Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

[Lucas Cordiviola]

> So I assumed it'd be working like that by now.

Is not.

Check line 137 on Dan's dejavu branch:
https://github.com/pure-data/pure-data/blob/dejavu/tcl/pd-gui.tcl


If you want to see how does pdx font metrics looks on your Mac you must edit 
your local “pd-gui.tcl”

change line 137 to:

set font_metrics {
5 11
6 13
7 16
10 19
14 29
22 44
Restart Pd.



Mensaje telepatico asistido por maquinas.



From: Alexandre Torres Porres 
Sent: Saturday, July 22, 2017 3:54 AM
To: Lucas Cordiviola
Cc: pd-list@lists.iem.at
Subject: Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

well, in the PR discussion, back in february, I did ask


will this branch also focus on geting the same metrics as Pd-Extended or not?

And Dan replied:

"Yes it does, it simply doesn't have that change applied yet. I'm not 100% 
convinced the smaller boxes make the most sense overall, maybe there are some 
tweaks needed as a couple patches I have seem too slim where the top of the 
text overlaps the object border. I'll dig that up later.

I'd like to get some feedback from 
@millerpuckette & 
@umlaeute in regard to both font & metrics 
adjustment."


So I assumed it'd be working like that by now.


cheers

2017-07-22 0:39 GMT-03:00 Alexandre Torres Porres 
mailto:por...@gmail.com>>:


2017-07-21 22:10 GMT-03:00 Lucas Cordiviola 
mailto:lucard...@hotmail.com>>:

pdx metrics are not yet hardcoded, there must be a good argument for that to 
became meaningful.

what do you mean?

Support it by just testing and giving opinions. But modify that tcl please.

My opinion is that things should look the same, out of the box, on vanilla, pd 
extended, and purr data. In the verge of bringing cyclone to purr data, I hope 
the same documentation works in both, cause doing two of them would be insane.

Now, is there any need for further testing? Hasn't this been tested already?

cheers



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

[Alexandre Torres Porres]

well, in the PR discussion, back in february, I did ask

will this branch also focus on geting the same metrics as Pd-Extended or
not?

And Dan replied:

"Yes it does, it simply doesn't have that change applied yet. I'm not 100%
convinced the smaller boxes make the most sense overall, maybe there are
some tweaks needed as a couple patches I have seem too slim where the top
of the text overlaps the object border. I'll dig that up later.

I'd like to get some feedback from @millerpuckette
 & @umlaeute
 in regard to both font & metrics adjustment."


So I assumed it'd be working like that by now.


cheers

2017-07-22 0:39 GMT-03:00 Alexandre Torres Porres :

>
>
> 2017-07-21 22:10 GMT-03:00 Lucas Cordiviola :
>
>> pdx metrics are not yet hardcoded, there must be a good argument for that
>> to became meaningful.
>>
>
> what do you mean?
>
>
>> Support it by just testing and giving opinions. But modify that tcl
>> please.
>>
>
> My opinion is that things should look the same, out of the box, on
> vanilla, pd extended, and purr data. In the verge of bringing cyclone to
> purr data, I hope the same documentation works in both, cause doing two of
> them would be insane.
>
> Now, is there any need for further testing? Hasn't this been tested
> already?
>
> cheers
>
>
>
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

[Alexandre Torres Porres]

2017-07-21 22:10 GMT-03:00 Lucas Cordiviola :

> pdx metrics are not yet hardcoded, there must be a good argument for that
> to became meaningful.
>

what do you mean?


> Support it by just testing and giving opinions. But modify that tcl
> please.
>

My opinion is that things should look the same, out of the box, on vanilla,
pd extended, and purr data. In the verge of bringing cyclone to purr data,
I hope the same documentation works in both, cause doing two of them would
be insane.

Now, is there any need for further testing? Hasn't this been tested already?

cheers
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

[Lucas Cordiviola]

One more thing:

@Alex:

All this pdx font metrics thing came from your cyclone help patches not 
rendering the same on diff platforms. While testing that we also came across 
the diff between vanilla and purr-data.

Personally I think is clever that pd patches render the same on all pd's but I 
can perfectly live with vanilla metrics.

So if you are not going to modify your pd-gui.tcl and test you are welcomed to 
redo all your 300 help patches so that *we* can see them nicely on vanilla 
metrics.

Pdx metrics are still not officially supported by Miller, IOhannes or Dan.



>Ok, this is very Deja Vu indeed... let me just ask, we will not have to do any 
>of this when this gets merged, right? So why, to this point, when I build from 
>that branch I still have to do other stuff in order to get what it is supposed 
>to give me in the end? Do you follow me?

See above, pdx metrics are not yet hardcoded, there must be a good argument for 
that to became meaningful.

Support it by just testing and giving opinions. But modify that tcl please.




Mensaje telepatico asistido por maquinas.



From: Alexandre Torres Porres 
Sent: Saturday, July 22, 2017 12:54 AM
To: Lucas Cordiviola
Cc: pd-list@lists.iem.at
Subject: Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

2017-07-21 21:26 GMT-03:00 Lucas Cordiviola 
mailto:lucard...@hotmail.com>>:

Hi Alex,

What you built only changes Monaco for DejaVU. The font metrics in pd-gui.tcl 
are just vanilla.

If you want to test you can modify "pd-gui.tcl" to meet pdx metrics. If you do 
so you most likely get the same as Purr-data.

Just tweak the values on the .tcl file and restart pd.

The pdx values and the how-to are here:

https://github.com/pure-data/pure-data/wiki/Crossplatform-font-metrics-%26-comparisons


Ok, this is very Deja Vu indeed... let me just ask, we will not have to do any 
of this when this gets merged, right? So why, to this point, when I build from 
that branch I still have to do other stuff in order to get what it is supposed 
to give me in the end? Do you follow me?

cheers
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

[Alexandre Torres Porres]

2017-07-21 21:26 GMT-03:00 Lucas Cordiviola :

> Hi Alex,
>
> What you built only changes Monaco for DejaVU. The font metrics in
> pd-gui.tcl are just vanilla.
>
> If you want to test you can modify "pd-gui.tcl" to meet pdx metrics. If
> you do so you most likely get the same as Purr-data.
>
> Just tweak the values on the .tcl file and restart pd.
>
> The pdx values and the how-to are here:
>
> https://github.com/pure-data/pure-data/wiki/Crossplatform-
> font-metrics-%26-comparisons
>


Ok, this is very Deja Vu indeed... let me just ask, we will not have to do
any of this when this gets merged, right? So why, to this point, when I
build from that branch I still have to do other stuff in order to get what
it is supposed to give me in the end? Do you follow me?

cheers
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

[Alexandre Torres Porres]

oh Lucas... see? When I sent the email, I could then see you had written
me...

let me check that now, sorry :)

2017-07-21 21:51 GMT-03:00 Alexandre Torres Porres :

> now, since we're talking about "Dejas Vu", I'm also having a Deja Vu of my
> own... it's not the first time I complain things look different, but then
> someone ends up saying I needed to do something else in order to test it...
> as it happened in this discussion a few months ago
> https://github.com/pure-data/pure-data/pull/42
>
> and I did ask if the idea is to be compatible to old extended (and purr
> data), and I seem to have been able to test, at one point, that it actually
> did and I couldn't be happier.
>
> But I also thought that, at this point, compiling from that branch would
> give me that result out of the box.
>
> So, did I miss something or not?
>
> cheers
>
> 2017-07-21 20:36 GMT-03:00 Alexandre Torres Porres :
>
>> So, i just built from https://github.com/pure-data/pure-data/tree/dejavu
>> and tested here on my mac. It looks good alright, seems to render better
>> and stuff, but it isn't backwards compatible, seems to require more real
>> estate for texts and all (object size and the like) - see attachment.
>>
>> So it doesn't really render the same as in purr data, like you say. That
>> actually looks the same as the current pd state, as well as old pd extended.
>>
>> As much as I like the idea of having a new cross platform bahaviour for
>> pd, I was hoping it'd be the same as purr data and all... also that I
>> wouldn't have to reformat over 300 help patches I'm working on, plus my
>> tutorials which is quite numerous as well - not to mentio regular
>> patches/abstractions.
>>
>> cheers
>>
>>
>>
>> 2017-07-21 18:48 GMT-03:00 Lucas Cordiviola :
>>
>>> @IOhannes
>>>
>>> Here Dan is saying he is becoming to like the font metrics thing:
>>>
>>> https://lists.puredata.info/pipermail/pd-list/2017-07/119495.html
>>>
>>>
>>>
>>>
>>> Mensaje telepatico asistido por maquinas.
>>>
>>>
>>> --
>>> *From:* Pd-list  on behalf of Lucas
>>> Cordiviola 
>>> *Sent:* Friday, July 21, 2017 7:24 PM
>>> *To:* pd-list@lists.iem.at
>>> *Subject:* Re: [PD] [PD-announce] pd 0.48-0test1 released
>>>
>>>
>>>
>>> @IOhannes:
>>>
>>> *>it seems that i remember dan (or was it miller) pointing to the newly*
>>>
>>> *introduced zoom feature which complicates things a bit more. and
>>> dynamic*
>>>
>>> *font managing. and what not. (probably check the archives)*
>>>
>>> *the conclusion for me is: there *are* downsides.*
>>>
>>>
>>> The zoom feature will work with the correct doubles from pdx metrics,
>>> this are the actual doubles of vanilla:
>>>
>>> set font_zoom2_metrics {
>>> 12 20
>>> 14 26
>>> 18 32
>>> 20 42
>>> 30 50
>>> 50 90
>>> }
>>>
>>> Does *downside* to you is having a little more white space?
>>>
>>> Pdx font metrics wont break vanilla patches.
>>>
>>> See:
>>> https://github.com/pure-data/pure-data/wiki/Crossplatform-fo
>>> nt-metrics-%26-comparisons
>>>
>>>
>>>
>>> Mensaje telepatico asistido por maquinas.
>>>
>>>
>>> --
>>> *From:* Pd-list  on behalf of IOhannes m
>>> zmoelnig 
>>> *Sent:* Friday, July 21, 2017 7:44 AM
>>> *To:* pd-list@lists.iem.at
>>> *Subject:* Re: [PD] [PD-announce] pd 0.48-0test1 released
>>>
>>> On 2017-07-21 04:50, Lucas Cordiviola wrote:
>>> >
>>> > I don't see any downside on pdx font metrics.
>>> >
>>>
>>> it seems that i remember dan (or was it miller) pointing to the newly
>>> introduced zoom feature which complicates things a bit more. and dynamic
>>> font managing. and what not. (probably check the archives)
>>> the conclusion for me is: there *are* downsides.
>>>
>>>
>>> On 2017-07-21 04:50, Lucas Cordiviola wrote:
>>> > This is a major change to everybody,
>>>
>>> certainly not.
>>>
>>> gmasdr
>>> IOhannes
>>>
>>>
>>> ___
>>> Pd-list@lists.iem.at mailing list
>>> UNSUBSCRIBE and account-management -> https://lists.puredata.info/li
>>> stinfo/pd-list
>>>
>>>
>>
>
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

[Alexandre Torres Porres]

now, since we're talking about "Dejas Vu", I'm also having a Deja Vu of my
own... it's not the first time I complain things look different, but then
someone ends up saying I needed to do something else in order to test it...
as it happened in this discussion a few months ago
https://github.com/pure-data/pure-data/pull/42

and I did ask if the idea is to be compatible to old extended (and purr
data), and I seem to have been able to test, at one point, that it actually
did and I couldn't be happier.

But I also thought that, at this point, compiling from that branch would
give me that result out of the box.

So, did I miss something or not?

cheers

2017-07-21 20:36 GMT-03:00 Alexandre Torres Porres :

> So, i just built from https://github.com/pure-data/pure-data/tree/dejavu
> and tested here on my mac. It looks good alright, seems to render better
> and stuff, but it isn't backwards compatible, seems to require more real
> estate for texts and all (object size and the like) - see attachment.
>
> So it doesn't really render the same as in purr data, like you say. That
> actually looks the same as the current pd state, as well as old pd extended.
>
> As much as I like the idea of having a new cross platform bahaviour for
> pd, I was hoping it'd be the same as purr data and all... also that I
> wouldn't have to reformat over 300 help patches I'm working on, plus my
> tutorials which is quite numerous as well - not to mentio regular
> patches/abstractions.
>
> cheers
>
>
>
> 2017-07-21 18:48 GMT-03:00 Lucas Cordiviola :
>
>> @IOhannes
>>
>> Here Dan is saying he is becoming to like the font metrics thing:
>>
>> https://lists.puredata.info/pipermail/pd-list/2017-07/119495.html
>>
>>
>>
>>
>> Mensaje telepatico asistido por maquinas.
>>
>>
>> --
>> *From:* Pd-list  on behalf of Lucas
>> Cordiviola 
>> *Sent:* Friday, July 21, 2017 7:24 PM
>> *To:* pd-list@lists.iem.at
>> *Subject:* Re: [PD] [PD-announce] pd 0.48-0test1 released
>>
>>
>>
>> @IOhannes:
>>
>> *>it seems that i remember dan (or was it miller) pointing to the newly*
>>
>> *introduced zoom feature which complicates things a bit more. and dynamic*
>>
>> *font managing. and what not. (probably check the archives)*
>>
>> *the conclusion for me is: there *are* downsides.*
>>
>>
>> The zoom feature will work with the correct doubles from pdx metrics,
>> this are the actual doubles of vanilla:
>>
>> set font_zoom2_metrics {
>> 12 20
>> 14 26
>> 18 32
>> 20 42
>> 30 50
>> 50 90
>> }
>>
>> Does *downside* to you is having a little more white space?
>>
>> Pdx font metrics wont break vanilla patches.
>>
>> See:
>> https://github.com/pure-data/pure-data/wiki/Crossplatform-fo
>> nt-metrics-%26-comparisons
>>
>>
>>
>> Mensaje telepatico asistido por maquinas.
>>
>>
>> --
>> *From:* Pd-list  on behalf of IOhannes m
>> zmoelnig 
>> *Sent:* Friday, July 21, 2017 7:44 AM
>> *To:* pd-list@lists.iem.at
>> *Subject:* Re: [PD] [PD-announce] pd 0.48-0test1 released
>>
>> On 2017-07-21 04:50, Lucas Cordiviola wrote:
>> >
>> > I don't see any downside on pdx font metrics.
>> >
>>
>> it seems that i remember dan (or was it miller) pointing to the newly
>> introduced zoom feature which complicates things a bit more. and dynamic
>> font managing. and what not. (probably check the archives)
>> the conclusion for me is: there *are* downsides.
>>
>>
>> On 2017-07-21 04:50, Lucas Cordiviola wrote:
>> > This is a major change to everybody,
>>
>> certainly not.
>>
>> gmasdr
>> IOhannes
>>
>>
>> ___
>> Pd-list@lists.iem.at mailing list
>> UNSUBSCRIBE and account-management -> https://lists.puredata.info/li
>> stinfo/pd-list
>>
>>
>
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

[Lucas Cordiviola]

Hi Alex,

What you built only changes Monaco for DejaVU. The font metrics in pd-gui.tcl 
are just vanilla.

