Yesterday I discussed the typical steps you would need to go through to set up a Revit 2015 add-in development environment.
My top priority right now is getting my Tech Summit talk prepared, which involves adding some functionality to the room editor application add-in.
I decided to migrate it to Revit 2015 as well.
Here are the steps I took:
Just like The Building Coder sample migration yesterday, the flat migration proved very easy.
I repeated the same steps:
That was it.
The flat migration compiles with zero errors.
The resulting add-in works fine.
This probably actually means that the Revit 2014 add-in would work fine in Revit 2015 as well.
The compilation does produce five warnings about deprecated API usage; five occurrences, actually, but only two distinct warnings, to be exact:
Let's take a look at those and work towards a completely clean build.
The external command causing the first warning is CmdUploadRooms, where the following code is used to check for pre-selected rooms:
List<ElementId> ids = null; Selection sel = uidoc.Selection; if( 0 < sel.Elements.Size ) { foreach( Element e in sel.Elements ) { if( !( e is Room ) ) { Util.ErrorMsg( "Please pre-select only room" + " elements before running this command." ); return Result.Failed; } if( null == ids ) { ids = new List<ElementId>( 1 ); } ids.Add( e.Id ); } }
Removing that warning is simple, and I can even shorten the code, like this:
Selection sel = uidoc.Selection; ICollection<ElementId> ids = sel.GetElementIds(); if( 0 < ids.Count ) { foreach( ElementId id in ids ) { if( !( doc.GetElement( id ) is Room ) ) { Util.ErrorMsg( "Please pre-select only room" + " elements before running this command." ); return Result.Failed; } } }
The second warning is also trivial to fix.
One of the occurrences is this simple loop:
foreach( View v in sheet.Views ) { GetViewTransform( v ); }
The obsolete Views property returns an obsolete ViewSet class instance.
The replacement method GetAllPlacedViews returns a set of element ids instead, in the shape of a generic set ISet<ElementId>
.
You can use the generic LINQ Select method to convert the element ids straight to View object instances like this, if you like:
foreach( View v in sheet.GetAllPlacedViews() .Select<ElementId, View>( id => doc.GetElement( id ) as View ) ) { GetViewTransform( v ); }
Similar conversions resolve the other two warnings concerning this property.
As you see, I presented some syntax coloured C# source code above, copied and pasted from Visual Studio 2012.
I am still using the J.T. Leigh CopySourceAsHtml utility as described in The Building Coder source code colour coder and my pycolorize.py Python script to strip some extraneous baggage from the HTML code before inserting it into the blog text.
Luckily, the description of how to use CopySourceAsHtml for Visual Studio 2010 also applies to Visual Studio 2012, more or less, as I found out by trial and error.
I now have three versions of the CopySourceAsHtml add-in file located in the following folders for Visual Studio 2008, 2010 and 2012:
The add-in file that I am using for Visual Studio 2012 looks like this:
<?xml version="1.0" encoding="utf-8" standalone="no"?> <Extensibility xmlns= "http://schemas.microsoft.com/AutomationExtensibility"> <HostApplication> <Name>Microsoft Visual Studio Macros</Name> <Version>11.0</Version> </HostApplication> <HostApplication> <Name>Microsoft Visual Studio</Name> <Version>11.0</Version> </HostApplication> <Addin> <FriendlyName>CopySourceAsHtml</FriendlyName> <Description> Adds support to Microsoft Visual Studio 2012 for copying source code, syntax highlighting, and line numbers as HTML. </Description> <Assembly> JTLeigh.Tools.Development.CopySourceAsHtml, Version=3.0.3215.1, Culture=neutral, PublicKeyToken=bb2a58bdc03d2e14, processorArchitecture=MSIL </Assembly> <FullClassName> JTLeigh.Tools.Development.CopySourceAsHtml.Connect </FullClassName> <LoadBehavior>1</LoadBehavior> <CommandPreload>0</CommandPreload> <CommandLineSafe>0</CommandLineSafe> </Addin> </Extensibility>
For the sake of completeness, the current version of my Python clean-up script looks like this:
#!/usr/bin/python # -*- coding: iso-8859-15 -*- # # pycolorize.py - massage colorised HTML source code copied from Visual Studio # # jeremy tammik, autodesk inc, 2009-02-05 # # History: # # 2009-02-05 initial version # 2009-05-22 updated to support Visual Studio 2008 # 2011-04-19 updated to support Visual Studio 2010 and CopySourceAsHtml 3.0 # 2013-05-21 migrated to mac os x unix, cf. http://coffeeghost.net/src/pyperclip.py # # read a block of text from a file or the windows clipboard # replace cb[12345] by the appropriate colour # remove the style and pre tags # import os, re color_map = { '#2b91af' : 'teal', '#a31515' : 'maroon' } def getTextMac(): outf = os.popen('pbpaste', 'r') content = outf.read() outf.close() return content def setTextMac(text): outf = os.popen('pbcopy', 'w') outf.write(text) outf.close() def getTextWin(): w.OpenClipboard() d = w.GetClipboardData( win32con.CF_TEXT ) w.CloseClipboard() return d def setTextWin( aType, aString ): w.OpenClipboard() w.EmptyClipboard() w.SetClipboardData( aType, aString ) w.CloseClipboard() _regexColor = re.compile( '\.(cb[1-9]) \{ color\: ([#0-9a-z]+); \}' ) _regexStyle = re.compile( '(<style type="text/css">.*</style>\s*<div class="cf">\s*)', re.DOTALL ) _regexEnd = re.compile( '(</pre>\s*</div>)', re.DOTALL ) #_regexPreEnd = re.compile( '(</pre>$)' ) def replace_cb_by_color( s ): "Search for '.cb1 { color: blue; }' and globally replace cb[1-9] by the appropriate colour." m = _regexColor.search( s ) if m: #print 'match found' a = m.groups() #print a if 2 == len( a ): color = a[1] #print color if color_map.has_key( color ): color = color_map[color] #print color return True, s.replace( a[0], color ) #else: # print 'no match found' return False, s def main(): 'Convert Visual Studio CopySourceAsHtml colour styles to a more compact form.' s = getTextMac() #s = '''<style type="text/css"> ... ''' #print s go = True while go: go, s = replace_cb_by_color( s ) #print s m = _regexStyle.match( s ) #print m if m: s = s.replace( m.group( 1 ), '' ) #print s s = s.replace( '<pre class="cl">', '' ) #print s m = _regexEnd.search( s ) #print m if m: s = s.replace( m.group( 1 ), '' ) #print s s = s.strip().replace( '</pre>', '' ) #print s setTextMac( s ) #print s if __name__ == '__main__': main()
The RoomEditorApp source code, Visual Studio solution and add-in manifest live in the RoomEditorApp GitHub repository.
The versions discussed above are release 2015.0.2.6 for the flat migration and release 2015.0.2.7 with the deprecated API usage cleaned up.
Next step: finally get going with the Tech Summit preparation!