Hi All,

test18.csd seems to go into an infinite loop, test19.csd hangs,
test20.csd is broken, and test21.csd hangs.  Test22 works though.  I'm
pretty sure those tests previously worked. I'm feeling pretty
jet-lagged at the moment, but will try to take a look tonight or
tomorrow.  In general, if anyone working with the new parser can run
the test suite before commiting changes so we know for sure new work
doesn't break anything, that'd be great.

Thanks!
steven

------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Does this mean anything to you? (CSD attached)
==================================================
csound_orcerror: syntax error, unexpected T_IDENT_K, expecting S_LB
Parsing failed due to invalid input!
AST - INITIAL
Printing Tree




Verifying AST (NEED TO IMPLEMENT)
[Begin Expanding Expressions in AST]
[End Expanding Expressions in AST]
AST - AFTER EXPANSION
Printing Tree




Optimizing AST
[End Optimizing AST]
create_opcode: No rule to handle statemnent of type 0
Printing Tree




Unknown TREE node of type 0 found in root.
Printing Tree




error:  perf-pass statements illegal in header blk


----- Original Message ----- 
From: "Steven Yi" 
To: "Developer discussions" 
Sent: Wednesday, April 22, 2009 1:13 AM
Subject: [Cs-dev] Some tests broken


> Hi All,
>
> test18.csd seems to go into an infinite loop, test19.csd hangs,
> test20.csd is broken, and test21.csd hangs.  Test22 works though.  I'm
> pretty sure those tests previously worked. I'm feeling pretty
> jet-lagged at the moment, but will try to take a look tonight or
> tomorrow.  In general, if anyone working with the new parser can run
> the test suite before commiting changes so we know for sure new work
> doesn't break anything, that'd be great.
>
> Thanks!
> steven
>
> ------------------------------------------------------------------------------
> Stay on top of everything new and different, both inside and
> around Java (TM) technology - register by April 22, and save
> $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
> 300 plus technical and hands-on sessions. Register today.
> Use priority code J9JMT32. http://p.sf.net/sfu/p
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel 

I think I have fixed the string problem -- it runs test26 now.
I too am seeing problems with test18 -- it did run once, but usually loops.

Need to do 0xfffb polymorphic type

==John ff


------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

This type of statement

if kform1 <= kfun then
kform1 = kfun
endif

is broken. That is what is breaking some of the code I am testing
Victor

----- Original Message ----- 
From: 
To: "Developer discussions" 
Sent: Wednesday, April 22, 2009 8:42 AM
Subject: Re: [Cs-dev] Some tests broken


>I think I have fixed the string problem -- it runs test26 now.
> I too am seeing problems with test18 -- it did run once, but usually 
> loops.
>
> Need to do 0xfffb polymorphic type
>
> ==John ff
>
>
> ------------------------------------------------------------------------------
> Stay on top of everything new and different, both inside and
> around Java (TM) technology - register by April 22, and save
> $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
> 300 plus technical and hands-on sessions. Register today.
> Use priority code J9JMT32. http://p.sf.net/sfu/p
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel 


------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

and labels too.

If I have a label , eg.

next:

It breaks the compilation; even if there are no goto statements.

Victor
----- Original Message ----- 
From: "victor" 
To: "Developer discussions" 
Sent: Wednesday, April 22, 2009 8:47 AM
Subject: Re: [Cs-dev] Some tests broken


> This type of statement
>
> if kform1 <= kfun then
> kform1 = kfun
> endif
>
> is broken. That is what is breaking some of the code I am testing
> Victor
>
> ----- Original Message ----- 
> From: 
> To: "Developer discussions" 
> Sent: Wednesday, April 22, 2009 8:42 AM
> Subject: Re: [Cs-dev] Some tests broken
>
>
>>I think I have fixed the string problem -- it runs test26 now.
>> I too am seeing problems with test18 -- it did run once, but usually
>> loops.
>>
>> Need to do 0xfffb polymorphic type
>>
>> ==John ff
>>
>>
>> ------------------------------------------------------------------------------
>> Stay on top of everything new and different, both inside and
>> around Java (TM) technology - register by April 22, and save
>> $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
>> 300 plus technical and hands-on sessions. Register today.
>> Use priority code J9JMT32. http://p.sf.net/sfu/p
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
> ------------------------------------------------------------------------------
> Stay on top of everything new and different, both inside and
> around Java (TM) technology - register by April 22, and save
> $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
> 300 plus technical and hands-on sessions. Register today.
> Use priority code J9JMT32. http://p.sf.net/sfu/p
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel 


------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Hi Victor,

I took a look at this one and I think what's breaking is what you
mentioned, having to do with the if statement that looks like this:

if ktrig == 1 then

Currently, the parser requires () around the if statement's condition, so:

if (ktrig ==1 ) then

would work. I just tested by adding those and indeed choir.csd
compiled then.  The grammar has it as required to have the parentheses
(it's in the rule for "statement" and "ifthen", search for "T_IF
S_LB").

I'm still looking at the general if-then problem (the code Iooked at
in csound_orc_expression.c was actually fine, just the tool I was
using made the diff look strange and I was looking different code than
I thought I was).

John, if you have a chance to fiddle with that that'd be great,
otherwise we can put it on the list of things to do.

Thanks Victor for reporting this!
steven

On Wed, Apr 22, 2009 at 12:14 AM, victor  wrote:
> Does this mean anything to you? (CSD attached)
> ==================================================
> csound_orcerror: syntax error, unexpected T_IDENT_K, expecting S_LB
> Parsing failed due to invalid input!
> AST - INITIAL
> Printing Tree
> 
> 
> 
> 
> Verifying AST (NEED TO IMPLEMENT)
> [Begin Expanding Expressions in AST]
> [End Expanding Expressions in AST]
> AST - AFTER EXPANSION
> Printing Tree
> 
> 
> 
> 
> Optimizing AST
> [End Optimizing AST]
> create_opcode: No rule to handle statemnent of type 0
> Printing Tree
> 
> 
> 
> 
> Unknown TREE node of type 0 found in root.
> Printing Tree
> 
> 
> 
> 
> error:  perf-pass statements illegal in header blk
>
>
> ----- Original Message ----- From: "Steven Yi" 
> To: "Developer discussions" 
> Sent: Wednesday, April 22, 2009 1:13 AM
> Subject: [Cs-dev] Some tests broken
>
>
>> Hi All,
>>
>> test18.csd seems to go into an infinite loop, test19.csd hangs,
>> test20.csd is broken, and test21.csd hangs.  Test22 works though.  I'm
>> pretty sure those tests previously worked. I'm feeling pretty
>> jet-lagged at the moment, but will try to take a look tonight or
>> tomorrow.  In general, if anyone working with the new parser can run
>> the test suite before commiting changes so we know for sure new work
>> doesn't break anything, that'd be great.
>>
>> Thanks!
>> steven
>>
>>
>> ------------------------------------------------------------------------------
>> Stay on top of everything new and different, both inside and
>> around Java (TM) technology - register by April 22, and save
>> $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
>> 300 plus technical and hands-on sessions. Register today.
>> Use priority code J9JMT32. http://p.sf.net/sfu/p
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> Stay on top of everything new and different, both inside and
> around Java (TM) technology - register by April 22, and save
> $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
> 300 plus technical and hands-on sessions. Register today.
> Use priority code J9JMT32. http://p.sf.net/sfu/p
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>

------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net