If you want to test you can modify "pd-gui.tcl" to meet pdx metrics. If you do 
so you most likely get the same as Purr-data.

Just tweak the values on the .tcl file and restart pd.

The pdx values and the how-to are here:

https://github.com/pure-data/pure-data/wiki/Crossplatform-font-metrics-%26-comparisons




Mensaje telepatico asistido por maquinas.



From: Alexandre Torres Porres 
Sent: Friday, July 21, 2017 11:36 PM
To: Lucas Cordiviola
Cc: pd-list@lists.iem.at
Subject: Re: [PD] 048 font metrics (was: pd 0.48-0test1 released)

So, i just built from https://github.com/pure-data/pure-data/tree/dejavu and 
tested here on my mac. It looks good alright, seems to render better and stuff, 
but it isn't backwards compatible, seems to require more real estate for texts 
and all (object size and the like) - see attachment.
[https://avatars3.githubusercontent.com/u/10228017?v=4&s=400]

pure-data/pure-data
github.com
pure-data - Pure Data - tracking Miller's SourceForge git repository (also used 
by libpd)



So it doesn't really render the same as in purr data, like you say. That 
actually looks the same as the current pd state, as well as old pd extended.

As much as I like the idea of having a new cross platform bahaviour for pd, I 
was hoping it'd be the same as purr data and all... also that I wouldn't have 
to reformat over 300 help patches I'm working on, plus my tutorials which is 
quite numerous as well - not to mentio regular patches/abstractions.

cheers



2017-07-21 18:48 GMT-03:00 Lucas Cordiviola 
mailto:lucard...@hotmail.com>>:

@IOhannes

Here Dan is saying he is becoming to like the font metrics thing:

https://lists.puredata.info/pipermail/pd-list/2017-07/119495.html





Mensaje telepatico asistido por maquinas.



From: Pd-list 
mailto:pd-list-boun...@lists.iem.at>> on behalf 
of Lucas Cordiviola mailto:lucard...@hotmail.com>>
Sent: Friday, July 21, 2017 7:24 PM
To: pd-list@lists.iem.at
Subject: Re: [PD] [PD-announce] pd 0.48-0test1 released


@IOhannes:


>it seems that i remember dan (or was it miller) pointing to the newly

introduced zoom feature which complicates things a bit more. and dynamic

font managing. and what not. (probably check the archives)

the conclusion for me is: there *are* downsides.

The zoom feature will work with the correct doubles from pdx metrics, this are 
the actual doubles of vanilla:

set font_zoom2_metrics {
12 20
14 26
18 32
20 42
30 50
50 90
}

Does *downside* to you is having a little more white space?

Pdx font metrics wont break vanilla patches.

See:
https://github.com/pure-data/pure-data/wiki/Crossplatform-font-metrics-%26-comparisons




Mensaje telepatico asistido por maquinas.



From: Pd-list 
mailto:pd-list-boun...@lists.iem.at>> on behalf 
of IOhannes m zmoelnig mailto:zmoel...@iem.at>>
Sent: Friday, July 21, 2017 7:44 AM
To: pd-list@lists.iem.at
Subject: Re: [PD] [PD-announce] pd 0.48-0test1 released

On 2017-07-21 04:50, Lucas Cordiviola wrote:
>
> I don't see any downside on pdx font metrics.
>

it seems that i remember dan (or was it miller) pointing to the newly
introduced zoom feature which complicates things a bit more. and dynamic
font managing. and what not. (probably check the archives)
the conclusion for me is: there *are* downsides.


On 2017-07-21 04:50, Lucas Cordiviola wrote:
> This is a major change to everybody,

certainly not.

gmasdr
IOhannes


___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list


___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

[PD] 048 font metrics (was: pd 0.48-0test1 released)

[Lucas Cordiviola]

@IOhannes

Here Dan is saying he is becoming to like the font metrics thing:

https://lists.puredata.info/pipermail/pd-list/2017-07/119495.html





Mensaje telepatico asistido por maquinas.



From: Pd-list  on behalf of Lucas Cordiviola 

Sent: Friday, July 21, 2017 7:24 PM
To: pd-list@lists.iem.at
Subject: Re: [PD] [PD-announce] pd 0.48-0test1 released


@IOhannes:


>it seems that i remember dan (or was it miller) pointing to the newly

introduced zoom feature which complicates things a bit more. and dynamic

font managing. and what not. (probably check the archives)

the conclusion for me is: there *are* downsides.

The zoom feature will work with the correct doubles from pdx metrics, this are 
the actual doubles of vanilla:

set font_zoom2_metrics {
12 20
14 26
18 32
20 42
30 50
50 90
}

Does *downside* to you is having a little more white space?

Pdx font metrics wont break vanilla patches.

See:
https://github.com/pure-data/pure-data/wiki/Crossplatform-font-metrics-%26-comparisons




Mensaje telepatico asistido por maquinas.



From: Pd-list  on behalf of IOhannes m zmoelnig 

Sent: Friday, July 21, 2017 7:44 AM
To: pd-list@lists.iem.at
Subject: Re: [PD] [PD-announce] pd 0.48-0test1 released

On 2017-07-21 04:50, Lucas Cordiviola wrote:
>
> I don't see any downside on pdx font metrics.
>

it seems that i remember dan (or was it miller) pointing to the newly
introduced zoom feature which complicates things a bit more. and dynamic
font managing. and what not. (probably check the archives)
the conclusion for me is: there *are* downsides.


On 2017-07-21 04:50, Lucas Cordiviola wrote:
> This is a major change to everybody,

certainly not.

gmasdr
IOhannes

___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 21:21, IOhannes m zmölnig wrote:

#2 alternatively, you could just rebuild jack2 and do a `sudo make
install` (which will (well "should"; i haven't manually checked) install
to /usr/local/). then run `sudo ldconfig`, and any newly started
application will start using libjack.so from /usr/local/lib/.
this is more brittle then #1, as it potentially uses a different libjack
version - potentially breaking your system (again).
in theory you could also *not* `make install`, and then start Pd with
LD_LIBRARY_PATH to force using the non-installed libjack.so

LD_LIBRARY_PATH=/path/to/src/jack/wherever/libjack.so/is/ gdb ./pd

that's probably even more error-prone, because you might end up using a
jackd server with a non-matching pd/libjack



ooh, that sounds like playing with fire. Not sure if I am ready for this.



another interesting test would be to switch your system to jackd1 (as
provided by kxstudio), and see whether that works. (you shouldn't need
to uninstall any packages, apart from some "-dev" packages which don't
change the runtime characteristics of your system).


seems easy - Starting Synaptic shows me now libjack-jackd2-0 wants to be 
upgraded with the same version - that's probably because I installed the 
debug symbols version locally (?)


however, marking jackd1 for installation results in broken packages 
which I don't know how to resolve. See attachment
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 0.48 W32 -listdev

[Lucas Cordiviola]

This is the correct output on 047:

C:\Users\Lucarda\Downloads\pd\bin>pd -listdev -stderr
audio input devices:
0. Microphone (Mixing Driver 2 for
1. Microphone (Mixing Driver 1 for
2. Microphone (2- High Definition
audio output devices:
0. Speakers (Mixing Driver 2 for U
1. Speakers (2- High Definition Au
2. Speakers (Mixing Driver 1 for U
API number 3

no midi input devices found
MIDI output devices:
1. Microsoft MIDI Mapper
2. Microsoft GS Wavetable Synth



Mensaje telepatico asistido por maquinas.



From: Pd-list  on behalf of Lucas Cordiviola 

Sent: Friday, July 21, 2017 9:16 PM
To: pd-list@lists.iem.at
Subject: Re: [PD] 0.48 W32 -listdev


I'm not sure, also there are strange chars before the first line.

Copied from command prompt:

pd -listdev

F☻└¿≥(F☻└¿≥(Ñ☺►ε(1Ç↔audio input devices:
0. Ñ☺
1.
2. ►ε(
audio output devices:
0. 1
1. Ç
2. ↔
API number 3

no midi input devices found
MIDI output devices:
1. Microsoft MIDI Mapper
2. Microsoft GS Wavetable Synth
ü┼▒i☻└x╖Ü↓ü┼▒i☻└x╖Ü↓r9Gÿ☺╪╚╩☺♥►r9Gÿ☺╪╚╩☺♥►



there are also strange chars when starting pd from the commline:

C:\Users\Lucarda\Downloads\pd0.48-test2\pd\bin>pd
f☻└¿≥(f☻└¿≥(f☻└x╖Üf☻└x╖Ür8gÿ☺►r8gÿ☺►




Mensaje telepatico asistido por maquinas.



From: Pd-list  on behalf of IOhannes m zmölnig 

Sent: Friday, July 21, 2017 9:02 PM
To: pd-list@lists.iem.at
Subject: Re: [PD] 0.48 W32 -listdev

On 07/21/2017 10:37 PM, Lucas Cordiviola wrote:
> Hi Miller,
>
> On a normal Pd startup (mmio) audio devices are not listed (not even in the 
> GUI), or wrong chars appear.
>
> OTOH starting with -asio -listdev audio devices show correct names.

what's the proper names, and how are they displayed?
could this be an issue with utf-8 vs utf-16 (that is: encoding of
strange characters)?

fgmadsr
IOhannes

___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 21:21, IOhannes m zmölnig wrote:

On 07/21/2017 08:19 PM, Max wrote:

On 2017년 07월 21일 18:47, Dan Wilcox wrote:

In this case it might have been better to build jack itself from
source and then tell pd to link it instead. I believe there is a
configure option for this, IOhannes will know for sure.


I'm willing to try this :)


there's basically two options.
none of them includes any special flags to compile Pd.

first i'd probably restore the old jack configuration, so you can have
audio again.

#1 rebuild the (proper, working) libjack packages with debugging symbols.
there's a description how to build your package with debug symbols if
you cannot obtain them otherwise on [1] (in retrospect, this would have
been the better choice).
this will create a number of packages, you only need to install the
newly created libjack-jackd2-0 (which ought to have the exact same
version as the original package).


Cool, thanks for holding my hand.

I think the rebuilding with debug symbols was successful.

(gdb) run
Starting program: /home/max/Code/pd-0.48-0test3/bin/pd
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x77f32700 (LWP 8921)]
[New Thread 0x77eb1700 (LWP 8923)]

Thread 3 "pd" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x77eb1700 (LWP 8923)]
0x778b8f96 in Jack::JackLinuxFutex::Connect (this=0x76a990, 
name=name@entry=0x77eb0be4 "system", 
server_name=server_name@entry=0x784610 "default") at 
../linux/JackLinuxFutex.cpp:181

181 if (! fPrivate && fFutex->wasInternal)

(gdb) thread apply all bt

Thread 3 (Thread 0x77eb1700 (LWP 8966)):
#0  0x778b8f96 in Jack::JackLinuxFutex::Connect (this=0x76a990, 
name=name@entry=0x77eb0be4 "system", 
server_name=server_name@entry=0x784610 "default")

at ../linux/JackLinuxFutex.cpp:181
#1  0x778b99c5 in Jack::JackLibClient::ClientNotifyImp 
(this=0x7844e0, refnum=0, name=0x77eb0be4 "system", notify=0, 
sync=, message=, value1=0, value2=0)

at ../common/JackLibClient.cpp:169
#2  0x77893fc8 in Jack::JackClient::ClientNotify (this=0x7844e0, 
refnum=0, name=0x77eb0be4 "system", notify=, sync=1, 
message=0x77eb0c3c "", value1=0, value2=0)

at ../common/JackClient.cpp:210
#3  0x778bbb6a in Jack::JackSocketClientChannel::Execute 
(this=0x764970) at ../posix/JackSocketClientChannel.cpp:143
#4  0x778b6aa0 in Jack::JackPosixThread::ThreadHandler 
(arg=0x764a98) at ../posix/JackPosixThread.cpp:59
#5  0x7766c6ba in start_thread (arg=0x77eb1700) at 
pthread_create.c:333
#6  0x76e953dd in clone () at 
../sysdeps/unix/sysv/linux/x86_64/clone.S:109


Thread 2 (Thread 0x77f32700 (LWP 8964)):
#0  pthread_cond_wait@@GLIBC_2.3.2 () at 
../sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185
#1  0x778b78cc in Jack::JackPosixProcessSync::Wait 
(this=this@entry=0x784458) at ../posix/JackPosixProcessSync.cpp:81
#2  0x778aa828 in Jack::JackMessageBuffer::Execute 
(this=0x77c220) at ../common/JackMessageBuffer.cpp:104
#3  0x778b6aa0 in Jack::JackPosixThread::ThreadHandler 
(arg=0x784438) at ../posix/JackPosixThread.cpp:59
#4  0x7766c6ba in start_thread (arg=0x77f32700) at 
pthread_create.c:333
#5  0x76e953dd in clone () at 
../sysdeps/unix/sysv/linux/x86_64/clone.S:109


