Hi All,

Just wanted to report the latest. I've just managed to get parser3 to
compile and run trapped.csd.  I'm now going through the csdtests
(currently 53 success, 30 fail).  I think the fails are mostly the
result of a few bugs that should get fixed quickly.  I'm hoping that
I'll be able to push up to github within the next couple of days for
more general testing.

To note, I had to temporarily remove the parallel markup code from the
parser.  The plan is to restore that as part of a tree pass that
happens after the semantic verification step.  I think this should
actually improve the markup compared to doing it earlier in the parser
as it will take into account any changes due to tree rewrites.

I'll report back once getting through the csdtests.

Thanks!
steven

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Good job!

2014-09-08 21:43 GMT+02:00 Steven Yi :
> Hi All,
>
> Just wanted to report the latest. I've just managed to get parser3 to
> compile and run trapped.csd.  I'm now going through the csdtests
> (currently 53 success, 30 fail).  I think the fails are mostly the
> result of a few bugs that should get fixed quickly.  I'm hoping that
> I'll be able to push up to github within the next couple of days for
> more general testing.
>
> To note, I had to temporarily remove the parallel markup code from the
> parser.  The plan is to restore that as part of a tree pass that
> happens after the semantic verification step.  I think this should
> actually improve the markup compared to doing it earlier in the parser
> as it will take into account any changes due to tree rewrites.
>
> I'll report back once getting through the csdtests.
>
> Thanks!
> steven
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-- 

Oeyvind Brandtsegg
Professor of Music Technology
NTNU
7491 Trondheim
Norway
Cell: +47 92 203 205

http://flyndresang.no/
http://www.partikkelaudio.com/
http://soundcloud.com/brandtsegg
http://soundcloud.com/t-emp

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Thanks Oeyvind! I'm down to 13 fails now.  Hope no tricky ones pop up
tomorrow. :)

On Mon, Sep 8, 2014 at 9:10 PM, Oeyvind Brandtsegg
 wrote:
> Good job!
>
> 2014-09-08 21:43 GMT+02:00 Steven Yi :
>> Hi All,
>>
>> Just wanted to report the latest. I've just managed to get parser3 to
>> compile and run trapped.csd.  I'm now going through the csdtests
>> (currently 53 success, 30 fail).  I think the fails are mostly the
>> result of a few bugs that should get fixed quickly.  I'm hoping that
>> I'll be able to push up to github within the next couple of days for
>> more general testing.
>>
>> To note, I had to temporarily remove the parallel markup code from the
>> parser.  The plan is to restore that as part of a tree pass that
>> happens after the semantic verification step.  I think this should
>> actually improve the markup compared to doing it earlier in the parser
>> as it will take into account any changes due to tree rewrites.
>>
>> I'll report back once getting through the csdtests.
>>
>> Thanks!
>> steven
>>
>> ------------------------------------------------------------------------------
>> Want excitement?
>> Manually upgrade your production database.
>> When you want reliability, choose Perforce
>> Perforce version control. Predictably reliable.
>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>
> --
>
> Oeyvind Brandtsegg
> Professor of Music Technology
> NTNU
> 7491 Trondheim
> Norway
> Cell: +47 92 203 205
>
> http://flyndresang.no/
> http://www.partikkelaudio.com/
> http://soundcloud.com/brandtsegg
> http://soundcloud.com/t-emp
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Update: I've pushed the latest code for parser3 up to Github.  It
passes all 83 tests here, plus the regression and unit tests.  Some
notes:

* There are some thing that the previous parser enforced that the
latest code does not, due to the changes in design.  Primarily, the
use of reserved variables (i.e. sr, ksmps, kr) is not prevented during
parse time.  It needs to be checked during semantic analysis time, but
I have not yet looked at that.

* Markup for parallel has been removed from the parser. I will be
putting that into a separate markup phase. I had started the markup
phase a while back (in cs_par_orc_semantic_analysis.c) and plan to
complete that and put in the markup stuff that was previously in the
parser.

* With parser3, we should be able to define variables with the same
name as opcodes.  That has been a long running concern if a developer
creates a new opcode, it could potentially clash with a variable name
used by a user a long time ago.  It might already work, but I have to
double check and add tests for this.

* Typed variables (i.e. mySignal:a) should be parsing now, but the
code for doing lookup has not yet been implemented.

