Log of the #nice channel on irc.freenode.net

Using timezone: Central European Time
* CIA-2 leaves10:49
* CIA-2 joins11:44
* dbr joins17:18
* ram_ joins17:41
* ram_ leaves17:42
* sriramalka joins17:43
Wondering if anyone's working on fixing nice with java 1.5?17:44
<dbr>not actively, but it should indeed be done...17:45
do you have the possibility to help?
<sriramalka>Yes, I am very interested indeed
<dbr>cool!17:48
what problems did you identify?17:49
don't hesitate to ask for help when you need ot
it
<sriramalka>Thanks. This was the ambiguity error about "append()" in Constraint.nice. Can't seem to get rid of it.17:50
<dbr>we are speaking about compiling nice with java 1.5, right?17:51
what is the ambiguity error?
<sriramalka>Yes. Compiling with java 1.5. The error is as follows:17:52
~/Software/nice-0.9.10.orig/src/bossa/syntax/constraint.nice: line 100, column 36:
Ambiguity for symbol append. Possibilities are :
java.lang.AbstractStringBuilder append(java.lang.AbstractStringBuilder, ?Object)
java.lang.Appendable append(java.lang.Appendable, nice.lang.char)
This particular error ought to be avoidable in the typing system because AbstractStringBuilder implements Appendable and hence is more specific.17:53
<dbr>yes, but char is more specific than ?Object17:55
so neither method is more specific than the other
<sriramalka>Any way to cast my way out of this?17:56
<dbr>i guess the second one should be called, right? to avoid the boxing
what about:
Appendable a = theBuilder;17:57
and the use a
that makes the first one unapplicable
<sriramalka>I believe I tried that, but I forget all the combinations I tried. I'll attempt it again.
<dbr>ah, maybe not because of type inference
the compiler is too smart ;-)
you could always call a method that return an Appendable (the same as its argument)17:59
convoluted but it should work
<sriramalka>Yes, I see the code. That did fix it, but of course there was a similar problem elsewhere. I guess I'll just use Appendable wherever I get this error and see if the tests work.18:03
<dbr>cool, thanks
that's indeed a workaround instead of a true solution, but it would be good to have for now18:04
you can leave TODO markers to remind us to fix it properly
<sriramalka>Just to introduce myself. I am Sriram Srinivasan, currently doing a PhD at Univ. of Cambridge after 18 years of industry experience. Have developed tons of middleware frameworks including WebLogic. I was looking for a better type system than java to base my ideas on (concurrency related ideas, not type system) and found scala very different from the java mindset. Hence Nice. The language looks very nice indeed.18:09
<dbr>great! nice to meet you :-)18:10
i have to go, but i'll be back on irc18:33
or you can contact me by email18:34
bonniot@users.sf.net
* dbr leaves18:35
* sriramalka leaves19:53
* sriramalka joins21:00
* CIA-2 leaves21:44
* CIA-16 joins21:49
* sriramalka leaves21:56