How to load referenced script libraries from a Roslyn script file? - c#

I figured out I cannot load one script library from another easily:
module.csx
string SomeFunction() {
return "something";
}
script.csx
ExecuteFile("module.csx");
SomeFunction() <-- causes compile error "SomeFunction" does not exist
This is because the compiler does not know of module.csx at the time it compiles script.csx afaiu. I can add another script to load the two files from that one, and that will work. However thats not that pretty.
Instead I like to make my scripthost check for a special syntax "load module" within my scripts, and execute those modules before actual script execution.
script.csx
// load "module.csx"
SomeFunction()
Now, with some basic string handling, I can figure out which modules to load (lines that contains // load ...) and load that files (gist here https://gist.github.com/4147064):
foreach(var module in scriptModules) {
session.ExecuteFile(module);
}
return session.Execute(script)
But - since we're talking Roslyn, there should be some nice way to parse the script for the syntax I'm looking for, right?
And it might even exist a way to handle module libraries of code?

Currently in Roslyn there is no way to reference another script file. We are considering moving #load from being a host command of the Interactive Window to being a part of the language (like #r), but it isn't currently implemented.
As to how to deal with the strings, you could parse it normally, and then look for pre-processor directives that are of an unknown type and delve into the structure that way.

Support for #load in script files has been added as of https://github.com/dotnet/roslyn/commit/f1702c.
This functionality will be available in Visual Studio 2015 Update 1.

Include the script:
#load "common.csx"
...
And configure the source resolver when you run the scripts:
Script<object> script = CSharpScript.Create(code, ...);
var options = ScriptOptions.Default.WithSourceResolver(new SourceFileResolver(new string[] { }, baseDirectory));
var func = script.WithOptions(options).CreateDelegate()
...

Related

How to output a file from a roslyn code analyzer?

I'm using the roslyn API to write a DiagnosticAnalyzer and CodeFix.
After I have collected all strings and string-interpolations, I want to write all of them to a file but I am not sure how to do this the best way.
Of course I can always simply do a File.WriteAllText(...) but I'd like to expose more control to the user.
I'm also not sure about how to best trigger the generation of this file, so my questions are:
I do not want to hard-code the filename, what would be the best way to expose this setting to the user of the code-analyzer? A config file? If so, how would I access that? ie: How do I know the directory?
If one string is missing from the file, I'd like to to suggest a code fix like "Project contains changed or new strings, regenerate string file". Is this the best way to do this? Or is it possible to add a button or something to visual studio?
I'm calling the devenv.com executable from the commandline to trigger builds, is there a way to force my code-fix to run either while building, or before/after? Or would I have to "manually" load the solution with roslyn and execute my codefix?
I've just completed a project on this. There are a few things that you will need to do / know.
You will probably need to switch you're portable class library to a class library. otherwise you will have trouble calling the File.WriteAllText()
You can see how to Convert a portable class library to a regular here
This will potentially not appropriately work for when trying to apply all changes to document/project/solution. When Calling from a document/project/solution, the changes are precalcuated and applied in a preview window. If you cancel, an undo action is triggered to undo all changes, if you write to a file during this time, and do not register an undo action you will not undo the changes to the file.
I've opened a bug with roslyn but you can handle instances by override the preview you can see how to do so here
And one more final thing you may need to know is how to access the Solution from the analyzer which, Currently there is a hack I've written to do so here
As Tamas said you can use additional files you can see how to do so here
You can use additional files, but I know on the version I'm using resource files, are not marked as additional files by default they are embeddedResources.
So, for my users to not have to manually mark the resource as additonalFiles I wrote a function to get out the Designer.cs files associated with resource files from the csproj file using xDoc you can use it as an example if you choose to parse the csproj file:
protected List<string> GetEmbeddedResourceResxDocumentPaths(Project project)
{
XDocument xmldoc = XDocument.Load(project.FilePath);
XNamespace msbuild = "http://schemas.microsoft.com/developer/msbuild/2003";
var resxFiles = new List<string>();
foreach (var resource in xmldoc.Descendants(msbuild + "EmbeddedResource"))
{
string includePath = resource.Attribute("Include").Value;
var includeExtension = Path.GetExtension(includePath);
if (0 == string.Compare(includeExtension, RESX_FILE_EXTENSION, StringComparison.OrdinalIgnoreCase))
{
var outputTag = resource.Elements(msbuild + LAST_GENERATED_TAG).FirstOrDefault();
if (null != outputTag)
{
resxFiles.Add(outputTag.Value);
}
}
}
return resxFiles;
}
For config files you can use the AdditionalFiles msbuild property, which is passed to the analyzers through the context. See here.