I'll be working on these various parts over the next few days.  If
anyone has a chance to pull the branch and can test it with various
project you might have there, I'd appreciate the feedback.

My plan is that once the above is complete to merge back into develop.
(Note, I'm about to do a merge the latest from develop into the
parser3 branch so that it has the latest fixes there too.)

Thanks!
steven

On Mon, Sep 8, 2014 at 11:24 PM, Steven Yi  wrote:
> Thanks Oeyvind! I'm down to 13 fails now.  Hope no tricky ones pop up
> tomorrow. :)
>
> On Mon, Sep 8, 2014 at 9:10 PM, Oeyvind Brandtsegg
>  wrote:
>> Good job!
>>
>> 2014-09-08 21:43 GMT+02:00 Steven Yi :
>>> Hi All,
>>>
>>> Just wanted to report the latest. I've just managed to get parser3 to
>>> compile and run trapped.csd.  I'm now going through the csdtests
>>> (currently 53 success, 30 fail).  I think the fails are mostly the
>>> result of a few bugs that should get fixed quickly.  I'm hoping that
>>> I'll be able to push up to github within the next couple of days for
>>> more general testing.
>>>
>>> To note, I had to temporarily remove the parallel markup code from the
>>> parser.  The plan is to restore that as part of a tree pass that
>>> happens after the semantic verification step.  I think this should
>>> actually improve the markup compared to doing it earlier in the parser
>>> as it will take into account any changes due to tree rewrites.
>>>
>>> I'll report back once getting through the csdtests.
>>>
>>> Thanks!
>>> steven
>>>
>>> ------------------------------------------------------------------------------
>>> Want excitement?
>>> Manually upgrade your production database.
>>> When you want reliability, choose Perforce
>>> Perforce version control. Predictably reliable.
>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>>
>>
>> --
>>
>> Oeyvind Brandtsegg
>> Professor of Music Technology
>> NTNU
>> 7491 Trondheim
>> Norway
>> Cell: +47 92 203 205
>>
>> http://flyndresang.no/
>> http://www.partikkelaudio.com/
>> http://soundcloud.com/brandtsegg
>> http://soundcloud.com/t-emp
>>
>> ------------------------------------------------------------------------------
>> Want excitement?
>> Manually upgrade your production database.
>> When you want reliability, choose Perforce
>> Perforce version control. Predictably reliable.
>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce.
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Sounds great Steven. I'll try it out if I get a chance, but I'm up to
my eyes with stuff right now.

On 10 September 2014 00:06, Steven Yi  wrote:
> Update: I've pushed the latest code for parser3 up to Github.  It
> passes all 83 tests here, plus the regression and unit tests.  Some
> notes:
>
> * There are some thing that the previous parser enforced that the
> latest code does not, due to the changes in design.  Primarily, the
> use of reserved variables (i.e. sr, ksmps, kr) is not prevented during
> parse time.  It needs to be checked during semantic analysis time, but
> I have not yet looked at that.
>
> * Markup for parallel has been removed from the parser. I will be
> putting that into a separate markup phase. I had started the markup
> phase a while back (in cs_par_orc_semantic_analysis.c) and plan to
> complete that and put in the markup stuff that was previously in the
> parser.
>
> * With parser3, we should be able to define variables with the same
> name as opcodes.  That has been a long running concern if a developer
> creates a new opcode, it could potentially clash with a variable name
> used by a user a long time ago.  It might already work, but I have to
> double check and add tests for this.
>
> * Typed variables (i.e. mySignal:a) should be parsing now, but the
> code for doing lookup has not yet been implemented.
>
> I'll be working on these various parts over the next few days.  If
> anyone has a chance to pull the branch and can test it with various
> project you might have there, I'd appreciate the feedback.
>
> My plan is that once the above is complete to merge back into develop.
> (Note, I'm about to do a merge the latest from develop into the
> parser3 branch so that it has the latest fixes there too.)
>
> Thanks!
> steven
>
> On Mon, Sep 8, 2014 at 11:24 PM, Steven Yi  wrote:
>> Thanks Oeyvind! I'm down to 13 fails now.  Hope no tricky ones pop up
>> tomorrow. :)
>>
>> On Mon, Sep 8, 2014 at 9:10 PM, Oeyvind Brandtsegg
>>  wrote:
>>> Good job!
>>>
>>> 2014-09-08 21:43 GMT+02:00 Steven Yi :
>>>> Hi All,
>>>>
>>>> Just wanted to report the latest. I've just managed to get parser3 to
>>>> compile and run trapped.csd.  I'm now going through the csdtests
>>>> (currently 53 success, 30 fail).  I think the fails are mostly the
>>>> result of a few bugs that should get fixed quickly.  I'm hoping that
>>>> I'll be able to push up to github within the next couple of days for
>>>> more general testing.
>>>>
>>>> To note, I had to temporarily remove the parallel markup code from the
>>>> parser.  The plan is to restore that as part of a tree pass that
>>>> happens after the semantic verification step.  I think this should
>>>> actually improve the markup compared to doing it earlier in the parser
>>>> as it will take into account any changes due to tree rewrites.
>>>>
>>>> I'll report back once getting through the csdtests.
>>>>
>>>> Thanks!
>>>> steven
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Want excitement?
>>>> Manually upgrade your production database.
>>>> When you want reliability, choose Perforce
>>>> Perforce version control. Predictably reliable.
>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>>
>>>
>>> --
>>>
>>> Oeyvind Brandtsegg
>>> Professor of Music Technology
>>> NTNU
>>> 7491 Trondheim
>>> Norway
>>> Cell: +47 92 203 205
>>>
>>> http://flyndresang.no/
>>> http://www.partikkelaudio.com/
>>> http://soundcloud.com/brandtsegg
>>> http://soundcloud.com/t-emp
>>>
>>> ------------------------------------------------------------------------------
>>> Want excitement?
>>> Manually upgrade your production database.
>>> When you want reliability, choose Perforce
>>> Perforce version control. Predictably reliable.
>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce.
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce.
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Thanks Rory!