Thread 1 (Thread 0x77f348c0 (LWP 8955)):
#0  0x7767551d in read () at ../sysdeps/unix/syscall-template.S:84
#1  0x778b81f5 in read (__nbytes=4, __buf=, 
__fd=) at /usr/include/x86_64-linux-gnu/bits/unistd.h:44
#2  Jack::JackClientSocket::Read (this=0x764950, data=, 
len=4) at ../posix/JackSocket.cpp:187
#3  0x778a1e4b in Jack::JackResult::Read (trans=0x764950, 
this=0x7fffd6e0) at ../common/JackRequest.h:143
#4  Jack::JackClientOpenResult::Read (trans=0x764950, 
this=0x7fffd6e0) at ../common/JackRequest.h:294
#5  Jack::JackGenericClientChannel::ServerSyncCall 
(result=0x7fffd7dc, res=0x7fffd6e0, req=0x7fffd700, 
this=0x764970) at ../common/JackGenericClientChannel.cpp:68
#6  Jack::JackGenericClientChannel::ClientOpen (this=0x764970, 
name=name@entry=0x7fffd7e0 "pure_data", pid=8955, 
uuid=uuid@entry=-1, shared_engine=shared_engine@entry=0x7fffd7d0,
shared_client=shared_client@entry=0x7fffd7d4, 
shared_graph=0x7fffd7d8, result=0x7fffd7dc) at 
../common/JackGenericClientChannel.cpp:113
#7  0x778b9aff in Jack::JackLibClient::Open (this=0x7844e0, 
server_name=0x778bcb9f "default", name=, uuid=-1, 
options=JackNoStartServer, status=0x7fffd9fc)

at ../common/JackLibClient.cpp:114
#8  0x778bb0da in jack_client_open_aux (ap=0x7fffd8b8, 
status=0x7fffd9fc, options=JackNoStartServer, client_name=0x763200 
"pure_data") at ../common/JackLibAPI.cpp:155
#9  jack_client_open (ext_client_name=0x763200 "pure_data", 
options=JackNoStartServer, status=

Re: [PD] 0.48 W32 -listdev

[Lucas Cordiviola]

I'm not sure, also there are strange chars before the first line.

Copied from command prompt:

pd -listdev

F☻└¿≥(F☻└¿≥(Ñ☺►ε(1Ç↔audio input devices:
0. Ñ☺
1.
2. ►ε(
audio output devices:
0. 1
1. Ç
2. ↔
API number 3

no midi input devices found
MIDI output devices:
1. Microsoft MIDI Mapper
2. Microsoft GS Wavetable Synth
ü┼▒i☻└x╖Ü↓ü┼▒i☻└x╖Ü↓r9Gÿ☺╪╚╩☺♥►r9Gÿ☺╪╚╩☺♥►



there are also strange chars when starting pd from the commline:

C:\Users\Lucarda\Downloads\pd0.48-test2\pd\bin>pd
f☻└¿≥(f☻└¿≥(f☻└x╖Üf☻└x╖Ür8gÿ☺►r8gÿ☺►




Mensaje telepatico asistido por maquinas.



From: Pd-list  on behalf of IOhannes m zmölnig 

Sent: Friday, July 21, 2017 9:02 PM
To: pd-list@lists.iem.at
Subject: Re: [PD] 0.48 W32 -listdev

On 07/21/2017 10:37 PM, Lucas Cordiviola wrote:
> Hi Miller,
>
> On a normal Pd startup (mmio) audio devices are not listed (not even in the 
> GUI), or wrong chars appear.
>
> OTOH starting with -asio -listdev audio devices show correct names.

what's the proper names, and how are they displayed?
could this be an issue with utf-8 vs utf-16 (that is: encoding of
strange characters)?

fgmadsr
IOhannes

___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] 0.48 W32 -listdev

[IOhannes m zmölnig]

On 07/21/2017 10:37 PM, Lucas Cordiviola wrote:
> Hi Miller,
> 
> On a normal Pd startup (mmio) audio devices are not listed (not even in the 
> GUI), or wrong chars appear.
> 
> OTOH starting with -asio -listdev audio devices show correct names.

what's the proper names, and how are they displayed?
could this be an issue with utf-8 vs utf-16 (that is: encoding of
strange characters)?

fgmadsr
IOhannes



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

[PD] 0.48 W32 -listdev

[Lucas Cordiviola]

Hi Miller,

On a normal Pd startup (mmio) audio devices are not listed (not even in the 
GUI), or wrong chars appear.

OTOH starting with -asio -listdev audio devices show correct names.



Mensaje telepatico asistido por maquinas.
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[IOhannes m zmölnig]

On 07/21/2017 09:33 PM, Max wrote:
> OK, still allow me to vent my frustration for a moment: For OS X and
> Windows you get precompiled binaries, while Linux users are expected to
> be almost fully qualified developers just for running an application.

while i understand your frustration about having a broken system, i
think you are barking up the wrong tree here.

your trouble is not, because you "just want to run the application".
you managed to compile the application fine, without having to be a
fully qualified developer.

the problem we are facing, however, is that you are experiencing a crash
(which nobody else so far has been able to reproduce yet), and we want
to debug that crash.
debugging is more related to "fully qualified programming" than to "just
running an application".
all the serious trouble you've seen so far, is because we want to debug
the problem; and your machine is so far the only one where the problem
shows itself; that's why we depend on you.

if your system was an OSX machine (say, a Macbook Air) then you had two
possibilities: either rely on "just running the application" and live
with your problem, or to "be almost a fully qualified developer" and
help finding the cause of the problem.
the very same holds true for w32.

to quote dan:
> This have never been an issue with OS X in the past so I *do not*
> think there is suddenly a major problem based on feedback from running
> Pd on *one machine*.

does this make you feel any better? (the problem *I* am facing right
now, is that my support for digging this problem will have to end
tonight, since more pressing things are on my agenda; which I regret
very much for your sake, but about which I cannot do much...).


gfmrdsa
IOhannes



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] expr update for 0.48

[Alexandre Torres Porres]

so, seems  atanh() and other can;t be a part of windows right now, so I
added a remark about it in the documentation, here's the update

2017-07-21 16:17 GMT-03:00 Alexandre Torres Porres :

> So, as I was saying since early this year, I was working on an update of
> the documentation of [expr]. This reflects all the new and old stuff that
> were never fully documented in the help file. I made a Pull Request at
> https://github.com/pure-data/pure-data/pull/132
>
> But I also have the help file attached here for anyone to check it out
>
> Basically, I kept all the original information, but reorganized things and
> added a LOT of missing information. Now we have every single function
> documented and separated into groups, that's the main addition!  I also
> added a version number and some useful examples/information on how to use
> [expr] with value/arrays and deal with "$0".
>
> cheers
>
> 2017-07-20 2:09 GMT-03:00 Alexandre Torres Porres :
>
>> 2017-07-20 2:08 GMT-03:00 Alexandre Torres Porres :
>>
>>>
>>>
>> but I realize v0.55 is here for good to be a part of the next 0.48, so
>>> this would be v0.6 for a later release, right?
>>>
>>
>> well, one of them is the bug where atanh() doesn't load for windows, I
>> really hope this makes it into the next release
>>
>>
>


expr-help.pd
Description: Binary data
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 20:49, IOhannes m zmölnig wrote:

On 07/21/2017 04:49 PM, Max wrote:

i have no more audio now...
weep.


more important: did the crash go away with the downgrade?


I ended up with a broken Jack so I could not test this.


I want to go back. Package management seems like an awesome idea until
this. Just give me a download including all the libraries statically
linked and I will happily download a few MB more


this won't help you at all in the given situation, where you have
multiple applications (like Pd) talking to a single server (like jackd).
they *must* share some common infrastructure.


OK, still allow me to vent my frustration for a moment: For OS X and 
Windows you get precompiled binaries, while Linux users are expected to 
be almost fully qualified developers just for running an application. I 
mean sure, I can go back to Pd from the repositories (thanks IOhannes 
for the packaging work by the way, we love you). If you are not running 
debian testing but some Ubuntu (-derivate) that's version 0.46.7 of Pd. 
The latest and greatest.


We've talked about this at the convention for a moment, with Liam and me 
describing this Linux user perspective. Remember?


Dan also put in a lot of work into the install.txt to make it easier to 
follow for non-devs. Thanks Dan, we love you too.


(For Ubuntu users,) the next step would be to have a .deb file to 
install, or alternatively a PPA from which I can install a stable, 
testing and daily branch of Pd.


Someone said that setting up a PPA is trivial, so I am eager to try 
this. Actually I was already reading up on it, that's why I tried to 
compile the test version in the first place.




___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] [PD-announce] pd 0.48-0test1 released

[Lucas Cordiviola]

@IOhannes:


>it seems that i remember dan (or was it miller) pointing to the newly

introduced zoom feature which complicates things a bit more. and dynamic

font managing. and what not. (probably check the archives)

the conclusion for me is: there *are* downsides.

The zoom feature will work with the correct doubles from pdx metrics, this are 
the actual doubles of vanilla:

set font_zoom2_metrics {
12 20
14 26
18 32
20 42
30 50
50 90
}

Does *downside* to you is having a little more white space?

Pdx font metrics wont break vanilla patches.

See:
https://github.com/pure-data/pure-data/wiki/Crossplatform-font-metrics-%26-comparisons




Mensaje telepatico asistido por maquinas.



From: Pd-list  on behalf of IOhannes m zmoelnig 

Sent: Friday, July 21, 2017 7:44 AM
To: pd-list@lists.iem.at
Subject: Re: [PD] [PD-announce] pd 0.48-0test1 released

On 2017-07-21 04:50, Lucas Cordiviola wrote:
>
> I don't see any downside on pdx font metrics.
>

it seems that i remember dan (or was it miller) pointing to the newly
introduced zoom feature which complicates things a bit more. and dynamic
font managing. and what not. (probably check the archives)
the conclusion for me is: there *are* downsides.


On 2017-07-21 04:50, Lucas Cordiviola wrote:
> This is a major change to everybody,

certainly not.

gmasdr
IOhannes

___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[IOhannes m zmölnig]

On 07/21/2017 08:19 PM, Max wrote:
> On 2017년 07월 21일 18:47, Dan Wilcox wrote:
>> In this case it might have been better to build jack itself from
>> source and then tell pd to link it instead. I believe there is a
>> configure option for this, IOhannes will know for sure.
> 
> I'm willing to try this :)

there's basically two options.
none of them includes any special flags to compile Pd.

first i'd probably restore the old jack configuration, so you can have
audio again.

#1 rebuild the (proper, working) libjack packages with debugging symbols.
there's a description how to build your package with debug symbols if
you cannot obtain them otherwise on [1] (in retrospect, this would have
been the better choice).
this will create a number of packages, you only need to install the
newly created libjack-jackd2-0 (which ought to have the exact same
version as the original package).

#2 alternatively, you could just rebuild jack2 and do a `sudo make
install` (which will (well "should"; i haven't manually checked) install
to /usr/local/). then run `sudo ldconfig`, and any newly started
application will start using libjack.so from /usr/local/lib/.
this is more brittle then #1, as it potentially uses a different libjack
version - potentially breaking your system (again).
in theory you could also *not* `make install`, and then start Pd with
LD_LIBRARY_PATH to force using the non-installed libjack.so
> LD_LIBRARY_PATH=/path/to/src/jack/wherever/libjack.so/is/ gdb ./pd
that's probably even more error-prone, because you might end up using a
jackd server with a non-matching pd/libjack


another interesting test would be to switch your system to jackd1 (as
provided by kxstudio), and see whether that works. (you shouldn't need
to uninstall any packages, apart from some "-dev" packages which don't
change the runtime characteristics of your system).

gfas
IOhannes


[1]
https://wiki.debian.org/HowToGetABacktrace#Rebuilding_the_package_you.2BIBk-re_debugging



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] expr update for 0.48

[Alexandre Torres Porres]

So, as I was saying since early this year, I was working on an update of
the documentation of [expr]. This reflects all the new and old stuff that
were never fully documented in the help file. I made a Pull Request at
https://github.com/pure-data/pure-data/pull/132

But I also have the help file attached here for anyone to check it out

Basically, I kept all the original information, but reorganized things and
added a LOT of missing information. Now we have every single function
documented and separated into groups, that's the main addition!  I also
added a version number and some useful examples/information on how to use
[expr] with value/arrays and deal with "$0".

cheers

2017-07-20 2:09 GMT-03:00 Alexandre Torres Porres :

> 2017-07-20 2:08 GMT-03:00 Alexandre Torres Porres :
>
>>
>>
> but I realize v0.55 is here for good to be a part of the next 0.48, so
>> this would be v0.6 for a later release, right?
>>
>
> well, one of them is the bug where atanh() doesn't load for windows, I
> really hope this makes it into the next release
>
>


expr-help.pd
Description: Binary data
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[IOhannes m zmölnig]

On 07/21/2017 06:11 PM, Max wrote:
> 
>> I also wonder if this is a "linking jack2 when we want jack1" issue.
> 
> I don't think so. I had a jack2 system until I just broke everything
> trying to help debugging..


i don't think so either.
jackd2 and jackd1 are really binary compatible.

they offer slightly different feature-sets (like dbus support), but
that's not something that is visible to the consuming applications
(those that link against libjack).

fmgards
IOhannes

PS: of course i feel your pain that you broke the system. i'm really
sorry about this, despite my "more important" remark in my other mail.



