets named ConvertTo-FixedByte and ConvertTo-FixedPath.  I’ll cover the former in another post; there’s a fair amount of detail to dig into, but its use here is rather mundane. 

ConvertTo-FixedPath is what really makes this example shine.  If you’re like me, you hate managing long paths in shell environments.  Wrapping onto multiple lines, truncating the front, and (especially) truncating the end of the path are all unacceptable.  Cmdlets to the rescue!  This one has two modes (“parameter sets” in PS-speak).  The first is obvious enough:

PS> convertto-fixedpath 12345678901234567890 -width 16

12345678...67890

The 20-character input string is truncated to 16, but only by omitting things in the middle.  (Note that this mode has an off-by-one bug.  Hint: odd numbers.)  This allows you to answer the important questions: what is the file name? what branch/workspace are they located in?  Usually we don’t care as much about the intermediate directories.  Maybe the same is true for other kinds of strings too – let me know if you find it useful!

The other mode is designed explicitly for use in a ps1xml formatter.  Consider:

PS> convertto-fixedpath 12345678901234567890 -otherwidths 30, 20, 10

12345678...67890

Same output, but what the heck are the “other widths”?  Imagine this string were going into a variable-length field.  We want the string to expand dynamically, showing us as much of the filename as possible.  Yet if it does need to be cropped, we want it done intelligently, as if we’d known the width in advance and passed it as a –width parameter.

As it turns out, we can know the final width, given a little hackery.  The Powershell API lets you access the raw console host details, including its BufferSize.  All you need to do is subtract the widths of the other columns, plus room for the separator character PS places between each column.  So imagine the code above is used in a four-column formatter: 80 - 30 - 20 - 10 - (4*1) = 16.  Sure enough, if you scroll back up to the XML, you’ll see the values passed to –otherwidths are no coincidence.  We hand-tweaked every column width so it fit the data type as closely as possible without wrapping*, then plugged in those widths to the convertto-fixedpath cmdlet driving the final variable-length column.

This hack may not always work as Powershell migrates away from the ancient CSRSS hosting environment.  But in theory, anyone who implements PSHostRawUserInterface should be compatible.  It works in the Powershell v2 ISE environment that ships with Win7, for instance.  Custom-PS-tool-lovers, let me know how it goes.

*Except for column titles, obviously.  They were a necessary casualty of two higher-priority design considerations (a) keeping labels identical to the underlying property name (b) supporting the default 80-character console width without squishing average-sized paths.

Unfortunately, the standard answer from the product team didn’t work for me. 

<FIELD type="Integer" name="AreaID" refname="System.AreaId">
  <HELPTEXT>AreaID</HELPTEXT>
</FIELD>
<FIELD type="Integer" name="IterationID" refname="System.IterationId">
  <HELPTEXT>IterationID</HELPTEXT>
</FIELD>
<FIELD type="String" name="Product Area - Validation" refname="Coatue.ProductAreaValidation">
  <HELPTEXT>Hidden field used to validate Product Area</HELPTEXT>
  <WHEN field="System.AreaId" value="84">
    <COPY from="value" value="Restricted" />
  </WHEN>
  <PROHIBITEDVALUES>
    <LISTITEM value="Restricted" />
  </PROHIBITEDVALUES>
</FIELD>
<FIELD type="String" name="Sprint or Release - Validation" refname="Coatue.SprintPathValidation">
  <HELPTEXT>Hidden field used to validate Sprint or Release</HELPTEXT>
  <WHEN field="System.IterationId" value="84">
    <COPY from="value" value="Restricted" />
  </WHEN>
  <PROHIBITEDVALUES>
    <LISTITEM value="Restricted" />
  </PROHIBITEDVALUES>
</FIELD>

It successfully blocked work items that had Area/Iteration paths I didn’t want, but I’d continue to get validation errors after I changed the paths in the form to something else.

At this point I started to wonder whether the AreaId and AreaPath fields were as intricately tied as I thought they were.  After all, I didn’t have any rules saying to update one when the other changed.  (The documentation kinda implies this but doesn’t spell out the mechanism.)  So I decided to test at the object model level: Get-TfsServer OM wrapper to the rescue!