Get assembly name at compile time and use it in code [duplicate]

Is there a way to find out the assembly name at design-time (i.e. not using reflection or runtime APIs such as System.Reflection.Assembly.GetEntryAssembly) from within Visual Studio?
The scenario requires a tool to get the assembly name that a Visual Studio project will eventually compile into.
This is like parsing the AssemblyName property of the .csproj - I am wondering if there are any APIs that can give this information reliably.
Please do not respond back with runtime APIs that use reflection - there is no assembly file present at the time I need the assembly name - just the metadata of the assembly in the csproj file.
if you are calling the tool via a post/pre-build event, this data is very easy to access.
Just go to the "project properties->Build Events" tab, then select either "edit pre-build" or "edit post-build", depending on when you want the tool to run. This should bring up an edit window with the ever helpful "Macros >>" button. Press this and you will be given a heap of macros to use and should be pretty much everything you need.
The "API" you could use is LINQ to XML after all the .csproj file is just xml. (and you can get the location of the .csproj file if you need from the solution file which for some reason is not XML but can be easily parsed)
You can use "TargetName" available in Macros for Post-build events. It will give you the assembly name for your project.
After a quick run through MSDN I found this article which might be a good start for some further research:
Accessing Project Type Specific Project, Project Item, and Configuration Properties
I think you will need to write some regular expression that will give you the value of "AssemblyTitle" attribute in AssemblyInfo.cs file.
Something like this:
public class Assembly
{
public static string GetTitle (string fileFullName) {
var contents = File.ReadAllText (fileFullName); //may raise exception if file doesn't exist
//regex string is: AssemblyTitle\x20*\(\x20*"(?<Title>.*)"\x20*\)
//loading from settings because it is annoying to type it in editor
var reg = new Regex (Settings.Default.Expression);
var match = reg.Match (contents);
var titleGroup = match.Groups["Title"];
return (match.Success && titleGroup.Success) ? titleGroup.Value : String.Empty;
}
}

Dynamically compiled project losing resources

I need to compile source code of big project dynamically and output type can be Windows Application or Class Library.
Code is nicely executed and its possible to make .dll or .exe files, but problem is that, when I'm trying to make .exe file - it's losing resources like project icon. Result file doesn't include assembly information to.
Any way to solve this? (Expected result should be the same, that manual Build function on project file in Visual Studio 2015).
Thank you!
var workspace = MSBuildWorkspace.Create();
//Locating project file that is WindowsApplication
var project = workspace.OpenProjectAsync(#"C:\RoslynTestProjectExe\RoslynTestProjectExe.csproj").Result;
var metadataReferences = project.MetadataReferences;
// removing all references
foreach (var reference in metadataReferences)
{
project = project.RemoveMetadataReference(reference);
}
//getting new path of dlls location and adding them to project
var param = CreateParamString(); //my own function that returns list of references
foreach (var par in param)
{
project = project.AddMetadataReference(MetadataReference.CreateFromFile(par));
}
//compiling
var projectCompilation = project.GetCompilationAsync().Result;
using (var stream = new MemoryStream())
{
var result = projectCompilation.Emit(stream);
if (result.Success)
{
/// Getting result
//writing exe file
using (var file = File.Create(Path.Combine(_buildPath, fileName)))
{
stream.Seek(0, SeekOrigin.Begin);
stream.CopyTo(file);
}
}
}
We never really designed the workspace API to include all the information you need to emit like this; in particular when you're calling Emit there's an EmitOptions you can pass that includes, amongst other things, resource information. But we don't expose that information since this scenario wasn't hugely considered. We've done some of the work in the past to enable this but ultimately never merged it. You might wish to consider filing a bug so we officially have the request somewhere.
So what can you do? I think there's a few options. You might consider not using Roslyn at all but rather modifying the project file and building that with the MSBuild APIs. Unfortunately I don't know what you're ultimately trying to achieve here (it would help if you mentioned it), but there's a lot more than just the compiler invocation that is involved in building a project. Changing references potentially changes other things too.
It'd also be possible, of course, to update MSBuildWorkspace yourself to pass this through. If you were to modify the Roslyn code, you'll see we implement a series of interfaces named "ICscHostObject#" (where # is a number) and we get passed the information from MSBuild to that. It looks like we already stash that in the command line arguments, so you might be able to pass that to our command line parser and get the data back you need that way.