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[IOhannes m zmölnig]

On 07/21/2017 04:49 PM, Max wrote:
> On 2017년 07월 21일 12:51, IOhannes m zmölnig wrote:
>> On 07/21/2017 11:28 AM, Max wrote:
>>> On 2017년 07월 21일 07:43, IOhannes m zmölnig wrote:
>> i think, you are still missing the debug packages of libjack.
>> you need to enable to debugsym repository [1], and then install  the
>> "libjack-jackd2-0-dbgsym" package (if libjack-jackd2-0 provides your the
>> libjack.so).
> 
> I think that KXStudio doesn't provide the debug symbol version, so I
> need to downgrade libjackd2-0 to
> 1.9.10+20150825git1ed50c92~dfsg-1ubuntu1 from 2:1.9.11~20161209-1~xenial5.
> However, that forces me to remove 20 depending packages, what I just
> went ahead with...
> 
> i have no more audio now...
> weep.

more important: did the crash go away with the downgrade?

> 
> I want to go back. Package management seems like an awesome idea until
> this. Just give me a download including all the libraries statically
> linked and I will happily download a few MB more

this won't help you at all in the given situation, where you have
multiple applications (like Pd) talking to a single server (like jackd).
they *must* share some common infrastructure.

mgfdst
IOhannes



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 18:47, Dan Wilcox wrote:
In this case it might have been better to build jack itself from source 
and then tell pd to link it instead. I believe there is a configure 
option for this, IOhannes will know for sure.


I'm willing to try this :)



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Max]

On 2017년 07월 21일 16:41, me.grimm wrote:
 >> Desperately installing the Adobe crack on OS X and in the process of 
hindering it of calling home deleting the localhost line?

not sure what this means... so delete this file?


There apparently is a crack out there for the Adobe suite that only 
works if your (illegally obtained) serial number is not sent to the 
Adobe servers for authentification. So the crack tells you to go and 
edit the hosts file and _ADD_ some lines so that DNS requests to 
adobe.com will not be resolved and thus your applications keep working. 
However, I have seen multiple instances where the user has mistaken the 
instructions of the cracker and has DELETED or REPLACED the whole hosts 
file, so that in the process localhost can't be resolved anymore, 
breaking Pd.



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Miller Puckette]

Can you try a desperate workaround?  in tcl/pd_connect.tcl, replace the
funciton ::pd_connect::to_pd this way:

# if pd opens first, it starts pd-gui, then pd-gui connects to the port pd sent
proc ::pd_connect::to_pd {port {host localhost}} {
variable pd_socket
::pdwindow::debug "'pd-gui' connecting to 'pd' on localhost $port ...\n"
if {[catch {set pd_socket [socket $host $port]}]} {
if {$filename ne "localhost" || [catch {set pd_socket [socket 127.0.0.1 
$port]}]} {
puts stderr "WARNING: connect to pd failed, retrying port 
$host:$port."
after 1000 ::pd_connect::to_pd $port $host
return
}
}
::pd_connect::configure_socket $pd_socket
}

If 'localhost" falist that retries with 127.0.0.1 :)

Miller

On Fri, Jul 21, 2017 at 10:41:12AM -0400, me.grimm wrote:
> >> Is the firewall on ...?
> firewall is off
> 
> >> ping localhost
> ping: cannot resolve localhost: Unknown host
> 
> >> with -nogui in Terminal ala
> n$ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -nogui -verbose
> input channels = 2, output channels = 2
> Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
> error: Ignoring invalid font-metrics from GUI!
> Pt_Start() called
> 
> >> is that the classic messed up /private/etc/hosts file?
> yeah this is what I found there:
> https://github.com/lennylxx/ipv6-hosts
> 
> >> Desperately installing the Adobe crack on OS X and in the process of
> hindering it of calling home deleting the localhost line?
> not sure what this means... so delete this file?
> 
> so, the /private/etc/hosts file "Hosts file which is used for improving
> IPv6 access speed to Google, YouTube, Facebook, Wikipedia, etc. in Mainland
> China."
> 
> ... and yes the student is from china so that explains this at least.
> 
> so what do I do now? :)
> 
> this gets us into privacy laws, china gov laws?, no idea, etc. which turns
> this to a much deeper discussion since I assume we are here to support as
> many students as possible but without fully understanding what a systems
> level change could potentially do once students return to native country I
> really have no experience or clue here
> 
> max, you have some experience here no?
> 
> well anyway, might there be a work around for now? or could pd issue a
> warning like "is your firewall on?" OR "sup with your /private/etc/hosts
> file?
> 
> thanks!
> m
> 
> 
> On Fri, Jul 21, 2017 at 10:09 AM, Max  wrote:
> 
> > is that the classic messed up /private/etc/hosts file? Desperately
> > installing the Adobe crack on OS X and in the process of hindering it of
> > calling home deleting the localhost line?
> >
> > ping localhost
> >
> >
> >
> >
> > On 2017년 07월 21일 16:01, me.grimm wrote:
> >
> >> same stuff going on here... what to be done?
> >>
> >> test3
> >> $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
> >> input channels = 2, output channels = 2
> >> Pd-0.48.0 ("test3") compiled 07:55:09 Jul 21 2017
> >> port 5400
> >> "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
> >> Waiting for connection request...
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> ^CPd: signal 2
> >> gui socket 0 - closing audio...
> >> closing MIDI...
> >> ... done.
> >>
> >> from this morn git pull
> >> $ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -verbose
> >> input channels = 2, output channels = 2
> >> Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
> >> port 5400
> >> "/Volumes/U/Pd-0.48-0test3.app/Contents/Resources/../MacOS/Pd" 5400
> >> Waiting for connection request...
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> ^CPd: signal 2
> >> gui socket 0 - closing audio...
> >> closing MIDI...
> >> ... done.
> >> $
> >>
> >> On Thu, Jul 20, 2017 at 11:37 AM, Miller Puckette  >> m...@ucsd.edu>> wrote:
> >>
> >> Thanks.. and how does "test3" behave?  It's compiled with 8.5-gen
> >> Tcl/TK - should behave like "test1" ded, except with better
> >> text spacing.
> >>
> >> M

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Dan Wilcox]

> On Jul 21, 2017, at 6:11 PM, Max  wrote:
> 
> On 2017년 07월 21일 18:03, Dan Wilcox wrote:
>> In the meantime, can't you install the latest stable release in the package 
>> manager? Thanks for helping test the new version at any rate.
> 
> I wish I could, but when choosing "force version" in synaptic, I end up with 
> broken packages and can't apply the changes any more.

Ouch. Yeah, I guess I don't have to tell you forcing things is a bad idea. 
Sorry.

>> I also wonder if this is a "linking jack2 when we want jack1" issue.
> 
> I don't think so. I had a jack2 system until I just broke everything trying 
> to help debugging..

In this case it might have been better to build jack itself from source and 
then tell pd to link it instead. I believe there is a configure option for 
this, IOhannes will know for sure.


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 18:03, Dan Wilcox wrote:
In the meantime, can't you install the latest stable release in the 
package manager? Thanks for helping test the new version at any rate.


I wish I could, but when choosing "force version" in synaptic, I end up 
with broken packages and can't apply the changes any more.



I also wonder if this is a "linking jack2 when we want jack1" issue.


I don't think so. I had a jack2 system until I just broke everything 
trying to help debugging..




___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Dan Wilcox]

> On Jul 21, 2017, at 5:43 PM, pd-list-requ...@lists.iem.at wrote:
> 
> From: Max mailto:abonneme...@revolwear.com>>
> Subject: Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux
> Date: July 21, 2017 at 4:49:50 PM GMT+2
> To: pd-list@lists.iem.at 
> 
> 
> I want to go back. Package management seems like an awesome idea until this. 
> Just give me a download including all the libraries statically linked and I 
> will happily download a few MB more
> /rant end./


In the meantime, can't you install the latest stable release in the package 
manager? Thanks for helping test the new version at any rate.

I also wonder if this is a "linking jack2 when we want jack1" issue.


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Max]

Related anecdote:
I had students buy a copy of Max, because they could not get Pd to run, 
because the pirated Adobe and messed up their localhost in the process. 
Free software irony.


m.



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Dan Wilcox]

This have never been an issue with OS X in the past so I *do not* think there 
is suddenly a major problem based on feedback from running Pd on *one machine*.

I suggest another tool or manual updates to the host file is the cause. It's 
also better to use localhost as it can resolve in both IPv4 and IPv6 instead of 
127.0.0.1 which is IPv4 only.

Also, by "firewall" I mean the built in Firewall utility in the OS, not the 
low-level hosts files or DNS settings.

> On Jul 21, 2017, at 5:43 PM, pd-list-requ...@lists.iem.at wrote:
> 
> From: Miller Puckette mailto:m...@ucsd.edu>>
> Subject: Re: [PD] Macbook Air Errors with 0.48 Test 1
> Date: July 21, 2017 at 5:43:27 PM GMT+2
> To: "me.grimm" mailto:megr...@gmail.com>>
> Cc: "pd-list@lists.iem.at " 
> mailto:pd-list@lists.iem.at>>
> 
> 
> Here's an idea: get into the tcl code (pd/tcl) and change all instances of
> 
> localhost
> 
> to
> 
> 127.0.0.1
> 
> Perhaps this is more robust.
> 
> cheers
> Miller
> 
> On Fri, Jul 21, 2017 at 10:41:12AM -0400, me.grimm wrote:
 Is the firewall on ...?
>> firewall is off
>> 
 ping localhost
>> ping: cannot resolve localhost: Unknown host


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Miller Puckette]

Here's an idea: get into the tcl code (pd/tcl) and change all instances of

localhost

to

127.0.0.1

Perhaps this is more robust.

cheers
Miller

On Fri, Jul 21, 2017 at 10:41:12AM -0400, me.grimm wrote:
> >> Is the firewall on ...?
> firewall is off
> 
> >> ping localhost
> ping: cannot resolve localhost: Unknown host
> 
> >> with -nogui in Terminal ala
> n$ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -nogui -verbose
> input channels = 2, output channels = 2
> Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
> error: Ignoring invalid font-metrics from GUI!
> Pt_Start() called
> 
> >> is that the classic messed up /private/etc/hosts file?
> yeah this is what I found there:
> https://github.com/lennylxx/ipv6-hosts
> 
> >> Desperately installing the Adobe crack on OS X and in the process of
> hindering it of calling home deleting the localhost line?
> not sure what this means... so delete this file?
> 
> so, the /private/etc/hosts file "Hosts file which is used for improving
> IPv6 access speed to Google, YouTube, Facebook, Wikipedia, etc. in Mainland
> China."
> 
> ... and yes the student is from china so that explains this at least.
> 
> so what do I do now? :)
> 
> this gets us into privacy laws, china gov laws?, no idea, etc. which turns
> this to a much deeper discussion since I assume we are here to support as
> many students as possible but without fully understanding what a systems
> level change could potentially do once students return to native country I
> really have no experience or clue here
> 
> max, you have some experience here no?
> 
> well anyway, might there be a work around for now? or could pd issue a
> warning like "is your firewall on?" OR "sup with your /private/etc/hosts
> file?
> 
> thanks!
> m
> 
> 
> On Fri, Jul 21, 2017 at 10:09 AM, Max  wrote:
> 
> > is that the classic messed up /private/etc/hosts file? Desperately
> > installing the Adobe crack on OS X and in the process of hindering it of
> > calling home deleting the localhost line?
> >
> > ping localhost
> >
> >
> >
> >
> > On 2017년 07월 21일 16:01, me.grimm wrote:
> >
> >> same stuff going on here... what to be done?
> >>
> >> test3
> >> $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
> >> input channels = 2, output channels = 2
> >> Pd-0.48.0 ("test3") compiled 07:55:09 Jul 21 2017
> >> port 5400
> >> "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
> >> Waiting for connection request...
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> ^CPd: signal 2
> >> gui socket 0 - closing audio...
> >> closing MIDI...
> >> ... done.
> >>
> >> from this morn git pull
> >> $ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -verbose
> >> input channels = 2, output channels = 2
> >> Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
> >> port 5400
> >> "/Volumes/U/Pd-0.48-0test3.app/Contents/Resources/../MacOS/Pd" 5400
> >> Waiting for connection request...
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> WARNING: connect to pd failed, retrying port localhost:5400.
> >> ^CPd: signal 2
> >> gui socket 0 - closing audio...
> >> closing MIDI...
> >> ... done.
> >> $
> >>
> >> On Thu, Jul 20, 2017 at 11:37 AM, Miller Puckette  >> m...@ucsd.edu>> wrote:
> >>
> >> Thanks.. and how does "test3" behave?  It's compiled with 8.5-gen
> >> Tcl/TK - should behave like "test1" ded, except with better
> >> text spacing.
> >>
> >> M
> >>
> >> On Thu, Jul 20, 2017 at 11:08:30AM -0400, me.grimm wrote:
> >>  > ok... got a chance to test this morn.
> >>  >
> >>  > miller test02 (pd icon just bounces in the dock)
> >>  >
> >>  > $ /Volumes/U/Pd-0.48-0test2.app/Contents/Resources/bin/pd -verbose
> >>  > input channels = 2, output channels = 2
> >>  > Pd-0.48.0 ("test2") compiled 08:40:39 Jul 19 2017
> >>  > port 5400
> >>  > "/Volumes/U/Pd-0.48-0test2.app/Contents/Resources/../MacOS/Pd"
> >> 5400
> >>  > Waiting for connection request...
> >>  > WARNING: connect to pd failed, retrying port localhost:5400.
> >>

Re: [PD] [PD-announce] pd 0.48-0test1 released

[Miller Puckette]

I took them out just now :)

M

On Fri, Jul 21, 2017 at 11:43:54AM +0200, Dan Wilcox wrote:
> 
> > On Jul 20, 2017, at 6:43 PM, pd-list-requ...@lists.iem.at wrote:
> > 
> > 1) the very first time you start up Pd, there's no log level set (and 
> > therefore you don't get any errors).
> 
> This is fixed. When I added log level saving I forget to check the case where 
> it wasn't set (in older files) and that defaulted to a level of 0. Whoops.
> 
> > 3) when I create [midiin] and [sysexin] I get the error 'windows: not 
> > supported' in the console. these objectes used to always work for me. did 
> > you change something? unfortunately I don't have any MIDI equipment right 
> > now to test it.
> 
> 
> FWIW Those are simple prints only for WIN32 in x_midi.c and not the result of 
> some dependency check. If things are working, they can probably be safely 
> removed. The warning about them not working on Windows can also be removed 
> from the midi object help file.
> 
> 
> Dan Wilcox
> @danomatika 
> danomatika.com 
> robotcowboy.com 
> 
> 
> 

> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list


___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Max]

On 2017년 07월 21일 16:41, me.grimm wrote:

 >> Is the firewall on ...?
firewall is off

 >> ping localhost
ping: cannot resolve localhost: Unknown host

 >> with -nogui in Terminal ala
n$ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -nogui -verbose
input channels = 2, output channels = 2
Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
error: Ignoring invalid font-metrics from GUI!
Pt_Start() called

 >> is that the classic messed up /private/etc/hosts file?
yeah this is what I found there:
https://github.com/lennylxx/ipv6-hosts

 >> Desperately installing the Adobe crack on OS X and in the process of 
hindering it of calling home deleting the localhost line?

not sure what this means... so delete this file?

so, the /private/etc/hosts file "Hosts file which is used for improving 
IPv6 access speed to Google, YouTube, Facebook, Wikipedia, etc. in 
Mainland China."