I already found one issue, which is that the i() check for not working
with expressions doesn't work as it's looking for a LABEL_TOKEN, which
I changed.  Also, that made me remember that there is an issue with
labels too I had forgotten to restore.  Labels can also be keywords,
such as else: or if: .  The current parser (I'll call it parser2 from
here on out) had a rule to catch this, but it was a bit permissive and
bubbled up through the parser.  I think I had come up with a way to
localize that to just after goto, kgoto, and igoto, but I had
forgotten to revisit that.  (will do tomorrow).   So those issues at
least are marked as to be taken care of.

Thanks!
steven

On Tue, Sep 9, 2014 at 11:08 PM, Rory Walsh  wrote:
> Sounds great Steven. I'll try it out if I get a chance, but I'm up to
> my eyes with stuff right now.
>
> On 10 September 2014 00:06, Steven Yi  wrote:
>> Update: I've pushed the latest code for parser3 up to Github.  It
>> passes all 83 tests here, plus the regression and unit tests.  Some
>> notes:
>>
>> * There are some thing that the previous parser enforced that the
>> latest code does not, due to the changes in design.  Primarily, the
>> use of reserved variables (i.e. sr, ksmps, kr) is not prevented during
>> parse time.  It needs to be checked during semantic analysis time, but
>> I have not yet looked at that.
>>
>> * Markup for parallel has been removed from the parser. I will be
>> putting that into a separate markup phase. I had started the markup
>> phase a while back (in cs_par_orc_semantic_analysis.c) and plan to
>> complete that and put in the markup stuff that was previously in the
>> parser.
>>
>> * With parser3, we should be able to define variables with the same
>> name as opcodes.  That has been a long running concern if a developer
>> creates a new opcode, it could potentially clash with a variable name
>> used by a user a long time ago.  It might already work, but I have to
>> double check and add tests for this.
>>
>> * Typed variables (i.e. mySignal:a) should be parsing now, but the
>> code for doing lookup has not yet been implemented.
>>
>> I'll be working on these various parts over the next few days.  If
>> anyone has a chance to pull the branch and can test it with various
>> project you might have there, I'd appreciate the feedback.
>>
>> My plan is that once the above is complete to merge back into develop.
>> (Note, I'm about to do a merge the latest from develop into the
>> parser3 branch so that it has the latest fixes there too.)
>>
>> Thanks!
>> steven
>>
>> On Mon, Sep 8, 2014 at 11:24 PM, Steven Yi  wrote:
>>> Thanks Oeyvind! I'm down to 13 fails now.  Hope no tricky ones pop up
>>> tomorrow. :)
>>>
>>> On Mon, Sep 8, 2014 at 9:10 PM, Oeyvind Brandtsegg
>>>  wrote:
>>>> Good job!
>>>>
>>>> 2014-09-08 21:43 GMT+02:00 Steven Yi :
>>>>> Hi All,
>>>>>
>>>>> Just wanted to report the latest. I've just managed to get parser3 to
>>>>> compile and run trapped.csd.  I'm now going through the csdtests
>>>>> (currently 53 success, 30 fail).  I think the fails are mostly the
>>>>> result of a few bugs that should get fixed quickly.  I'm hoping that
>>>>> I'll be able to push up to github within the next couple of days for
>>>>> more general testing.
>>>>>
>>>>> To note, I had to temporarily remove the parallel markup code from the
>>>>> parser.  The plan is to restore that as part of a tree pass that
>>>>> happens after the semantic verification step.  I think this should
>>>>> actually improve the markup compared to doing it earlier in the parser
>>>>> as it will take into account any changes due to tree rewrites.
>>>>>
>>>>> I'll report back once getting through the csdtests.
>>>>>
>>>>> Thanks!
>>>>> steven
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> Want excitement?
>>>>> Manually upgrade your production database.
>>>>> When you want reliability, choose Perforce
>>>>> Perforce version control. Predictably reliable.
>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>>> _______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>
>>>>
>>>>
>>>> --
>>>>
>>>> Oeyvind Brandtsegg
>>>> Professor of Music Technology
>>>> NTNU
>>>> 7491 Trondheim
>>>> Norway
>>>> Cell: +47 92 203 205
>>>>
>>>> http://flyndresang.no/
>>>> http://www.partikkelaudio.com/
>>>> http://soundcloud.com/brandtsegg
>>>> http://soundcloud.com/t-emp
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Want excitement?
>>>> Manually upgrade your production database.
>>>> When you want reliability, choose Perforce
>>>> Perforce version control. Predictably reliable.
>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> Want excitement?
>> Manually upgrade your production database.
>> When you want reliability, choose Perforce.
>> Perforce version control. Predictably reliable.
>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce.
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce.
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

One thing looks like it's working here, which is that variables can
have the same name as opcodes:



;
;
; ==============================================


sr    =    44100
ksmps    =    1
nchnls    =    2
0dbfs    =    1

instr 1

ampdb = vco2(0.1, 440)
outs ampdb, ampdb

endin


; ==============================================


i1 0 0.2




On Tue, Sep 9, 2014 at 11:44 PM, Steven Yi  wrote:
> Thanks Rory!
>
> I already found one issue, which is that the i() check for not working
> with expressions doesn't work as it's looking for a LABEL_TOKEN, which
> I changed.  Also, that made me remember that there is an issue with
> labels too I had forgotten to restore.  Labels can also be keywords,
> such as else: or if: .  The current parser (I'll call it parser2 from
> here on out) had a rule to catch this, but it was a bit permissive and
> bubbled up through the parser.  I think I had come up with a way to
> localize that to just after goto, kgoto, and igoto, but I had
> forgotten to revisit that.  (will do tomorrow).   So those issues at
> least are marked as to be taken care of.
>
> Thanks!
> steven
>
> On Tue, Sep 9, 2014 at 11:08 PM, Rory Walsh  wrote:
>> Sounds great Steven. I'll try it out if I get a chance, but I'm up to
>> my eyes with stuff right now.
>>
>> On 10 September 2014 00:06, Steven Yi  wrote:
>>> Update: I've pushed the latest code for parser3 up to Github.  It
>>> passes all 83 tests here, plus the regression and unit tests.  Some
>>> notes:
>>>
>>> * There are some thing that the previous parser enforced that the
>>> latest code does not, due to the changes in design.  Primarily, the
>>> use of reserved variables (i.e. sr, ksmps, kr) is not prevented during
>>> parse time.  It needs to be checked during semantic analysis time, but
>>> I have not yet looked at that.
>>>
>>> * Markup for parallel has been removed from the parser. I will be
>>> putting that into a separate markup phase. I had started the markup
>>> phase a while back (in cs_par_orc_semantic_analysis.c) and plan to
>>> complete that and put in the markup stuff that was previously in the
>>> parser.
>>>
>>> * With parser3, we should be able to define variables with the same
>>> name as opcodes.  That has been a long running concern if a developer
>>> creates a new opcode, it could potentially clash with a variable name
>>> used by a user a long time ago.  It might already work, but I have to
>>> double check and add tests for this.
>>>
>>> * Typed variables (i.e. mySignal:a) should be parsing now, but the
>>> code for doing lookup has not yet been implemented.
>>>
>>> I'll be working on these various parts over the next few days.  If
>>> anyone has a chance to pull the branch and can test it with various
>>> project you might have there, I'd appreciate the feedback.
>>>
>>> My plan is that once the above is complete to merge back into develop.
>>> (Note, I'm about to do a merge the latest from develop into the
>>> parser3 branch so that it has the latest fixes there too.)
>>>
>>> Thanks!
>>> steven
>>>
>>> On Mon, Sep 8, 2014 at 11:24 PM, Steven Yi  wrote:
>>>> Thanks Oeyvind! I'm down to 13 fails now.  Hope no tricky ones pop up
>>>> tomorrow. :)
>>>>
>>>> On Mon, Sep 8, 2014 at 9:10 PM, Oeyvind Brandtsegg
>>>>  wrote:
>>>>> Good job!
>>>>>
>>>>> 2014-09-08 21:43 GMT+02:00 Steven Yi :
>>>>>> Hi All,
>>>>>>
>>>>>> Just wanted to report the latest. I've just managed to get parser3 to
>>>>>> compile and run trapped.csd.  I'm now going through the csdtests
>>>>>> (currently 53 success, 30 fail).  I think the fails are mostly the
>>>>>> result of a few bugs that should get fixed quickly.  I'm hoping that
>>>>>> I'll be able to push up to github within the next couple of days for
>>>>>> more general testing.
>>>>>>
>>>>>> To note, I had to temporarily remove the parallel markup code from the
>>>>>> parser.  The plan is to restore that as part of a tree pass that
>>>>>> happens after the semantic verification step.  I think this should
>>>>>> actually improve the markup compared to doing it earlier in the parser
>>>>>> as it will take into account any changes due to tree rewrites.
>>>>>>
>>>>>> I'll report back once getting through the csdtests.
>>>>>>
>>>>>> Thanks!
>>>>>> steven
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> Want excitement?
>>>>>> Manually upgrade your production database.
>>>>>> When you want reliability, choose Perforce
>>>>>> Perforce version control. Predictably reliable.
>>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>>>> _______________________________________________
>>>>>> Csound-devel mailing list
>>>>>> Csound-devel@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>>
>>>>> Oeyvind Brandtsegg
>>>>> Professor of Music Technology
>>>>> NTNU
>>>>> 7491 Trondheim
>>>>> Norway
>>>>> Cell: +47 92 203 205
>>>>>
>>>>> http://flyndresang.no/
>>>>> http://www.partikkelaudio.com/
>>>>> http://soundcloud.com/brandtsegg
>>>>> http://soundcloud.com/t-emp
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> Want excitement?
>>>>> Manually upgrade your production database.
>>>>> When you want reliability, choose Perforce
>>>>> Perforce version control. Predictably reliable.
>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>>> _______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>> ------------------------------------------------------------------------------
>>> Want excitement?
>>> Manually upgrade your production database.
>>> When you want reliability, choose Perforce.
>>> Perforce version control. Predictably reliable.
>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> Want excitement?
>> Manually upgrade your production database.
>> When you want reliability, choose Perforce.
>> Perforce version control. Predictably reliable.
>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Syntax highlighting just got a little more interesting :)