PS C:\Users\rberg> $tfs = get-tfsserver njtfs –all
PS C:\Users\rberg> $bug = $tfs.wit.GetWorkItem(1333)
PS C:\Users\rberg> $bug.AreaId; $bug.IterationId
84
84  

PS C:\workspaces\ws1> $bug.AreaPath; $bug.IterationPath
Test-ConchangoV2\test
Test-ConchangoV2\Release 1\Sprint 4

PS C:\Users\rberg> $bug.fields | where { $_.name.contains("Validation") } | select value
Value
-----
Restricted Restricted                                                                                                                       

PS C:\Users\rberg> $bug.AreaId = 104; $bug.IterationId = 91
PS C:\workspaces\ws1> $bug.AreaPath; $bug.IterationPath
Test-ConchangoV2\test
Test-ConchangoV2\Release 1\Sprint 4 

PS C:\Users\rberg> $bug.fields | where { $_.name.contains("Validation") } | select value
Value
-----
Restricted
Restricted

Nope, not a synchronization issue between path on the form <–> underlying ID.  Looks like I really do need to tweak Gregg’s advice.  All you need is a corresponding WHENNOT rule to cancel out each WHEN rule as it’s no longer needed.  Here’s what the hidden validator for Area looks like now:

<FIELD type="String" name="Product Area - Validation" refname="Coatue.ProductAreaValidation">
  <HELPTEXT>Hidden field used to validate Product Area</HELPTEXT>
  <WHEN field="System.AreaId" value="84">
    <COPY from="value" value="Restricted" />
  </WHEN>
  <WHENNOT field="System.AreaId" value="84">
    <COPY from="value" value="Ok" />
  </WHENNOT>
  <PROHIBITEDVALUES>
    <LISTITEM value="Restricted" />
  </PROHIBITEDVALUES>
</FIELD>

Being relatively new to WIT customization, the answer wasn’t immediately obvious.  TFS doesn’t let you stick an ALLOWEDVALUES rule on fields like System.CreatedDate that the system populates on creation.  And anything you stick on a user-editable field will inevitably block users from navigating an existing work item through the rest of its legitimate workflow.  The path I took was modifying the initial transition from “”.  XML excerpt:

        <TRANSITION from="" to="Active">
          <REASONS>
            <REASON value="Build Failure" />
            <DEFAULTREASON value="New" />
          </REASONS>
          <FIELDS>
            <FIELD refname="Microsoft.VSTS.Common.ActivatedBy">
              <COPY from="currentuser" />
              <VALIDUSER />
              <REQUIRED />
            </FIELD>
            <FIELD refname="Microsoft.VSTS.Common.ActivatedDate">
              <SERVERDEFAULT from="clock" />
            </FIELD>
            <FIELD refname="System.AssignedTo">
              <DEFAULT from="currentuser" />
            </FIELD>
            <FIELD refname="System.Title">
              <ALLOWEDVALUES>
                <LISTITEM value="Before 2009-02-12" />
              </ALLOWEDVALUES>
            </FIELD>
          </FIELDS>
        </TRANSITION>

Most of this comes from the stock MSF Agile v4.2 template; the System.Title field with the strange rule toward the end is my addition.  Doesn’t really matter what you put in the value, so long as you don’t tell your users what it is :)

The arrival of the Fall 2008 Powertools hinted at yet another possibility based on Powershell.  Script-friendly unlocking is particularly enticing for scenarios like the one I found myself in last week: transitioning a team from exclusive checkouts to an edit-merge-commit workflow.  Turning off the exclusive settings on the Team Project is the obvious first step, but what to do about the hundreds of existing pending changes?  I can’t ask everyone to suddenly checkin their work (which may be far from ready to share, or even build), but I also can’t reorganize the tree unless every lock is removed.  Thus:

del temp.hat 2> $null
foreach ($pendingSet in get-tfspendingchange -user *)
{
    $workspace = '"' + $pendingSet.Name + '";' + $pendingSet.OwnerName
    $pendingSet.PendingChanges | where {$_.IsLock} | foreach {
      $item = '"' + $_.ServerItem + '"'
      "lock /lock:none /workspace:$workspace $item" | out-file temp.hat -append
    }
}
tf "@temp.hat"
del temp.hat

This isn’t the cleanest solution in the world.  You couldn’t just wire up the “find all locks in the system” example I provided for Brian’s post to a Remove-TfsLock cmdlet -- even if the Power Tools provided one (which they don’t yet).  By convention, a pipeline of TFS cmdlets transmits QualifiedItems, but to unlock an arbitrary item we need two additional pieces of information from the PendingSet object.  So we fall back on good old procedural loops around tf lock.

Could we improve the script?  Sure.  We could batch ~12 items at a time* for a better performance, though the gain is not nearly as big as using a @hatfile in the first place.  We could also be a little more robust.  Get-TfsPendingChange returns a full PendingSet in the general case, but sometimes the return type decays to a simple PendingChange[].  If our query returns exactly 1 workspace, then our logic for populating $workspace won’t work. 