... and yes the student is from china so that explains this at least.

so what do I do now? :)

this gets us into privacy laws, china gov laws?, no idea, etc. which 
turns this to a much deeper discussion since I assume we are here to 
support as many students as possible but without fully understanding 
what a systems level change could potentially do once students return to 
native country I really have no experience or clue here


max, you have some experience here no?

well anyway, might there be a work around for now? or could pd issue a 
warning like "is your firewall on?" OR "sup with your /private/etc/hosts 
file?


according to the web, this is the default hosts file on OS X:

---begin--
##
##
# Host Database
#
# localhost is used to configure the loopback interface
# when the system is booting. Do not change this entry.
##
127.0.0.1 localhost
255.255.255.255 broadcasthost
::1 localhost
fe80::1%lo0 localhost
--end---


there is even a warning in there that those users seem to ignore.
I'm pretty sure those four lines need to be in there, you can add all 
the stuff from

https://raw.githubusercontent.com/lennylxx/ipv6-hosts/master/hosts
if you want to "optimize" things for use in China, but keep those 4 lines!!






___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 12:51, IOhannes m zmölnig wrote:

On 07/21/2017 11:28 AM, Max wrote:

On 2017년 07월 21일 07:43, IOhannes m zmölnig wrote:

i think, you are still missing the debug packages of libjack.
you need to enable to debugsym repository [1], and then install  the
"libjack-jackd2-0-dbgsym" package (if libjack-jackd2-0 provides your the
libjack.so).


I think that KXStudio doesn't provide the debug symbol version, so I 
need to downgrade libjackd2-0 to 
1.9.10+20150825git1ed50c92~dfsg-1ubuntu1 from 2:1.9.11~20161209-1~xenial5.
However, that forces me to remove 20 depending packages, what I just 
went ahead with...


i have no more audio now...
weep.

I want to go back. Package management seems like an awesome idea until 
this. Just give me a download including all the libraries statically 
linked and I will happily download a few MB more

/rant end./

___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Macbook Air Errors with 0.48 Test 1

[me.grimm]

>> Is the firewall on ...?
firewall is off

>> ping localhost
ping: cannot resolve localhost: Unknown host

>> with -nogui in Terminal ala
n$ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -nogui -verbose
input channels = 2, output channels = 2
Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
error: Ignoring invalid font-metrics from GUI!
Pt_Start() called

>> is that the classic messed up /private/etc/hosts file?
yeah this is what I found there:
https://github.com/lennylxx/ipv6-hosts

>> Desperately installing the Adobe crack on OS X and in the process of
hindering it of calling home deleting the localhost line?
not sure what this means... so delete this file?

so, the /private/etc/hosts file "Hosts file which is used for improving
IPv6 access speed to Google, YouTube, Facebook, Wikipedia, etc. in Mainland
China."

... and yes the student is from china so that explains this at least.

so what do I do now? :)

this gets us into privacy laws, china gov laws?, no idea, etc. which turns
this to a much deeper discussion since I assume we are here to support as
many students as possible but without fully understanding what a systems
level change could potentially do once students return to native country I
really have no experience or clue here

max, you have some experience here no?

well anyway, might there be a work around for now? or could pd issue a
warning like "is your firewall on?" OR "sup with your /private/etc/hosts
file?

thanks!
m


On Fri, Jul 21, 2017 at 10:09 AM, Max  wrote:

> is that the classic messed up /private/etc/hosts file? Desperately
> installing the Adobe crack on OS X and in the process of hindering it of
> calling home deleting the localhost line?
>
> ping localhost
>
>
>
>
> On 2017년 07월 21일 16:01, me.grimm wrote:
>
>> same stuff going on here... what to be done?
>>
>> test3
>> $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
>> input channels = 2, output channels = 2
>> Pd-0.48.0 ("test3") compiled 07:55:09 Jul 21 2017
>> port 5400
>> "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
>> Waiting for connection request...
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> ^CPd: signal 2
>> gui socket 0 - closing audio...
>> closing MIDI...
>> ... done.
>>
>> from this morn git pull
>> $ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -verbose
>> input channels = 2, output channels = 2
>> Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
>> port 5400
>> "/Volumes/U/Pd-0.48-0test3.app/Contents/Resources/../MacOS/Pd" 5400
>> Waiting for connection request...
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> ^CPd: signal 2
>> gui socket 0 - closing audio...
>> closing MIDI...
>> ... done.
>> $
>>
>> On Thu, Jul 20, 2017 at 11:37 AM, Miller Puckette > m...@ucsd.edu>> wrote:
>>
>> Thanks.. and how does "test3" behave?  It's compiled with 8.5-gen
>> Tcl/TK - should behave like "test1" ded, except with better
>> text spacing.
>>
>> M
>>
>> On Thu, Jul 20, 2017 at 11:08:30AM -0400, me.grimm wrote:
>>  > ok... got a chance to test this morn.
>>  >
>>  > miller test02 (pd icon just bounces in the dock)
>>  >
>>  > $ /Volumes/U/Pd-0.48-0test2.app/Contents/Resources/bin/pd -verbose
>>  > input channels = 2, output channels = 2
>>  > Pd-0.48.0 ("test2") compiled 08:40:39 Jul 19 2017
>>  > port 5400
>>  > "/Volumes/U/Pd-0.48-0test2.app/Contents/Resources/../MacOS/Pd"
>> 5400
>>  > Waiting for connection request...
>>  > WARNING: connect to pd failed, retrying port localhost:5400.
>>  > WARNING: connect to pd failed, retrying port localhost:5400.
>>  > WARNING: connect to pd failed, retrying port localhost:5400.
>>  > ^CPd: signal 2
>>  > gui socket 0 - closing audio...
>>  > closing MIDI...
>>  > ... done.
>>  > $
>>  >
>>  >
>>  > my compile with tk 8.6.6 (tk "feather" icon just bounces in the
>> dock
>>  >
>>  > $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
>>  > inp

Re: [PD] One click too many on macOS

[Roman Haefeli]

On Mit, 2017-07-19 at 18:30 +0200, Dan Wilcox wrote:
> Ah ok, I misunderstood. What I was referring to was *very* buggy
> behavior such as creating a bang with the key binding creates two
> bangs...
> 
> What you're talking about is the fact that macOS does not have "focus
> follows mouse", which it sounds like you are used to.

No, at least: I'm not used to windows coming into focus just by
hovering them, if that's what you mean.

>  Yes, clicking to focus the window is normal. You can hover over an
> unfocused window and scroll it but not click in without the initial
> focus click.

Hm.. I don't have a Mac nearby right now, but last time I checked this:
When Finder is not in focus, I can click on folder and the Finder
window comes to the front AND the folder is selected. The same with any
button in qjackctl. I didn't test any other apps. However, with Pd
triggering a [bng] of a window that is not in focus requires two
clicks. As far as I can tell, that behavior is specific to Pd and seems
not to be a Mac paradigm (what I initially thought it would be). 

> I'd suggest using CMD+` to cycle the currently active app's windows.

Thanks, that is a handy shortcut.

Roman


signature.asc
Description: This is a digitally signed message part
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Max]

is that the classic messed up /private/etc/hosts file? Desperately 
installing the Adobe crack on OS X and in the process of hindering it of 
calling home deleting the localhost line?


ping localhost



On 2017년 07월 21일 16:01, me.grimm wrote:

same stuff going on here... what to be done?

test3
$ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
input channels = 2, output channels = 2
Pd-0.48.0 ("test3") compiled 07:55:09 Jul 21 2017
port 5400
"/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
Waiting for connection request...
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
^CPd: signal 2
gui socket 0 - closing audio...
closing MIDI...
... done.

from this morn git pull
$ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -verbose
input channels = 2, output channels = 2
Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
port 5400
"/Volumes/U/Pd-0.48-0test3.app/Contents/Resources/../MacOS/Pd" 5400
Waiting for connection request...
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
^CPd: signal 2
gui socket 0 - closing audio...
closing MIDI...
... done.
$

On Thu, Jul 20, 2017 at 11:37 AM, Miller Puckette > wrote:


Thanks.. and how does "test3" behave?  It's compiled with 8.5-gen
Tcl/TK - should behave like "test1" ded, except with better
text spacing.

M

On Thu, Jul 20, 2017 at 11:08:30AM -0400, me.grimm wrote:
 > ok... got a chance to test this morn.
 >
 > miller test02 (pd icon just bounces in the dock)
 >
 > $ /Volumes/U/Pd-0.48-0test2.app/Contents/Resources/bin/pd -verbose
 > input channels = 2, output channels = 2
 > Pd-0.48.0 ("test2") compiled 08:40:39 Jul 19 2017
 > port 5400
 > "/Volumes/U/Pd-0.48-0test2.app/Contents/Resources/../MacOS/Pd" 5400
 > Waiting for connection request...
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > ^CPd: signal 2
 > gui socket 0 - closing audio...
 > closing MIDI...
 > ... done.
 > $
 >
 >
 > my compile with tk 8.6.6 (tk "feather" icon just bounces in the dock
 >
 > $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
 > input channels = 2, output channels = 2
 > Pd-0.48.0 ("test1") compiled 16:38:24 Jul 19 2017
 > port 5400
 > "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
 > Waiting for connection request...
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > WARNING: connect to pd failed, retrying port localhost:5400.
 > ^CPd: signal 2
 > gui socket 0 - closing audio...
 > closing MIDI...
 > ... done.
 > $
 >
 >
 > tell me what to try and i will since i have access to student
computer next
 > couple weeks
 >
 > cheers
 > m
 >
 >
 > On Wed, Jul 19, 2017 at 1:32 PM, Miller Puckette mailto:m...@ucsd.edu>> wrote:
 >
 > > Can you try

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Dan Wilcox]

Also, can you confirm the pd core is working by starting it with -nogui in 
Terminal ala:

/path/to/Pd.app/Contents/Resources/bin/pd -nogui -verbose

> On Jul 21, 2017, at 4:03 PM, Dan Wilcox  wrote:
> 
> Is the firewall on ...?
> 
>> On Jul 21, 2017, at 4:01 PM, me.grimm > > wrote:
>> 
>> same stuff going on here... what to be done?
>> 
>> test3
>> $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
>> input channels = 2, output channels = 2
>> Pd-0.48.0 ("test3") compiled 07:55:09 Jul 21 2017
>> port 5400
>> "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
>> Waiting for connection request... 
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> ^CPd: signal 2
>> gui socket 0 - closing audio...
>> closing MIDI...
>> ... done.
>> 
>> from this morn git pull
>> $ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -verbose
>> input channels = 2, output channels = 2
>> Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
>> port 5400
>> "/Volumes/U/Pd-0.48-0test3.app/Contents/Resources/../MacOS/Pd" 5400
>> Waiting for connection request... 
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> WARNING: connect to pd failed, retrying port localhost:5400.
>> ^CPd: signal 2
>> gui socket 0 - closing audio...
>> closing MIDI...
>> ... done.
>> $ 
>> 
>> On Thu, Jul 20, 2017 at 11:37 AM, Miller Puckette > > wrote:
>> Thanks.. and how does "test3" behave?  It's compiled with 8.5-gen
>> Tcl/TK - should behave like "test1" ded, except with better
>> text spacing.
>> 
>> M
>> 
>> On Thu, Jul 20, 2017 at 11:08:30AM -0400, me.grimm wrote:
>> > ok... got a chance to test this morn.
>> >
>> > miller test02 (pd icon just bounces in the dock)
>> >
>> > $ /Volumes/U/Pd-0.48-0test2.app/Contents/Resources/bin/pd -verbose
>> > input channels = 2, output channels = 2
>> > Pd-0.48.0 ("test2") compiled 08:40:39 Jul 19 2017
>> > port 5400
>> > "/Volumes/U/Pd-0.48-0test2.app/Contents/Resources/../MacOS/Pd" 5400
>> > Waiting for connection request...
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > ^CPd: signal 2
>> > gui socket 0 - closing audio...
>> > closing MIDI...
>> > ... done.
>> > $
>> >
>> >
>> > my compile with tk 8.6.6 (tk "feather" icon just bounces in the dock
>> >
>> > $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
>> > input channels = 2, output channels = 2
>> > Pd-0.48.0 ("test1") compiled 16:38:24 Jul 19 2017
>> > port 5400
>> > "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
>> > Waiting for connection request...
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > WARNING: connect to pd failed, retrying port localhost:5400.
>> > ^CPd: signal 2
>> > gui socket 0 - closing audio...
>> > closing MIDI...
>> > ... done.
>> > $
>> >
>> >
>> > tell me what to try and i will since i have access to student computer next
>> > couple weeks
>> >
>> > cheers
>> > m
>> >
>> >
>> > On Wed, Jul 19, 2017 a

Re: [PD] Macbook Air Errors with 0.48 Test 1

[Dan Wilcox]

Is the firewall on ...?