On 10 September 2014 11:27, Steven Yi  wrote:
> One thing looks like it's working here, which is that variables can
> have the same name as opcodes:
>
>
> 
> ;
> ;
> ; ==============================================
> 
>
> sr    =    44100
> ksmps    =    1
> nchnls    =    2
> 0dbfs    =    1
>
> instr 1
>
> ampdb = vco2(0.1, 440)
> outs ampdb, ampdb
>
> endin
>
> 
> ; ==============================================
> 
>
> i1 0 0.2
>
> 
> 
>
> On Tue, Sep 9, 2014 at 11:44 PM, Steven Yi  wrote:
>> Thanks Rory!
>>
>> I already found one issue, which is that the i() check for not working
>> with expressions doesn't work as it's looking for a LABEL_TOKEN, which
>> I changed.  Also, that made me remember that there is an issue with
>> labels too I had forgotten to restore.  Labels can also be keywords,
>> such as else: or if: .  The current parser (I'll call it parser2 from
>> here on out) had a rule to catch this, but it was a bit permissive and
>> bubbled up through the parser.  I think I had come up with a way to
>> localize that to just after goto, kgoto, and igoto, but I had
>> forgotten to revisit that.  (will do tomorrow).   So those issues at
>> least are marked as to be taken care of.
>>
>> Thanks!
>> steven
>>
>> On Tue, Sep 9, 2014 at 11:08 PM, Rory Walsh  wrote:
>>> Sounds great Steven. I'll try it out if I get a chance, but I'm up to
>>> my eyes with stuff right now.
>>>
>>> On 10 September 2014 00:06, Steven Yi  wrote:
>>>> Update: I've pushed the latest code for parser3 up to Github.  It
>>>> passes all 83 tests here, plus the regression and unit tests.  Some
>>>> notes:
>>>>
>>>> * There are some thing that the previous parser enforced that the
>>>> latest code does not, due to the changes in design.  Primarily, the
>>>> use of reserved variables (i.e. sr, ksmps, kr) is not prevented during
>>>> parse time.  It needs to be checked during semantic analysis time, but
>>>> I have not yet looked at that.
>>>>
>>>> * Markup for parallel has been removed from the parser. I will be
>>>> putting that into a separate markup phase. I had started the markup
>>>> phase a while back (in cs_par_orc_semantic_analysis.c) and plan to
>>>> complete that and put in the markup stuff that was previously in the
>>>> parser.
>>>>
>>>> * With parser3, we should be able to define variables with the same
>>>> name as opcodes.  That has been a long running concern if a developer
>>>> creates a new opcode, it could potentially clash with a variable name
>>>> used by a user a long time ago.  It might already work, but I have to
>>>> double check and add tests for this.
>>>>
>>>> * Typed variables (i.e. mySignal:a) should be parsing now, but the
>>>> code for doing lookup has not yet been implemented.
>>>>
>>>> I'll be working on these various parts over the next few days.  If
>>>> anyone has a chance to pull the branch and can test it with various
>>>> project you might have there, I'd appreciate the feedback.
>>>>
>>>> My plan is that once the above is complete to merge back into develop.
>>>> (Note, I'm about to do a merge the latest from develop into the
>>>> parser3 branch so that it has the latest fixes there too.)
>>>>
>>>> Thanks!
>>>> steven
>>>>
>>>> On Mon, Sep 8, 2014 at 11:24 PM, Steven Yi  wrote:
>>>>> Thanks Oeyvind! I'm down to 13 fails now.  Hope no tricky ones pop up
>>>>> tomorrow. :)
>>>>>
>>>>> On Mon, Sep 8, 2014 at 9:10 PM, Oeyvind Brandtsegg
>>>>>  wrote:
>>>>>> Good job!
>>>>>>
>>>>>> 2014-09-08 21:43 GMT+02:00 Steven Yi :
>>>>>>> Hi All,
>>>>>>>
>>>>>>> Just wanted to report the latest. I've just managed to get parser3 to
>>>>>>> compile and run trapped.csd.  I'm now going through the csdtests
>>>>>>> (currently 53 success, 30 fail).  I think the fails are mostly the
>>>>>>> result of a few bugs that should get fixed quickly.  I'm hoping that
>>>>>>> I'll be able to push up to github within the next couple of days for
>>>>>>> more general testing.
>>>>>>>
>>>>>>> To note, I had to temporarily remove the parallel markup code from the
>>>>>>> parser.  The plan is to restore that as part of a tree pass that
>>>>>>> happens after the semantic verification step.  I think this should
>>>>>>> actually improve the markup compared to doing it earlier in the parser
>>>>>>> as it will take into account any changes due to tree rewrites.
>>>>>>>
>>>>>>> I'll report back once getting through the csdtests.
>>>>>>>
>>>>>>> Thanks!
>>>>>>> steven
>>>>>>>
>>>>>>> ------------------------------------------------------------------------------
>>>>>>> Want excitement?
>>>>>>> Manually upgrade your production database.
>>>>>>> When you want reliability, choose Perforce
>>>>>>> Perforce version control. Predictably reliable.
>>>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>>>>> _______________________________________________
>>>>>>> Csound-devel mailing list
>>>>>>> Csound-devel@lists.sourceforge.net
>>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>>
>>>>>> Oeyvind Brandtsegg
>>>>>> Professor of Music Technology
>>>>>> NTNU
>>>>>> 7491 Trondheim
>>>>>> Norway
>>>>>> Cell: +47 92 203 205
>>>>>>
>>>>>> http://flyndresang.no/
>>>>>> http://www.partikkelaudio.com/
>>>>>> http://soundcloud.com/brandtsegg
>>>>>> http://soundcloud.com/t-emp
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> Want excitement?
>>>>>> Manually upgrade your production database.
>>>>>> When you want reliability, choose Perforce
>>>>>> Perforce version control. Predictably reliable.
>>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>>>> _______________________________________________
>>>>>> Csound-devel mailing list
>>>>>> Csound-devel@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Want excitement?
>>>> Manually upgrade your production database.
>>>> When you want reliability, choose Perforce.
>>>> Perforce version control. Predictably reliable.
>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>> ------------------------------------------------------------------------------
>>> Want excitement?
>>> Manually upgrade your production database.
>>> When you want reliability, choose Perforce.
>>> Perforce version control. Predictably reliable.
>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Well, it should still be advised not to use the same name for a var as
an opcode, but it's not a requirement now.  With this we're
future-proofing the system a bit better.  I'm actually a little
worried once we can use opcodes as arguments, such as:

map(oscil, kamps, kfreqs)

O_O





On Wed, Sep 10, 2014 at 10:32 AM, Rory Walsh  wrote:
> Syntax highlighting just got a little more interesting :)
>
> On 10 September 2014 11:27, Steven Yi  wrote:
>> One thing looks like it's working here, which is that variables can
>> have the same name as opcodes:
>>
>>
>> 
>> ;
>> ;
>> ; ==============================================
>> 
>>
>> sr    =    44100
>> ksmps    =    1
>> nchnls    =    2
>> 0dbfs    =    1
>>
>> instr 1
>>
>> ampdb = vco2(0.1, 440)
>> outs ampdb, ampdb
>>
>> endin
>>
>> 
>> ; ==============================================
>> 
>>
>> i1 0 0.2
>>
>> 
>> 
>>
>> On Tue, Sep 9, 2014 at 11:44 PM, Steven Yi  wrote:
>>> Thanks Rory!
>>>
>>> I already found one issue, which is that the i() check for not working
>>> with expressions doesn't work as it's looking for a LABEL_TOKEN, which
>>> I changed.  Also, that made me remember that there is an issue with
>>> labels too I had forgotten to restore.  Labels can also be keywords,
>>> such as else: or if: .  The current parser (I'll call it parser2 from
>>> here on out) had a rule to catch this, but it was a bit permissive and
>>> bubbled up through the parser.  I think I had come up with a way to
>>> localize that to just after goto, kgoto, and igoto, but I had
>>> forgotten to revisit that.  (will do tomorrow).   So those issues at
>>> least are marked as to be taken care of.
>>>
>>> Thanks!
>>> steven
>>>
>>> On Tue, Sep 9, 2014 at 11:08 PM, Rory Walsh  wrote:
>>>> Sounds great Steven. I'll try it out if I get a chance, but I'm up to
>>>> my eyes with stuff right now.
>>>>
>>>> On 10 September 2014 00:06, Steven Yi  wrote:
>>>>> Update: I've pushed the latest code for parser3 up to Github.  It
>>>>> passes all 83 tests here, plus the regression and unit tests.  Some
>>>>> notes:
>>>>>
>>>>> * There are some thing that the previous parser enforced that the
>>>>> latest code does not, due to the changes in design.  Primarily, the
>>>>> use of reserved variables (i.e. sr, ksmps, kr) is not prevented during
>>>>> parse time.  It needs to be checked during semantic analysis time, but
>>>>> I have not yet looked at that.
>>>>>
>>>>> * Markup for parallel has been removed from the parser. I will be
>>>>> putting that into a separate markup phase. I had started the markup
>>>>> phase a while back (in cs_par_orc_semantic_analysis.c) and plan to
>>>>> complete that and put in the markup stuff that was previously in the
>>>>> parser.
>>>>>
>>>>> * With parser3, we should be able to define variables with the same
>>>>> name as opcodes.  That has been a long running concern if a developer
>>>>> creates a new opcode, it could potentially clash with a variable name
>>>>> used by a user a long time ago.  It might already work, but I have to
>>>>> double check and add tests for this.
>>>>>
>>>>> * Typed variables (i.e. mySignal:a) should be parsing now, but the
>>>>> code for doing lookup has not yet been implemented.
>>>>>
>>>>> I'll be working on these various parts over the next few days.  If
>>>>> anyone has a chance to pull the branch and can test it with various
>>>>> project you might have there, I'd appreciate the feedback.
>>>>>
>>>>> My plan is that once the above is complete to merge back into develop.
>>>>> (Note, I'm about to do a merge the latest from develop into the
>>>>> parser3 branch so that it has the latest fixes there too.)