Why do 'requires' statements fail when loading (iron)ruby script via a C# program?

IronRuby and VS2010 noob question:
I'm trying to do a spike to test the feasibility of interop between a C# project and an existing RubyGem rather than re-invent that particular wheel in .net. I've downloaded and installed IronRuby and the RubyGems package, as well as the gem I'd ultimately like to use.
Running .rb files or working in the iirb Ruby console is without problems. I can load the both the RubyGems package, and the gem itself and use it, so, at least for that use case, my environment is set up correctly.
However, when I try to do the same sort of thing from within a C# (4.0) console app, it complains about the very first line:
require 'RubyGems'
With the error:
no such file to load -- rubygems
My Console app looks like this:
using System;
using IronRuby;
namespace RubyInteropSpike
{
class Program
{
static void Main(string[] args)
{
var runtime = Ruby.CreateRuntime();
var scope = runtime.ExecuteFile("test.rb");
Console.ReadKey();
}
}
}
Removing the dependencies and just doing some basic self-contained Ruby stuff works fine, but including any kind of 'requires' statement seems to cause it to fail.
I'm hoping that I just need to pass some additional information (paths, etc) to the ruby runtime when I create it, and really hoping that this isn't some kind of limitation, because that would make me sad.
Short answer: Yes, this will work how you want it to.You need to use the engine's SetSearchPaths method to do what you wish.
A more complete example
(Assumes you loaded your IronRuby to C:\IronRubyRC2 as the root install dir)
var engine = IronRuby.Ruby.CreateEngine();
engine.SetSearchPaths(new[] {
#"C:\IronRubyRC2\Lib\ironruby",
#"C:\IronRubyRC2\Lib\ruby\1.8",
#"C:\IronRubyRC2\Lib\ruby\site_ruby\1.8"
});
engine.Execute("require 'rubygems'"); // without SetSearchPaths, you get a LoadError
/*
engine.Execute("require 'restclient'"); // install through igem, then check with igem list
engine.Execute("puts RestClient.get('http://localhost/').body");
*/
Console.ReadKey();

Simplfying DSL written for a C# app with IronPython

Thanks to suggestions from a previous question, I'm busy trying out IronPython, IronRuby and Boo to create a DSL for my C# app. Step one is IronPython, due to the larger user and knowledge base. If I can get something to work well here, I can just stop.
Here is my problem:
I want my IronPython script to have access to the functions in a class called Lib. Right now I can add the assembly to the IronPython runtime and import the class by executing the statement in the scope I created:
// load 'ScriptLib' assembly
Assembly libraryAssembly = Assembly.LoadFile(libraryPath);
_runtime.LoadAssembly(libraryAssembly);
// import 'Lib' class from 'ScriptLib'
ScriptSource imports = _engine.CreateScriptSourceFromString("from ScriptLib import Lib", SourceCodeKind.Statements);
imports.Execute(_scope);
// run .py script:
ScriptSource script = _engine.CreateScriptSourceFromFile(scriptPath);
script.Execute(_scope);
If I want to run Lib::PrintHello, which is just a hello world style statement, my Python script contains:
Lib.PrintHello()
or (if it's not static):
library = new Lib()
library.PrintHello()
How can I change my environment so that I can just have basic statments in the Python script like this:
PrintHello
TurnOnPower
VerifyFrequency
TurnOffPower
etc...
I want these scripts to be simple for a non-programmer to write. I don't want them to have to know what a class is or how it works. IronPython is really just there so that some basic operations like for, do, if, and a basic function definition don't require my writing a compiler for my DSL.
You should be able to do something like:
var objOps = _engine.Operations;
var lib = new Lib();
foreach (string memberName in objOps.GetMemberNames(lib)) {
_scope.SetVariable(memberName, objOps.GetMember(lib, memberName));
}
This will get all of the members from the lib objec and then inject them into the ScriptScope. This is done w/ the Python ObjectOperations class so that the members you get off will be Python members. So if you then do something similar w/ IronRuby the same code should basically work.

Categories