Of course, we could also do a lot worse!  I originally tried to add the wrapper quotes in-place, but quickly got into syntax trouble.  After consulting the relevant Powershell blog post I eventually got it right, but it was an unreadable mess.  For example, line 4 read:

$workspace = "`"$($pendingSet.Name)`";$($pendingSet.OwnerName)"

Yuck.

*3200 character limit on each line of a tf.exe script file; TFS paths are 260 chars max

PS C:\workspaces\ws1\MAIN> tfstatus

Version CreationDa ChangeType      ServerItem
                te
------- ---------- ----------      ----------
   7415  2/13/2009 Edit            $/Research Platform (R...rLife.v2 Web.vssscc
   8024  2/13/2009 Edit            $/Research Platform (R...ckerLife.v2 Web.sln

PS C:\workspaces\ws1\MAIN> tfstatus -user rwilliams

Name                Computer            OwnerName           Type
----                --------            ---------           ----
CoatueResearch      CMXP123             COATUECAP\rwilliams Workspace
BuildTypeEditor_... CMXP123             COATUECAP\rwilliams Workspace

The second command is actually returning a PendingSet[] array.  Each PendingSet contains a PendingChange[] array that’s specific to a particular workspace or shelveset, plus some metadata about the workspace or shelveset itself.

What gives?  Actually, the underlying APIs always return PendingSet[].  If Get-TfsPendingChange merely pushed the result of the API call onto the pipeline, as most other TFS cmdlets do, then every time you ran it you’d get something more like this:

PS C:\workspaces\ws1\MAIN> $ws = get-tfsworkspace .
PS C:\workspaces\ws1\MAIN> $ws.QueryPendingSets($null, $ws.Name, $ws.OwnerName, $false)

Name                Computer            OwnerName           Type
----                --------            ---------           ----
ws1                 RICHARD490          COATUECAP\rberg     Workspace

…which is clearly not as useful as the output of the first “tfstatus” example above.  Furthermore, people only use Get-TfsPendingChange (and its tf.exe predecessor) to query across multiple workspaces 5% of the time, if that.  I decided to optimize for the 95% case.  That meant checking to see how many PendingSets were returned from the server, pushing its PendingChange[] array onto the pipeline instead if there was only 1.

Note: the QueryShelvedChanges API we use takes a single shelveset.  Unfortunately that means the cmdlet does not support queries like “show me everyone who has shelved $/project/foo.cs” in a single server call.  (QueryShelvedChanges() does have the necessary overloads; we simply inherited this limitation from tf.exe and didn’t have time to rewrite it).  Thus, whenever you specify the –shelveset parameter to Get-TfsPendingChange, you are assured to get a PendingChange[].

for /f "usebackq tokens=2 delims=$" %i in (`tf changeset 1256 /i ^| find "$/"`) do tf label goodbuild $%i;C1256

Yuck (not to mention slow).

And so an idea was born.  Much progress came “free” from Powershell itself: it’s far, far easier to script than cmd.exe is, and when scripting fails, you can drop right into the TFS .Net API without missing a beat.  Some more came from conventions -- standardized verbs, standardized parameters, etc. – and examining how tf commands should be merged, split, or otherwise molded to fit them.  (I admit to being dragged kicking & screaming through this process!)  But ultimately, the biggest value would come from the ability to seamlessly string together multiple commands.

How to reconcile these goals?  To retain their power, cmdlets must output native API objects wherever possible.  The user receives the full data set in structured, strongly-typed form, including the ability to call properties and methods.  Yet for simplicity, the input parameters are rarely more than a couple strings.  In the pipeline I conceived, a form of weakly-typed items are the “glue” that allow rich outputs to feed limited inputs.

Recall that a QualifiedItem represents a simple tuple of {path, deletionID, versionspec}.  By qualifying individual items on the command line, tf.exe commands let you attach a deletion ID and/or versionspec directly to each.  This is the most general way to control the underlying API that tf offers.

The prototype I wrote in early 2007 was just a crude layer on top of tf.exe plus a “magic” extract-items.ps1 script that could turn any suitable object in the public API back into tf’s internal format – QualifiedItem[].  It worked like this:

tfps changeset 1256 | extract-items | tfps label goodbuild

The syntax was nowhere near idiomatic Powershell; the capabilities unlocked by inserting magic scripts in various places were not at all discoverable.  But I was hooked.  Henceforth, I used this little shell as my daily work environment for testing & debugging TFS.  (Direct programmatic access to the data returned by any tf.exe command sure sped things up.)  All the while I collected real-world use cases, tightened/tweaked/fixed, and shopped the demo around in search of a sympathetic ear.  As anyone who’s worked in a huge organization knows, nothing moves without a lot of pushing…and suffice to say that wasn’t my strong suit.  Eventually I did find myself in Brian’s office in spring 2008.  He gave me a green light, conditioned on fixing the fit & finish to Jeffrey’s satisfaction.

