and another bug:
(I've updated to the correct stable version as per your instructions, since my last post included). Not sure how to reproduce this bug but it just happened after switching back to MD (i'm on OSX):

A fatal error has occurred

Details of this error have been automatically sent to Xamarin for analysis. MonoDevelop will now close.

System.ArgumentOutOfRangeException: Argument is out of range.
  at MonoDevelop.D.Formatting.Indentation.IndentStack.Push (Inside inside, Byte keyword, Int32 lineNr, Int32 nSpaces) [0x00000] in <filename unknown>:0
  at MonoDevelop.D.Formatting.Indentation.DIndentEngine.PushHash (Inside inside) [0x00000] in <filename unknown>:0
  at MonoDevelop.D.Formatting.Indentation.DIndentEngine.Push (Char c) [0x00000] in <filename unknown>:0
  at MonoDevelop.Ide.Gui.Content.DocumentStateTracker`1[MonoDevelop.D.Formatting.Indentation.DIndentEngine].UpdateEngine (Int32 position) [0x0015b] in /Users/builder/data/lanes/monodevelop-lion-monodevelop-3.0-series/90a53d19/source/monodevelop/main/src/core/MonoDevelop.Ide/MonoDevelop.Ide.Gui.Content/DocumentStateTracker.cs:130
  at MonoDevelop.D.Formatting.Indentation.DIndentationTracker.GetIndentationString (Int32 offset, DocumentLocation loc) [0x00000] in <filename unknown>:0
  at MonoDevelop.D.Formatting.Indentation.DIndentationTracker.GetIndentationString (Int32 lineNumber, Int32 column) [0x00000] in <filename unknown>:0
  at MonoDevelop.D.Formatting.Indentation.DIndentationTracker.GetVirtualIndentationColumn (Int32 lineNumber, Int32 column) [0x00000] in <filename unknown>:0
  at Mono.TextEditor.TextEditorData.GetVirtualIndentationColumn (DocumentLocation loc) [0x00000] in /Users/builder/data/lanes/monodevelop-lion-monodevelop-3.0-series/90a53d19/source/monodevelop/main/src/core/Mono.Texteditor/Mono.TextEditor/TextEditorData.cs:1073
  at Mono.TextEditor.TextViewMargin.MousePressed (Mono.TextEditor.MarginMouseEventArgs args) [0x001fc] in /Users/builder/data/lanes/monodevelop-lion-monodevelop-3.0-series/90a53d19/source/monodevelop/main/src/core/Mono.Texteditor/Mono.TextEditor/Gui/TextViewMargin.cs:1640
  at Mono.TextEditor.TextEditor.OnButtonPressEvent (Gdk.EventButton e) [0x000a3] in /Users/builder/data/lanes/monodevelop-lion-monodevelop-3.0-series/90a53d19/source/monodevelop/main/src/core/Mono.Texteditor/Mono.TextEditor/Gui/TextEditor.cs:1025
  at Gtk.Widget.buttonpressevent_cb (IntPtr widget, IntPtr evnt) [0x00000] in <filename unknown>:0

On Wednesday, 16 January 2013 at 07:13:01 UTC, timotheecour wrote:
> and another bug:
> (I've updated to the correct stable version as per your instructions, since my last post included). Not sure how to reproduce this bug but it just happened after switching back to MD (i'm on OSX):
>
> A fatal error has occurred
>
> Details of this error have been automatically sent to Xamarin for analysis. MonoDevelop will now close.
>
> System.ArgumentOutOfRangeException: Argument is out of range.
>   at MonoDevelop.D.Formatting.Indentation.IndentStack.Push (Inside inside, Byte keyword, Int32 lineNr, Int32 nSpaces) [0x00000] in <filename unknown>:0

Next time, please dump the upper-most entry of the stack trace only ;)

This is fixed now. It'll be part of the next bug-fix release.