> On Jul 21, 2017, at 4:01 PM, me.grimm  wrote:
> 
> same stuff going on here... what to be done?
> 
> test3
> $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
> input channels = 2, output channels = 2
> Pd-0.48.0 ("test3") compiled 07:55:09 Jul 21 2017
> port 5400
> "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
> Waiting for connection request... 
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> ^CPd: signal 2
> gui socket 0 - closing audio...
> closing MIDI...
> ... done.
> 
> from this morn git pull
> $ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -verbose
> input channels = 2, output channels = 2
> Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
> port 5400
> "/Volumes/U/Pd-0.48-0test3.app/Contents/Resources/../MacOS/Pd" 5400
> Waiting for connection request... 
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> WARNING: connect to pd failed, retrying port localhost:5400.
> ^CPd: signal 2
> gui socket 0 - closing audio...
> closing MIDI...
> ... done.
> $ 
> 
> On Thu, Jul 20, 2017 at 11:37 AM, Miller Puckette  > wrote:
> Thanks.. and how does "test3" behave?  It's compiled with 8.5-gen
> Tcl/TK - should behave like "test1" ded, except with better
> text spacing.
> 
> M
> 
> On Thu, Jul 20, 2017 at 11:08:30AM -0400, me.grimm wrote:
> > ok... got a chance to test this morn.
> >
> > miller test02 (pd icon just bounces in the dock)
> >
> > $ /Volumes/U/Pd-0.48-0test2.app/Contents/Resources/bin/pd -verbose
> > input channels = 2, output channels = 2
> > Pd-0.48.0 ("test2") compiled 08:40:39 Jul 19 2017
> > port 5400
> > "/Volumes/U/Pd-0.48-0test2.app/Contents/Resources/../MacOS/Pd" 5400
> > Waiting for connection request...
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > ^CPd: signal 2
> > gui socket 0 - closing audio...
> > closing MIDI...
> > ... done.
> > $
> >
> >
> > my compile with tk 8.6.6 (tk "feather" icon just bounces in the dock
> >
> > $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
> > input channels = 2, output channels = 2
> > Pd-0.48.0 ("test1") compiled 16:38:24 Jul 19 2017
> > port 5400
> > "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
> > Waiting for connection request...
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > ^CPd: signal 2
> > gui socket 0 - closing audio...
> > closing MIDI...
> > ... done.
> > $
> >
> >
> > tell me what to try and i will since i have access to student computer next
> > couple weeks
> >
> > cheers
> > m
> >
> >
> > On Wed, Jul 19, 2017 at 1:32 PM, Miller Puckette  > > wrote:
> >
> > > Can you try with "test2"? (on my site, just now uploaded it).
> > >
> > > thanks
> > > Miller
> > >
> > > On Wed, Jul 19, 2017 at 06:00:36PM +0200, Dan Wilcox wrote:
> > > > The error indicates the pd core has crashed or is not running. Sometimes
> > > this also happens if the pd

Re: [PD] Macbook Air Errors with 0.48 Test 1

[me.grimm]

same stuff going on here... what to be done?

test3
$ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
input channels = 2, output channels = 2
Pd-0.48.0 ("test3") compiled 07:55:09 Jul 21 2017
port 5400
"/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
Waiting for connection request...
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
^CPd: signal 2
gui socket 0 - closing audio...
closing MIDI...
... done.

from this morn git pull
$ /Volumes/U/Pd-0.48-0test3.app/Contents/Resources/bin/pd -verbose
input channels = 2, output channels = 2
Pd-0.48.0 ("test3") compiled 22:16:32 Jul 19 2017
port 5400
"/Volumes/U/Pd-0.48-0test3.app/Contents/Resources/../MacOS/Pd" 5400
Waiting for connection request...
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
WARNING: connect to pd failed, retrying port localhost:5400.
^CPd: signal 2
gui socket 0 - closing audio...
closing MIDI...
... done.
$

On Thu, Jul 20, 2017 at 11:37 AM, Miller Puckette  wrote:

> Thanks.. and how does "test3" behave?  It's compiled with 8.5-gen
> Tcl/TK - should behave like "test1" ded, except with better
> text spacing.
>
> M
>
> On Thu, Jul 20, 2017 at 11:08:30AM -0400, me.grimm wrote:
> > ok... got a chance to test this morn.
> >
> > miller test02 (pd icon just bounces in the dock)
> >
> > $ /Volumes/U/Pd-0.48-0test2.app/Contents/Resources/bin/pd -verbose
> > input channels = 2, output channels = 2
> > Pd-0.48.0 ("test2") compiled 08:40:39 Jul 19 2017
> > port 5400
> > "/Volumes/U/Pd-0.48-0test2.app/Contents/Resources/../MacOS/Pd" 5400
> > Waiting for connection request...
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > ^CPd: signal 2
> > gui socket 0 - closing audio...
> > closing MIDI...
> > ... done.
> > $
> >
> >
> > my compile with tk 8.6.6 (tk "feather" icon just bounces in the dock
> >
> > $ /Volumes/U/Pd.app/Contents/Resources/bin/pd -verbose
> > input channels = 2, output channels = 2
> > Pd-0.48.0 ("test1") compiled 16:38:24 Jul 19 2017
> > port 5400
> > "/Volumes/U/Pd.app/Contents/Resources/../MacOS/Pd" 5400
> > Waiting for connection request...
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > WARNING: connect to pd failed, retrying port localhost:5400.
> > ^CPd: signal 2
> > gui socket 0 - closing audio...
> > closing MIDI...
> > ... done.
> > $
> >
> >
> > tell me what to try and i will since i have access to student computer
> next
> > couple weeks
> >
> > cheers
> > m
> >
> >
> > On Wed, Jul 19, 2017 at 1:32 PM, Miller Puckette  wrote:
> >
> > > Can you try with "test2"? (on my site, just now uploaded it).
> > >
> > > thanks
> > > Miller
> > >
> > > On Wed, Jul 19, 2017 at 06:00:36PM +0200, Dan Wilcox wrote:
> > > > The error indicates the pd core has crashed or is not running.
> Sometimes
> > > this also happens if the pd proves has hung in the background. Open
> > > Activity Monitor and make sure there is no "pd" processing running, if
> so
> > > then kill it.
> > > >
> > > > As a test, try running the pd core from the command line 

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Roman Haefeli]

On Fre, 2017-07-21 at 13:12 +0200, IOhannes m zmölnig wrote:
> - why don't you get the "Cannot lock down XXX byte memory area
> (Operation not permitted)" error (on the console) that i get?

I get that, too:

$ pd -jack -channels 2 -rt
Cannot lock down 1186 byte memory area (Operation not permitted)
Cannot lock down 82274202 byte memory area (Operation not permitted)
Cannot lock down 422 byte memory area (Operation not permitted)

And was wondering: Why does it appear suddenly with recent versions of
Pd?

Roman

signature.asc
Description: This is a digitally signed message part
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 13:12, IOhannes m zmölnig wrote:

On 07/21/2017 12:51 PM, IOhannes m zmölnig wrote:


better?


somewhat, but not much.



apart from that:
- what's the output of 'ulimit -a'
- is there a difference if you start jackd with/without realtime settings?
- is there a difference if you start jackd with/without memory locking?
- does it crash if you start Pd without jackd running?
- why don't you get the "Cannot lock down XXX byte memory area
(Operation not permitted)" error (on the console) that i get?
- what's the output of Pd (on the terminal and/or console) if you use a
different backend: './pd -verbose -rt -alsa'



IOhannes, thanks for your patience.

dpkg -S /usr/lib/x86_64-linux-gnu/libjack.so.0 | sed -e 's|:.*|-dbgsym|'

gives

libjack-jackd2-0-dbgsym

Following https://wiki.ubuntu.com/Debug%20Symbol%20Packages
the variable $(lsb_release -cs) gives me "sonya" on Mint 18.2, had to 
change this manually to xenial.
Thinking of not installing Mint the next time or going for something 
like LMDE, debian or arch... anyway, that's OT.
However, I have the KXStudio repositories installed, so the standard 
Ubuntu debug symbols don't help me, because they are for a different 
version. I should try downgrading jack, maybe that already solves the issue.



 ulimit -a
core file size  (blocks, -c) 0
data seg size   (kbytes, -d) unlimited
scheduling priority (-e) 0
file size   (blocks, -f) unlimited
pending signals (-i) 30618
max locked memory   (kbytes, -l) unlimited
max memory size (kbytes, -m) unlimited
open files  (-n) 1024
pipe size(512 bytes, -p) 8
POSIX message queues (bytes, -q) 819200
real-time priority  (-r) 95
stack size  (kbytes, -s) 8192
cpu time   (seconds, -t) unlimited
max user processes  (-u) 30618
virtual memory  (kbytes, -v) unlimited
file locks  (-x) unlimited

- is there a difference if you start jackd with/without realtime settings?

no, still segfaults

- what's the output of Pd (on the terminal and/or console) if you use a
different backend: './pd -verbose -rt -alsa'

/pd-0.48-0test3/bin$ ./pd -verbose -rt -alsa
sys_audioapi set to 1
input channels = 2, output channels = 2
Pd-0.48.0 ("test3") compiled 22:35:07 Jul 20 2017
port 5401
TCL_LIBRARY="../lib/tcl/library" TK_LIBRARY="../lib/tk/library"   wish 
"../tcl//pd-gui.tcl" 5401

Waiting for connection request...
... connected
../bin/pd-watchdog
open: /etc/pd/gem.conf: No such file or directory
open: /home/max/.pd/gem.conf: No such file or directory
open: ./gem.conf: No such file or directory


Then switching to jack output in Pd for the fun:

Segmentation fault

And in the jack log:
Fri Jul 21 13:27:43 2017: ERROR: Cannot read socket fd = 28 err = Broken 
pipe

Fri Jul 21 13:27:43 2017: ERROR: Could not read notification result
Fri Jul 21 13:27:43 2017: ERROR: ClientNotify fails name = system 
notification = 0 val1 = 0 val2 = 0
Fri Jul 21 13:27:43 2017: ERROR: NotifyAddClient new_client fails name = 
pure_data

Fri Jul 21 13:27:43 2017: ERROR: Cannot notify add client
Fri Jul 21 13:27:43 2017: ERROR: Cannot create new client
Fri Jul 21 13:27:43 2017: ERROR: Cannot write socket fd = 26 err = 
Broken pipe

Fri Jul 21 13:27:43 2017: ERROR: CheckRes error
Fri Jul 21 13:27:43 2017: ERROR: JackRequest::ClientOpen write error 
name = pure_data



- does it crash if you start Pd without jackd running?

no:

./pd
Cannot connect to server socket err = No such file or directory
Cannot connect to server request channel
jack server is not running or cannot be started
JackShmReadWritePtr::~JackShmReadWritePtr - Init not done for -1, 
skipping unlock
JackShmReadWritePtr::~JackShmReadWritePtr - Init not done for -1, 
skipping unlock

open: /etc/pd/gem.conf: No such file or directory
open: /home/max/.pd/gem.conf: No such file or directory
open: ./gem.conf: No such file or directory

- is there a difference if you start jackd with/without memory locking?
how?



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[IOhannes m zmölnig]

On 07/21/2017 12:51 PM, IOhannes m zmölnig wrote:
>>
>> better?
>>
> somewhat, but not much.
> 

apart from that:
- what's the output of 'ulimit -a'
- is there a difference if you start jackd with/without realtime settings?
- is there a difference if you start jackd with/without memory locking?
- does it crash if you start Pd without jackd running?
- why don't you get the "Cannot lock down XXX byte memory area
(Operation not permitted)" error (on the console) that i get?
- what's the output of Pd (on the terminal and/or console) if you use a
different backend: './pd -verbose -rt -alsa'


gfmards
IOhannes



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[IOhannes m zmölnig]

On 07/21/2017 11:28 AM, Max wrote:
> On 2017년 07월 21일 07:43, IOhannes m zmölnig wrote:
>> On 07/20/2017 10:40 PM, Max wrote:
>>>
>>>
>>> Thread 3 "pd" received signal SIGSEGV, Segmentation fault.
>>> [Switching to Thread 0x77eb1700 (LWP 21699)]
>>> 0x778b8f96 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
>>> (gdb) bt
>>> #0  0x778b8f96 in ?? () from
>>> /usr/lib/x86_64-linux-gnu/libjack.so.0
>>> #1  0x778b99c5 in ?? () from
>>> /usr/lib/x86_64-linux-gnu/libjack.so.0
>>> #2  0x77893fc8 in ?? () from
>>> /usr/lib/x86_64-linux-gnu/libjack.so.0
>>> #3  0x778bbb6a in ?? () from
>>> /usr/lib/x86_64-linux-gnu/libjack.so.0
>>> #4  0x778b6aa0 in ?? () from
>>> /usr/lib/x86_64-linux-gnu/libjack.so.0
>>> #5  0x7766c6ba in start_thread (arg=0x77eb1700) at
>>> pthread_create.c:333
>>> #6  0x76e953dd in clone () at
>>> ../sysdeps/unix/sysv/linux/x86_64/clone.S:109
>>>
>>> is that helpful or should I try one of these options?
>>
>>
>> no, it's not helpful :-(
>> please install the debug package(s) for libjack,
>> and read the section about backtraces for multithreaded programs.
> 
> I see.
> 
> 
> 
> 
> (gdb) run
> Starting program: /path/pd-0.48-0test3/bin/pd
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
> [New Thread 0x77f32700 (LWP 27779)]
> [New Thread 0x77eb1700 (LWP 27781)]
> 
> Thread 3 "pd" received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 0x77eb1700 (LWP 27781)]
> 0x778b8f96 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> 
> (gdb) thread apply all bt
> Thread 3 (Thread 0x77eb1700 (LWP 27781)):
> #0  0x778b8f96 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> #1  0x778b99c5 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> #2  0x77893fc8 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> #3  0x778bbb6a in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> #4  0x778b6aa0 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> #5  0x7766c6ba in start_thread (arg=0x77eb1700)
> at pthread_create.c:333
> #6  0x76e953dd in clone ()
> at ../sysdeps/unix/sysv/linux/x86_64/clone.S:109
> 
> Thread 2 (Thread 0x77f32700 (LWP 27779)):
> #0  pthread_cond_wait@@GLIBC_2.3.2 ()
> at ../sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185
> #1  0x778b78cc in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> #2  0x778aa828 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> #3  0x778b6aa0 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
> #4  0x7766c6ba in start_thread (arg=0x77f32700)
> at pthread_create.c:333
> #5  0x76e953dd in clone ()
> at ../sysdeps/unix/sysv/linux/x86_64/clone.S:109
> 
> 
> better?
> 

somewhat, but not much.

