Boo Programming Language Google Group

Convention based automatic compilation of imports

2013-05-22T07:10:28Z

Hi guys,

I have implemented a really dirty prof of concept to automatically
compile unresolved namespaces on imports [1]. The good news is that it
works beautifully. The bad news is that the devil is in the details and
I'm not completely sure of how it should work :)

The algorithm is something like this:

Re: Positional parameter placeholders

2013-05-21T09:23:45Z

I was investigating gradle and gradle-dotnet this weekend for some hours
but couldn't get it to a point where it actually was an improvement over
xbuild/nant/make. Perhaps I understood it all wrong, but all that
sourceSets and incremental build bless seems to be only available for
the Java plugin.

Re: Positional parameter placeholders

2013-05-20T13:07:41Z

That sounds interesting indeed. I'm a big fan of convention over configuration.

Indeed a setup very similar to what you describe is already possible
with the assembly-bundle gradle plugin in gradle-dotnet-plugins. With
the gradle daemon option it should be fast enough for interactive use.
Coupled with the ability to consume nuget and ivy packages + all the

Re: Positional parameter placeholders

2013-05-20T08:04:15Z

The Go build environment works very much like that; if you say 'go run
my.go' it will recompile any imports if needed. It is a nice feature, and
the result is a fast-compiling language that _feels_ very dynamic, for much
the same reasons as Boo does - extensive use of type inference. Model is

Re: Hints server

2013-05-16T17:22:02Z

Well done!

Re: Hints server

2013-05-16T17:08:36Z

Just a brief follow up... I've refactored the hints server logic into
its own repository [1] and updated the SublimeText plugin [2] for
compatibility with version 3 of the editor (have to check if it supports
version 2 still).

Results are pretty good, certainly an improvement over the experience

Re: Positional parameter placeholders

2013-05-16T16:52:03Z

I'm not a big fan of optional features, Python has the `from __future__
import` syntax to expose new language features which has been more a
headache than a solution in my experience. I understand though the need
to keep the core compiler lean and I agree that this feature is just
syntax sugar and as such it makes sense to have it decoupled from the

Re: Positional parameter placeholders

2013-05-16T15:18:03Z

I like that and I think it would make a lot of sense to have it as an
opt-in feature. I think having more opt-in features in general is
better than the current approach (mea maxima culpa) of adding new
steps to the main pipeline.

I think a good way to enable opt-in features would be through
attributes because they already provide a good way to define the

Re: Macros that expand to expressions

2013-05-16T13:30:50Z

Macros are always statements, since they can yield multiple statements
when executed.

You can either slightly modify the syntax so that it follows this pattern:

mymacro x:
whatever
blah

Or explore the "meta methods", which also work with the compiler AST
nodes but are executed at a later stage, once the compiler has annotated

Re: Macros that expand to expressions

2013-05-16T13:26:12Z

You are correct, macros are currently limited to statements.
Overcoming that limitation will require changes to the parser and
compiler.

Meta methods however can be used as expressions and while more limited
in the types of nodes they can produce - not types nor type members -
they are still very powerful:

Macros that expand to expressions

2013-05-16T12:22:12Z

Maybe I am misunderstanding something, but as far as I can see, macro
expansions have to be statements, meaning that one cannot do something like
the following:

x = mymacro:
whatever
blah

Am I right? Can this be done without compiler changes? If so, how?

Otherwise, would this be worth implementing? I might pick it up if that's

Re: deleteing notes from files+attachment

2013-05-14T05:50:07Z

Hi,

/\/\/.*$|\/\*.*\*\//m.Replace( text, '')

Be aware that this will match every // and /* */ block in the text. If
you need to parse quoted strings so that it can consume something like
"my string contains // but is not a comment", you would need to
implement some kind of parser.

regards,

deleteing notes from files+attachment

2013-05-13T23:48:19Z

hi everyone!
ive been working on this for ages and i didn't mange to come up with a
result ill really appreciate ur help!
im trying to write a function that removes all notes from a file.
both types of notes...
1) anything that is written beyond //
2) anything that is written between /* */
for example.. if the file contains:

Re: Positional parameter placeholders

2013-05-12T15:50:09Z

Hi, Ivan
This feature looks very interesting. I would personally prefer to see it
relying on type inference or dependent on compiler's option Ducky.
It is interesting that we already have placeholders in generic type
definition reference. It uses stars which is not the case for closures
because star is also binary operator. I think may be it could be useful to

Re: Boo Succinctly Revealed (second edition)

2013-05-12T15:42:34Z

My intent with this document is to provide a programmer with the jest of
Boo (so they can get going on it), or to nudge the memory of a Boo
programmer.
As such it should be compact, but not unreadably so.
I don't intend to teach Boo programming, just provide examples.

Thank you for the "property Age as int" form (I'll include it). It brings