>>>>>
>>>>> Thanks!
>>>>> steven
>>>>>
>>>>> On Mon, Sep 8, 2014 at 11:24 PM, Steven Yi  wrote:
>>>>>> Thanks Oeyvind! I'm down to 13 fails now.  Hope no tricky ones pop up
>>>>>> tomorrow. :)
>>>>>>
>>>>>> On Mon, Sep 8, 2014 at 9:10 PM, Oeyvind Brandtsegg
>>>>>>  wrote:
>>>>>>> Good job!
>>>>>>>
>>>>>>> 2014-09-08 21:43 GMT+02:00 Steven Yi :
>>>>>>>> Hi All,
>>>>>>>>
>>>>>>>> Just wanted to report the latest. I've just managed to get parser3 to
>>>>>>>> compile and run trapped.csd.  I'm now going through the csdtests
>>>>>>>> (currently 53 success, 30 fail).  I think the fails are mostly the
>>>>>>>> result of a few bugs that should get fixed quickly.  I'm hoping that
>>>>>>>> I'll be able to push up to github within the next couple of days for
>>>>>>>> more general testing.
>>>>>>>>
>>>>>>>> To note, I had to temporarily remove the parallel markup code from the
>>>>>>>> parser.  The plan is to restore that as part of a tree pass that
>>>>>>>> happens after the semantic verification step.  I think this should
>>>>>>>> actually improve the markup compared to doing it earlier in the parser
>>>>>>>> as it will take into account any changes due to tree rewrites.
>>>>>>>>
>>>>>>>> I'll report back once getting through the csdtests.
>>>>>>>>
>>>>>>>> Thanks!
>>>>>>>> steven
>>>>>>>>
>>>>>>>> ------------------------------------------------------------------------------
>>>>>>>> Want excitement?
>>>>>>>> Manually upgrade your production database.
>>>>>>>> When you want reliability, choose Perforce
>>>>>>>> Perforce version control. Predictably reliable.
>>>>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>>>>>> _______________________________________________
>>>>>>>> Csound-devel mailing list
>>>>>>>> Csound-devel@lists.sourceforge.net
>>>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>>
>>>>>>> Oeyvind Brandtsegg
>>>>>>> Professor of Music Technology
>>>>>>> NTNU
>>>>>>> 7491 Trondheim
>>>>>>> Norway
>>>>>>> Cell: +47 92 203 205
>>>>>>>
>>>>>>> http://flyndresang.no/
>>>>>>> http://www.partikkelaudio.com/
>>>>>>> http://soundcloud.com/brandtsegg
>>>>>>> http://soundcloud.com/t-emp
>>>>>>>
>>>>>>> ------------------------------------------------------------------------------
>>>>>>> Want excitement?
>>>>>>> Manually upgrade your production database.
>>>>>>> When you want reliability, choose Perforce
>>>>>>> Perforce version control. Predictably reliable.
>>>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>>>>> _______________________________________________
>>>>>>> Csound-devel mailing list
>>>>>>> Csound-devel@lists.sourceforge.net
>>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> Want excitement?
>>>>> Manually upgrade your production database.
>>>>> When you want reliability, choose Perforce.
>>>>> Perforce version control. Predictably reliable.
>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>>> _______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Want excitement?
>>>> Manually upgrade your production database.
>>>> When you want reliability, choose Perforce.
>>>> Perforce version control. Predictably reliable.
>>>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> Want excitement?
>> Manually upgrade your production database.
>> When you want reliability, choose Perforce
>> Perforce version control. Predictably reliable.
>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net