> #1  0x778b78cc in ?? () from
/usr/lib/x86_64-linux-gnu/libjack.so.0

this basically tells us, that it is crashing somewhere in libjack, but
the actual place/function in which this is happening is unknown ("??").

i think, you are still missing the debug packages of libjack.
you need to enable to debugsym repository [1], and then install  the
"libjack-jackd2-0-dbgsym" package (if libjack-jackd2-0 provides your the
libjack.so).
to you are unsure about the libjack.so provider, install the package
printed out by this:

dpkg -S /usr/lib/x86_64-linux-gnu/libjack.so.0 | sed -e 's|:.*|-dbgsym|'

fgmdsr
IOhannes

[1] https://wiki.ubuntu.com/Debug%20Symbol%20Packages




signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Roman Haefeli]

On Fre, 2017-07-21 at 12:16 +0200, Max wrote:
> On 2017년 07월 21일 11:35, Roman Haefeli wrote:
> > 
> > On Don, 2017-07-20 at 22:30 +0200, Max wrote:
> > > 
> > > On 2017년 07월 20일 22:19, IOhannes m zmölnig wrote:
> > 
> > > 
> > > libjack-jackd2-0 2:1.9.11~20161209-1~xenial5 is what I am
> > > running.
> > libjack-jackd2-0 1.9.10+20150825git1ed50c92~dfsg-1ubuntu1
> > 
> > This is on Ubuntu 16.04.
> > 
> > Last time I had Pd segfault (it segfaulted only wit -jack), the
> > list
> > of linked libraries helped troubleshooting the issue:
> > 
> > [...]
> 
> sorry about my ignorance Roman, but could you explain _how_ it
> helped?
> 
> [...]

Sorry for not being more clear. My Pd was not linked against jack,
although it was compiled with Jack support. As soon as I used '-jack'
or switched to 'Jack' through menu, it crashed. But your Pd _is_ linked
to libjack, so this is not the culprit.

Roman





signature.asc
Description: This is a digitally signed message part
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Icon Update?

[Dan Wilcox]

After tweaking (fixing inlet & rounded corners), the SVG output looks very 
similar: https://www.flickr.com/photos/danomatika/36062245585/in/datetaken/ 


The only difference is that I increased the font size a bit in the first 
version in order to make it more legible at smaller sizes. I think it's safe to 
say any exported icon will need to be tweak a little.

Also, I'm not doing all of this because I want *my icon* or whatever to be used 
at some point. I'm more trying to get feedback on what the community would like 
and we can choose from some proposals. The main thing is that we should keep 
the source files for the icon and it needs to be generated for all the 
platforms (I have this set up).

> On Jul 21, 2017, at 12:31 PM, Dan Wilcox  wrote:
> 
> The SVG output looks just like my tweak minus the tweaks, mainly the outlet 
> box is not tall enough and the object corners are rounded. Otherwise the 
> proportions are the same. It does save the step of needing the ES to SVG 
> conversion.
> 
>> On Jul 21, 2017, at 12:10 PM, Dan Wilcox > > wrote:
>> 
>> 
>>> On Jul 21, 2017, at 12:03 PM, Dan Wilcox >> > wrote:
>>> 
 Your icon version looks a bit slimmer (Font and lines of the box) then 
 what I got. I'm not sure why this is, it might be either because you are 
 on Mac OS, or because you used the PostScript export.
>> 
>> 
>> Also, not that I took the original output and tweaked it in Inkscape. I did 
>> not take the ES -> SVG and use that directly. This also how I created the 
>> PdParty logo: http://danomatika.com/code/pdparty 
>>  The idea was the keep the Pd 
>> proportions as much as possible.
>> 
>> I didn't know about the svg export plugin, I can see if it works on my 
>> system.


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[IOhannes m zmölnig]

On 07/21/2017 12:16 PM, Max wrote:
> sorry about my ignorance Roman, but could you explain _how_ it helped?
> 


mostly because libjack.so wasn't there at all.
there was some problem with underlinking/weak-symbols, which became
(somewhat) apparent by looking at the output of ldd.

gasmdr
IOhannes



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Icon Update?

[Dan Wilcox]

The SVG output looks just like my tweak minus the tweaks, mainly the outlet box 
is not tall enough and the object corners are rounded. Otherwise the 
proportions are the same. It does save the step of needing the ES to SVG 
conversion.

> On Jul 21, 2017, at 12:10 PM, Dan Wilcox  wrote:
> 
> 
>> On Jul 21, 2017, at 12:03 PM, Dan Wilcox > > wrote:
>> 
>>> Your icon version looks a bit slimmer (Font and lines of the box) then what 
>>> I got. I'm not sure why this is, it might be either because you are on Mac 
>>> OS, or because you used the PostScript export.
> 
> 
> Also, not that I took the original output and tweaked it in Inkscape. I did 
> not take the ES -> SVG and use that directly. This also how I created the 
> PdParty logo: http://danomatika.com/code/pdparty 
>  The idea was the keep the Pd proportions 
> as much as possible.
> 
> I didn't know about the svg export plugin, I can see if it works on my system.


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 11:35, Roman Haefeli wrote:

On Don, 2017-07-20 at 22:30 +0200, Max wrote:

On 2017년 07월 20일 22:19, IOhannes m zmölnig wrote:



libjack-jackd2-0 2:1.9.11~20161209-1~xenial5 is what I am running.

libjack-jackd2-0 1.9.10+20150825git1ed50c92~dfsg-1ubuntu1

This is on Ubuntu 16.04.

Last time I had Pd segfault (it segfaulted only wit -jack), the list
of linked libraries helped troubleshooting the issue:

$ ldd ~/.local/bin/pd linux-vdso.so.1 =>  (0x7ffd9c33a000) 
libasound.so.2 => /usr/lib/x86_64-linux-gnu/libasound.so.2

(0x7fa015b57000) libjack.so.0 =>
/usr/lib/x86_64-linux-gnu/libjack.so.0 (0x7fa01590e000) 
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0

(0x7fa0156f) libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2
(0x7fa0154ec000) libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6
(0x7fa0151e3000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6
(0x7fa014e18000) librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1
(0x7fa014c1) /lib64/ld-linux-x86-64.so.2
(0x555ed511d000) libstdc++.so.6 =>
/usr/lib/x86_64-linux-gnu/libstdc++.so.6 (0x7fa01488e000) 
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1

(0x7fa014677000)



sorry about my ignorance Roman, but could you explain _how_ it helped?

path/pd-0.48-0test3/bin$ ldd ./pd
linux-vdso.so.1 =>  (0x7ffd55d15000)
libasound.so.2 => /usr/lib/x86_64-linux-gnu/libasound.so.2
(0x7fa456d28000)
libjack.so.0 => /usr/lib/x86_64-linux-gnu/libjack.so.0 
(0x7fa456ad3000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0
(0x7fa4568b5000)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x7fa4566b1000)
libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x7fa4563a8000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7fa455fdd000)
librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x7fa455dd5000)
/lib64/ld-linux-x86-64.so.2 (0x55bdcba37000)
libdbus-1.so.3 => /lib/x86_64-linux-gnu/libdbus-1.so.3 
(0x7fa455b89000)
libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6
(0x7fa455806000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 
(0x7fa4555f)
libsystemd.so.0 => /lib/x86_64-linux-gnu/libsystemd.so.0
(0x7fa45556a000)
libselinux.so.1 => /lib/x86_64-linux-gnu/libselinux.so.1
(0x7fa455348000)
liblzma.so.5 => /lib/x86_64-linux-gnu/liblzma.so.5 (0x7fa455126000)
libgcrypt.so.20 => /lib/x86_64-linux-gnu/libgcrypt.so.20
(0x7fa454e44000)
libpcre.so.3 => /lib/x86_64-linux-gnu/libpcre.so.3 (0x7fa454bd4000)
libgpg-error.so.0 => /lib/x86_64-linux-gnu/libgpg-error.so.0
(0x7fa4549c)

___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Icon Update?

[Dan Wilcox]

> On Jul 21, 2017, at 12:03 PM, Dan Wilcox  wrote:
> 
>> Your icon version looks a bit slimmer (Font and lines of the box) then what 
>> I got. I'm not sure why this is, it might be either because you are on Mac 
>> OS, or because you used the PostScript export.


Also, not that I took the original output and tweaked it in Inkscape. I did not 
take the ES -> SVG and use that directly. This also how I created the PdParty 
logo: http://danomatika.com/code/pdparty  
The idea was the keep the Pd proportions as much as possible.

I didn't know about the svg export plugin, I can see if it works on my system.


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Icon Update?

[Dan Wilcox]

> On Jul 21, 2017, at 11:47 AM, pd-list-requ...@lists.iem.at wrote:
> 
> Dan, that looks very similar to what I posted on Pd-dev a few weeks back.
> I agree with your mention of lacking discussion, because there was no reply 
> to my mails either.
> 
> https://lists.puredata.info/pipermail/pd-dev/2017-07/021087.html 
> 
> https://lists.puredata.info/pipermail/pd-dev/2017-07/021084.html 
> 
> 
> The fabulous SVG export plugin [1] from IOhannes not only saves you one step 
> of conversion, it also had better results for me that actually looked like 
> the Pd gui, not some abstract version of it like when using the PostScript 
> output.

Yeah, there were several things put out there but they got buried a bit and not 
everyone saw them which is why I'm bringing this up.

> Your icon version looks a bit slimmer (Font and lines of the box) then what I 
> got. I'm not sure why this is, it might be either because you are on Mac OS, 
> or because you used the PostScript export.

The font size of the patch is 16 to get the proportions right which is why the 
outline seems slimmer. 

> Your version also has not the typical shadow that Mac OS icons seem to have.


This icon is meant for all platforms (some of which do not use shadows) and 
there is no hard requirement for mac icons to have a shadow. My proposal 
reflects the most basic/conservative update in keeping with the current vanilla 
icon.


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Icon Update?

[Max]

Dan, that looks very similar to what I posted on Pd-dev a few weeks back.
I agree with your mention of lacking discussion, because there was no 
reply to my mails either.


https://lists.puredata.info/pipermail/pd-dev/2017-07/021087.html
https://lists.puredata.info/pipermail/pd-dev/2017-07/021084.html

The fabulous SVG export plugin [1] from IOhannes not only saves you one 
step of conversion, it also had better results for me that actually 
looked like the Pd gui, not some abstract version of it like when using 
the PostScript output.


Your icon version looks a bit slimmer (Font and lines of the box) then 
what I got. I'm not sure why this is, it might be either because you are 
on Mac OS, or because you used the PostScript export. Your version also 
has not the typical shadow that Mac OS icons seem to have.


[1] https://puredata.info/downloads/patch2svg-plugin


cheers



On 2017년 07월 21일 03:56, Dan Wilcox wrote:
Here's a prelim stab at updating the Pd icon: 
https://github.com/pure-data/pd-icon


Nothing fancy, simply recreated the current vanilla icon using DejaVu 
Sans Mono from an actual Pd patch print export. Now in *high* 
resolution. This repo generates all icons for Windows, Mac, & Linux (I 
assume): there is also gif that TK can load natively. I believe I got 
the sizing and padding around the icon pretty well (at least on mac).


Not to discount Paul Rankin's mac icon refresh which got this going (and 
is arguably better than the flat vanilla approach I've taken):


* https://github.com/pure-data/pure-data/pull/53 (I dig the original 
simple aesthetic with the shadows)

* https://github.com/pure-data/pure-data/pull/56

The makefile can generate the icon from a 1024 master which can be 
changed from my prelim export.


One issue with the original proposal is that there wasn't really much of 
a discussion overall. Now would be a good time to revisit this. Once 
something is agreed upon, we just need an SVG and 1024x1024 PNG to 
generate the icon sets.
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] [PD-announce] pd 0.48-0test1 released

[Dan Wilcox]

> On Jul 20, 2017, at 6:43 PM, pd-list-requ...@lists.iem.at wrote:
> 
> 1) the very first time you start up Pd, there's no log level set (and 
> therefore you don't get any errors).

This is fixed. When I added log level saving I forget to check the case where 
it wasn't set (in older files) and that defaulted to a level of 0. Whoops.

> 3) when I create [midiin] and [sysexin] I get the error 'windows: not 
> supported' in the console. these objectes used to always work for me. did you 
> change something? unfortunately I don't have any MIDI equipment right now to 
> test it.


FWIW Those are simple prints only for WIN32 in x_midi.c and not the result of 
some dependency check. If things are working, they can probably be safely 
removed. The warning about them not working on Windows can also be removed from 
the midi object help file.