Redesigning the cmdlets based on PS conventions wasn’t a technical challenge per se (modulo the kicking :)).  But having to stick a magic cmdlet in between each pipe wasn’t going to fly, no matter its pretty new name Select-TfsItem and incorporation into the main (read: discoverable) C# snap-in.  The final innovation came from Lee Holmes, who suggested we use Powershell’s built-in type coercion to help the pipeline link up automatically: all you need is a constructor on the input type with a single parameter of the output type.  Hyung implemented it as QualifiedItemSpec:

A QualifiedItemSpec represents the collection of QIs extracted from a single input object.  Cmdlets then take an array of these collections as input.  Where appropriate, some cmdlets process each collection as a single entity and yield control; most, however, batch up the entire list of lists and execute a single server call in EndProcessing().  There are also subtleties in the way different object types are coerced into the much weaker QI type.  In the future we’ll look at the various ways such “chunks” of data stream along a TFS pipeline.

Select-TfsItem remains in the toolkit for completeness, but you can see in Reflector that its implementation has degenerated to nothing; the QualifiedItemSpec[] parameter does all the heavy lifting.  It’s still useful for quickly dumping the contents of a huge object like PendingSet into the console or a file.  Meanwhile, let’s see what the final version of our “label a changeset” script looks like:

Get-TfsChangeset 1256 | New-TfsLabel goodbuild

Easy as pie!  Except for the fact that *-TfsLabel commands haven’t actually shipped :)  I coded cmdlets replacing nearly all of tf.exe before leaving, and I know that power tools development remain in good hands, so I’ll leave you with the above example in the hopes it’ll work after the next release or two.

1) tf get foo.cs  # not qualified – falls back to default
2) tf get foo.cs /version:1234  # not qualified – falls back to /version parameter
3) tf get foo.cs;1234  # qualified
4) tf get foo.cs;X56  # qualified with deletion ID
5) tf get foo.cs;X56;LmyLabel  # qualified with both – betcha didn’t know this was possible
6) tf get foo.cs;1234 bar.cs;X56 baz.cs;789 *quux* /version:LmyLabel  # mix & match
7) tf history foo.cs;1234~5678  # qualified with two versions
8) tf history foo.cs;~5678  # default versionFrom, qualified versionTo
9) tf history foo.cs;5678  # same as above
10) tf history foo.cs;1234~  # qualified versionFrom, default versionTo

“Qualifying” an item means telling tf.exe exactly what version to operate on.  We do this by combining a name with a versionspec.  Most tf commands take qualified items as their main piece of input.  (Commands that don’t include things like Checkin, Shelve, Undo, or Status that need to use whatever’s already in the workspace; and of course ones like Workfold or Configure that don’t touch the repository at all).

Implementation

If you examine tf.exe in Reflector, you’ll see that QualifiedItem is a simple class encapsulating a Path, a DeletionId, and a VersionSpec[] array.  The array almost always has just one member; only a few commands take two versionspecs, and none takes more.  Most items have a deletion ID of 0 to indicate “not deleted.”

To see how it’s used, start by drilling into the VersionControlCommand class to find the methods named ParseQualifiedItem().  I’ll leave this journey as an exercise for the reader.  Suffice to say here’s what you’d find:

Behavior - single versionspec

The goal is to give the user fine-grained control over tf operations.  Examples #1-4 are straightforward.  #5 is a quirk of syntax with no real import.  #6 shows the power of this approach: a single command pulls different versions of foo, bar, baz, while applying a fallback versionspec to all non-qualified items – in this case, the set intersection of all files in the current directory matching the *quux* pattern with the contents of the “myLabel” label.

If not specified by a /version parameter, the default versionspec is T.

Behavior – version range (2 versionspecs)

Example #7 is straightforward.  #8-10 show off some syntax sugar.  When you have a command that expects a version range, you need only type one end of the range if the default suffices for the other.  The default versionFrom spec is C1 and the default versionTo is T.

Note that there is no opportunity for fallback.  You can use the /version parameter or qualified items, but not both.  All of the commands that operate on a version range have additional limitations that make the general syntax seen in #6 impossible.

• Diff – takes exactly one item with a version range (foo.cs;10~20), or two items with one versionspec each (foo.cs;10 bar.cs;20), but no other variations.
• History – takes exactly one item with a version range.
• Merge – takes exactly two items.  The first one has a version range and the second has no version.
• Rollback – introduces some new parameters that are beyond the scope of this post, not to mention still subject to change before TFS 2010’s release.