Back to Questions
10.0k
views6
comments
0
score		+
-
F# Pretty Printer and C# to F# translator
f#c#

Hello!

All are invited to test Pretty Printer to format the source files of projects on F #. The utility is a console application (-h for help screen). Requirement for utilities is the availability of set. NetFramework 2.0 and that all files *. fs, the proposed treatment are compiled without errors.

Known issues: conditional compilation directives must only contain within itself the full expression (eg, full function definition).

Suggestions for improvement, comments and description of errors are welcome.

Package of tools can be downloaded from
[link:refsharp.codeplex.com]

.Reply
Hide this comment

Hi,

I am in the middle of converting a large ocaml project into F#. I am also hope to change to light syntax after a while, so I hope to be able to use your program to fix up the source files. The files are created with emacs and uses a lot of tabs.

Some initial comments

  1. I found no source at github, what did I miss?
  2. If the file contains a tab, I get "2001: Unexpected `<Tab>` char in source"
  3. Would be nice if you would allow command line like "-file *.fs" or why not just *.fs. I assume you are not using the Arg-package today?
  4. I do not get source overwriting to work, I send "/r+ -file settings.fs" as arguments
  5. Moves comment to other part of ELSE, see below

Note how the "// default is UTF-8 encoding without a Byte-Order " comment is moved

Before:

1
2
3
4
5

let write_all_text (filename, contents, default_encoding) =
    if utf_8_regexp.Match(contents, 0, min contents.Length 1000 ).Success then // only look in first 1000 chars
        System.IO.File.WriteAllText(filename, contents) // default is UTF-8 encoding without a Byte-Order Mark (BOM),
    else 
        System.IO.File.WriteAllText(filename, contents, default_encoding)

After:

1
2
3
4
5
6
7

et write_all_text ( filename, contents, default_encoding ) = 
    if utf_8_regexp.Match ( contents, 0, min contents.Length 1000 ).Success then 
        // only look in first 1000 chars
        System.IO.File.WriteAllText ( filename, contents )
    else 
        // default is UTF-8 encoding without a Byte-Order Mark (BOM),
        System.IO.File.WriteAllText ( filename, contents, default_encoding )
By on 3/16/2011 3:52 AM ()Reply
Hide this comment

Hi,

1. No. You didn't missed anything. The project is under heavy refactoring and reconstructing, so I'm not ready yet to publish sources. Current version had changed from 0.9 near 80% of code. I think, at this stage, the publish of sources is useless.

2. Yes. Translator is not full ML compatible. So you need replace all `Tab`, that is not permitted in F# sources, to 2 or 4 spaces (depends of your IDE preferences).

3. Will be done.

4. Strange. Maybe your source files is locked when you try to use /r+ switch?

5. Unfortunately, I cannot change this behavior at this time, due specifics of generated AST. This another limitation that I forgot to mention. All comment must be before of source codes, which they describe.

On this time the project has moved to
[link:refsharp.codeplex.com]
and has version 1.0.0317

Thanks for investigation

By on 3/16/2011 10:32 PM ()Reply
Hide this comment

I was hoping to be able to use the prettyprinter to format non-light-syntax code inte light-syntax. Can it be used for this? I.e. it needs to respect parenthesis and begin/end

By on 3/18/2011 12:09 AM ()Reply
Hide this comment

I tried to describe non-light syntax in lexer file FSharp.lex. In most cases it will do tranform non-light to light automaticaly, but I never tested it. You can try to update by yourself grammar rules in FSharp.lex file by eliminating rules of indentation (^#....). Sorry, I have not described the *.lex files yet.

By on 3/25/2011 9:35 AM ()Reply
Hide this comment

By now "C# to F# translator" is ready. Translator does not make any intellectual job, like convert for loops with `break` statements into recursive functions, but it can convert almost all code of C# to F# (linq is not supported). You can try convertor as toy, or as tool for helping conversion from C# code.

[link:refsharp.codeplex.com]

By on 4/8/2011 11:12 PM ()Reply
Hide this comment

The overwriting error only occurs for one file. Maybe it was perfectly formatted before.

By on 3/18/2011 12:05 AM ()Reply

    Topic tags

    See all tags
    IntelliFactory Offices Copyright (c) 2011-2012 IntelliFactory. All rights reserved.
    Home | Products | Consulting | Trainings | Blogs | Jobs | Contact Us | Terms of Use | Privacy Policy | Cookie Policy     		Built with WebSharper

    Logging in...