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Roman Haefeli]

On Don, 2017-07-20 at 22:30 +0200, Max wrote:
> On 2017년 07월 20일 22:19, IOhannes m zmölnig wrote:
> > 
> > On 07/20/2017 10:09 PM, Max wrote:
> > > 
> > > It stayed the same with test3. Is anyone else running it fine
> > > with Jack
> > > on Linux? I'd like to compare the setup (jack versions??) to help
> > > narrow
> > > it down.
> > 
> > yes, i'm running Pd with jack without a problem (Debian/sid)
> Jack2?
> 
> libjack-jackd2-0 2:1.9.11~20161209-1~xenial5 is what I am running.


I have:

libjack-jackd2-0 1.9.10+20150825git1ed50c92~dfsg-1ubuntu1

This is on Ubuntu 16.04.

Last time I had Pd segfault (it segfaulted only wit -jack), the list of
linked libraries helped troubleshooting the issue:

$ ldd ~/.local/bin/pd
linux-vdso.so.1 =>  (0x7ffd9c33a000)
libasound.so.2 => /usr/lib/x86_64-linux-gnu/libasound.so.2 
(0x7fa015b57000)
libjack.so.0 => /usr/lib/x86_64-linux-gnu/libjack.so.0 
(0x7fa01590e000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 
(0x7fa0156f)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x7fa0154ec000)
libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x7fa0151e3000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7fa014e18000)
librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x7fa014c1)
/lib64/ld-linux-x86-64.so.2 (0x555ed511d000)
libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6 
(0x7fa01488e000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 
(0x7fa014677000)


Roman


signature.asc
Description: This is a digitally signed message part
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] PD 0.48-0 test3 segfaults with Jack and rt on Linux

[Max]

On 2017년 07월 21일 07:43, IOhannes m zmölnig wrote:

On 07/20/2017 10:40 PM, Max wrote:



Thread 3 "pd" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x77eb1700 (LWP 21699)]
0x778b8f96 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
(gdb) bt
#0  0x778b8f96 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#1  0x778b99c5 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#2  0x77893fc8 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#3  0x778bbb6a in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#4  0x778b6aa0 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#5  0x7766c6ba in start_thread (arg=0x77eb1700) at
pthread_create.c:333
#6  0x76e953dd in clone () at
../sysdeps/unix/sysv/linux/x86_64/clone.S:109

is that helpful or should I try one of these options?



no, it's not helpful :-(
please install the debug package(s) for libjack,
and read the section about backtraces for multithreaded programs.


I see.




(gdb) run
Starting program: /path/pd-0.48-0test3/bin/pd
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x77f32700 (LWP 27779)]
[New Thread 0x77eb1700 (LWP 27781)]

Thread 3 "pd" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x77eb1700 (LWP 27781)]
0x778b8f96 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0

(gdb) thread apply all bt
Thread 3 (Thread 0x77eb1700 (LWP 27781)):
#0  0x778b8f96 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#1  0x778b99c5 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#2  0x77893fc8 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#3  0x778bbb6a in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#4  0x778b6aa0 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#5  0x7766c6ba in start_thread (arg=0x77eb1700)
at pthread_create.c:333
#6  0x76e953dd in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:109

Thread 2 (Thread 0x77f32700 (LWP 27779)):
#0  pthread_cond_wait@@GLIBC_2.3.2 ()
at ../sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185
#1  0x778b78cc in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#2  0x778aa828 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#3  0x778b6aa0 in ?? () from /usr/lib/x86_64-linux-gnu/libjack.so.0
#4  0x7766c6ba in start_thread (arg=0x77f32700)
at pthread_create.c:333
#5  0x76e953dd in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:109


better?


___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] [PD-announce] pd 0.48-0test1 released

[Dan Wilcox]

Testing consensus with DejaVu Sans mono is that normal weight is best across 
platforms for similar rendering. I suppose we'll find out if everyone agrees. 
You can, of course, add -font-weight bold to your startup flags for presenting 
:)

> On Jul 21, 2017, at 10:23 AM, Andy Farnell  wrote:
> 
> Yes! Readability at a distance is really important in class.
> I vote for the heaviest dejavu typeface that will work.


Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] problem with translation : WAS Re: testing some 0.48.0 GUI updates

[cyrille henry]

thanks Baptiste for the review and Dan for the commit!


Le 20/07/2017 à 12:53, Dan Wilcox a écrit :

Ok. I think I have most of those now: 
https://github.com/pure-data/pure-data/pull/116/commits/217faea622f63ef972069bc5e78671092486906e

Also removed the #fuzzy keywords from things which are fixed and added some 
missing colons.


On Jul 20, 2017, at 12:23 PM, baptiste chatel mailto:baptiste.cha...@gmail.com>> wrote:

I'd like to find the time to properly review the french translation of the new 
strings, but alas...

"paramettres" > "paramètres"
"deplacement" > "déplacement"
"CARACTERE MANQUANT" > "CARACTÈRE MANQUANT"
"Charger a partir de" > "Charger à partir de"
"Sauver toutes les Préférences" > "Sauver toutes les préférences"
"Réglage" > "Réglages"
"Nombre de chiffre à afficher:" > "Nombre de chiffres à afficher:"
"Temps du flash (ms):" > "Durée du flash (ms):"
"échele_de_sortie" >"échelle_de_sortie"


2017-07-19 16:06 GMT+02:00 cyrille henry mailto:c...@chnry.net>>:

hello,

I'm sorry, i don't have time, so here is only a diff.

Since I made small modification of the existing translation, I copy this 
mail to the 2 other author of the original translation for verification.

thanks and cheers
c


Le 19/07/2017 à 10:46, Dan Wilcox a écrit :


On Jul 19, 2017, at 10:10 AM, pd-list-requ...@lists.iem.at 
 > wrote:

Le 17/07/2017 à 12:13, Dan Wilcox a écrit :
[...]

* The translation build system has been updated as it wasn't 
automatically refreshing the translation files. For translators, there should 
be new strings to add and other to update (see po/README.txt for details). For 
users, the menu titles should now be translated for those languages which do so.


I'm updating the french translation file, but I've got a problem 
with the line :

msgid "file edit put find media window help"

if I translate this line, the pg GUI will not start.


That line should be removed as it's been replaced by each menu item as 
a separate string. The reason why Pd can't start is that the strings in that 
list are used to call procedures ala pd_start_file, pd_start_edit, etc. 
Changing them means tcl errors early in the startup and nothing seems to 
happen. I figured this out the hard way too :)

should I just ignore this line?


Yes. I will check the Tcl to make sure it shouldn't appear in the 
translations. That might be an artifact to my earlier experiments.

What is the best process to share this file : forking pd on 
sourforce git, changing the file and doing a push request?


Whichever works best for Miller, really. We've been working 
successfully by doing small feature Pos via Github as well. Since it's likely a 
small update, you could generate a patch and send to pd-list as well. I'm 
finding GitHub easiest for my workflow, but use whatever works for you :)


Dan Wilcox
@danomatika 
danomatika.com  >
robotcowboy.com  >




___
Pd-list@lists.iem.at  mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list 






Dan Wilcox
@danomatika 
danomatika.com 
robotcowboy.com 





___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] [PD-announce] pd 0.48-0test1 released

[Andy Farnell]

Yes! Readability at a distance is really important in class.
I vote for the heaviest dejavu typeface that will work.

Also it only just occured to me to say this, so a feature request 
for some future release; It would be great to make a slightly 
stronger visual distinction between objects and message boxes. 
The indent on the right side of message boxes is often misread 
by students who are tired or sitting at the back.  

cheers,
Andy


On Fri, Jul 21, 2017 at 02:50:11AM +, Lucas Cordiviola wrote:
> Based on :
> https://github.com/pure-data/pure-data/wiki/Crossplatform-font-metrics-%26-comparisons
> 
> 
> I think that dejavu is pretty Ok.
> 
> And pdx font metrics is also Ok. This is a major change to everybody, it 
> looks smaller on all Os, but has benefits, one is that patches render the 
> same on Purr-data/Pd-L2ork.
> 
> I don't see any downside on pdx font metrics.
> 
> 
> 
> Mensaje telepatico asistido por maquinas.
> 
> 
> 
> From: Pd-list  on behalf of Dan Wilcox 
> 
> Sent: Thursday, July 20, 2017 6:49 AM
> To: Miller Puckette
> Cc: Pd-List
> Subject: Re: [PD] [PD-announce] pd 0.48-0test1 released
> 
> Hah, funny that never occurred to me. I tried all sorts of rounding with the 
> font metrics discovery.
> 
> It does work, but the default Monaco rendering is too thin and the text is 
> not very readable when the font size is 10 or lower.
> 
> However, I just tried with Deja Vu Sans Mono and it looks *great*. I mean, 
> really. I brought the dejavu branch up to date and it's working with the mac 
> build. We can add Windows font support later.
> 
> On Jul 20, 2017, at 7:25 AM, Miller Puckette 
> mailto:m...@ucsd.edu>> wrote:
> 
> Ok... I think I have a workaround that will allow us to ship with
> TCL/TK 8.5 (at last.)
> 
> I've pushed a compiled Mac app and the source to my webpage and
> git-pushed.
> 
> cheers
> Miller
> 
> On Sun, Jul 16, 2017 at 12:38:44PM -0700, Miller Puckette wrote:
> Sure enough - with TK 8.5, font sizes are no longer integers; the
> "font metrics" command appears to return the next-higher integer
> sizes (thus throwing off box sizes and messing up selecting text within
> boxes).
> 
> I'm not sure what to do - try to adapt Pd to deal with fractional font
> sizes, or find some way to force TK 5 to use integer-sized fonts, or
> just give up and use 8.4 for another cycle...
> 
> Incidentally, 8.4 isn't yet building on my MacOS 10.6... maybe I can just
> include the system-provided one :)
> 
> M
> 
> On Sat, Jul 15, 2017 at 11:22:35PM +0200, Dan Wilcox wrote:
> I've added  a macOS 32bit 0.48.0-test1 build using the normal (old, included) 
> Tk 8.4 wish: http://docs.danomatika.com/pdbuilds/ 
> 
> 
> On Jul 15, 2017, at 11:02 AM, 
> pd-list-requ...@lists.iem.at 
>  wrote:
> 
> I notice it looks quite differently in Mac OS, like there is a new font, but 
> it still seems it is Monaco. But the font size is smaller and the boxes are 
> thinner and wider. A subpatch will also have an extra space to the right 
> after the text/name of the subpatch.
> 
> You are seeing the retina rendering with a newer Tk, which still has those 
> font issues as the rendered box size does not quite fit the font. I haven't 
> been able to track down a fix yet.
> 
> It's weird, though, as this shouldn't be the default yet and Pd should still 
> be using 8.4 which is pixelated on retina screens but has correct object 
> sizing.
> 
> Miller, how did you build the mac distribution app?
> 
> 
> Dan Wilcox
> @danomatika 
> danomatika.com 
> robotcowboy.com 
> 
> 
> 
> 
> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list
> 
> 
> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list
> 
> 
> Dan Wilcox
> @danomatika
> danomatika.com
> robotcowboy.com
> 
> 
> 

> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list



signature.asc
Description: Digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] Icon Update

[jlistshit]

Thanx for the stabs repost, hadn’t seen that before. 
Of those two layouts i like Dan’s version more, as the “Retina” version gets 
too foggy here. I am not using etina, but my eyes like maximum contrast. 

The new font, DejaVu Sans Mono works better in the Retina version, though, as 
it has sufficient weight there to display well at smaller icon sizes. 

Eyes on an overfilled dashboard love this.

Thank you both for working on this,

jayrope
- - -
www.jayrope.com
www.aircushionfinish.com
www.ello.co/jayrope



On Jul 21, 2017, at 6:05, pd-list-requ...@lists.iem.at wrote:

> Message: 1
> Date: Fri, 21 Jul 2017 03:56:16 +0200
> From: Dan Wilcox 
> To: Pd-List 
> Subject: [PD] Icon Update?
> Message-ID: <8cd8c1e9-9c55-44c9-868a-8b3a9002f...@gmail.com>
> Content-Type: text/plain; charset="utf-8"
> 
> Here's a prelim stab at updating the Pd icon: 
> https://github.com/pure-data/pd-icon 
> 
> Nothing fancy, simply recreated the current vanilla icon using DejaVu Sans 
> Mono from an actual Pd patch print export. Now in *high* resolution. This 
> repo generates all icons for Windows, Mac, & Linux (I assume): there is also 
> gif that TK can load natively. I believe I got the sizing and padding around 
> the icon pretty well (at least on mac).
> 
> Not to discount Paul Rankin's mac icon refresh which got this going (and is 
> arguably better than the flat vanilla approach I've taken):
> 
> * https://github.com/pure-data/pure-data/pull/53 
>  (I dig the original simple 
> aesthetic with the shadows)
> * https://github.com/pure-data/pure-data/pull/56 
> 
> 
> The makefile can generate the icon from a 1024 master which can be changed 
> from my prelim export.
> 
> One issue with the original proposal is that there wasn't really much of a 
> discussion overall. Now would be a good time to revisit this. Once something 
> is agreed upon, we just need an SVG and 1024x1024 PNG to generate the icon 
> sets.
> 
> 
> Dan Wilcox
> @danomatika 
> danomatika.com 
> robotcowboy.com 
> 


___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] [PD-announce] pd 0.48-0test1 released

[IOhannes m zmoelnig]

On 2017-07-21 04:50, Lucas Cordiviola wrote:
> 
> I don't see any downside on pdx font metrics.
> 

it seems that i remember dan (or was it miller) pointing to the newly
introduced zoom feature which complicates things a bit more. and dynamic
font managing. and what not. (probably check the archives)
the conclusion for me is: there *are* downsides.


On 2017-07-21 04:50, Lucas Cordiviola wrote:
> This is a major change to everybody,

certainly not.

gmasdr
IOhannes



signature.asc
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list