On Wednesday, 16 January 2013 at 06:47:49 UTC, timotheecour wrote:
> A)
>> Awesome, again these problems - which MD version do you've got
>> installed?
>> 3.0.6 stable? Then please switch to the
>> mono-d.alexanderbothe.com/stableMD repository
>> 3.1.0 beta? Then switch to the mono-d.alexanderbothe.com repo.
>
> OK, that made it work. Very low priority, but maybe it's possible
> to provide 1 package that would work in both cases (if it were D,
> something like version(MD_3_0_6){}else version(MD_310_beta)
> else{...} but i guess it's C#).

I could modify the addin definition - so yes, let's try to do
this.
>
> B)
>> With code indent I'm not meaning code formatting but the
>> automated indentation generation after e.g. you've pressed
>> <return> and it inserts 3 tabs because you're in a method of a
>> class of an other class.
>
> thanks for finally tackling that! Now that the hard part is done,
> couldn't there be a way to reindent the entire file (or
> selection)  upon request (control + I on OSX for format doc) by
> calling this function on each line or something similar?
> (temporary hack and might be a bit slow but a MUCH needed
> feature).

A solution would be to let the indent engine run through the
entire code and catch the indents of each line..but well, I'll
rather finish my AST visitor, then I can continue with putting a
real formatting engine upon it.

>
> C)
> A question about your choice for C# vs D: wouldn't it have been
> easier to write the bulk of your code in D and interface with the
> MD codebase through some C wrappers? Assuming there's less than
> 100 or so functions you need to interface with... That way you'd
> be able to benefit from hopefully soon to be written code
> formatting, parsers, etc.

C# can interface C dlls. One problem: The huge code base written
in C# and C dlls. The single fact that a huge bunch of
cross-compilation problems etc will stop me even trying to
implement even one e.g. completion feature.
I mean, I'm not the one who sets up a cross-compilation
environment that compiles and packages all the code to a bunch of
binaries. I've chosen C# just _because_ I can distribute it on
all plattforms w/o having to recompile it. Btw, I think there's
no real performance boost when switching to a native code base.
So even a rough effort vs. productivity gain analysis would
clearly speak pro C#. I know that a virtual in-D implementation
of the D language would be great, but well, if you want to start,
I can try to support you ;)

> D)
> BUG: I unchecked the option "insert * or + at comment new line"
> but it still inserts "+" after a new line in a "/+" comment.

Will be fixed in the next version.
>
> E)
> BUG: create a new file, open it in MD (by double clicking for
> example), press a key (eg "a"), it gives the following error (see
> below). We can press OK and it'll work, but that's still a bug:

Fixed.

alex wrote:
> @Fil: I haven't changed anything regarding GDB support (yet). I'm
> sorry but I get myself desperate too about what the guys from MD
> tend to do sometimes. ;)

Okay, I'll ask the Arch community then. Thanks


> -- What is linux-vdso.so.1 ?

It's one of the many magical, compressed quantum alternate-reality links that Linux and GDB use to run smoothly. 'VDSO' stands for "Very Dense Singularity Object". Actually I just made all that up and I have no idea what it is, but I have a feeling I'll have to find out soon in order to fix this problem.

Vdso seems to be a way to make kernel function calls without using system
calls by making the function available in a virtual dynamic shared object
file.
m.linuxjournal.com/content/creating-vdso-colonels-other-chicken
On 16 Jan 2013 17:25, "F i L" <witte2008@gmail.com> wrote:

> alex wrote:
>
>> @Fil: I haven't changed anything regarding GDB support (yet). I'm sorry but I get myself desperate too about what the guys from MD tend to do sometimes. ;)
>>
>
> Okay, I'll ask the Arch community then. Thanks
>
>
>  -- What is linux-vdso.so.1 ?
>>
>
> It's one of the many magical, compressed quantum alternate-reality links that Linux and GDB use to run smoothly. 'VDSO' stands for "Very Dense Singularity Object". Actually I just made all that up and I have no idea what it is, but I have a feeling I'll have to find out soon in order to fix this problem.
>