Re: [MSEide-MSEgui-talk] MSEcoupon ready for testing

Hello Martin. Before to jump into your demo, I have a question... Does your demo use espeak-ng as executable or as a library ? Sorry to insist with this but it is a matter of license. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] MSEcoupon ready for testing

> The demo is GPL, so it can use the eSpeakNG library. Huh, Martin,... you did not answer to the question... So, if I understand ok, the demo uses eSpeakNG-library directly, not via eSpeakNG-executable. If it is the case, do you have plan, like we did for sak, to also provide a "

[MSEide-MSEgui-talk] Cross compile from Intel 64 Linux to ARM 64 Linux ?

Hello everybody. Is is possible to cross-compile a library (Java native) from a Intel 64 Linux machine for target ARM 64 Linux ? I did compile/test the library (Java native) on a Rpi Linux ARM 32 and all ok (loaded from a Java class). But how to compile it for a ARM 64 ? The goal is to use the

Re: [MSEide-MSEgui-talk] MSEcoupon ready for testing

>> >What is it? It doesn't look like a win32 binary. It is a win32 appliation, compiled with fpc 3.1.1. (via wine on Linux Mint). It works "out of the box" on all the windows os tested. Included screen capture of

Re: [MSEide-MSEgui-talk] MSEcoupon ready for testing

> Windows applications also need the "*.exe" extension Ooops, the extension is missing, sorry and fixed. > Ah, I see. 7zip probably did not unpack it correct. OK, included a traditional zip file. demo_sak_mse_win32.zip

Re: [MSEide-MSEgui-talk] Cross compile from Intel 64 Linux to ARM 64 Linux ?

> You compile with FPC (Free Pascal) a library written in Java? No, I compile Pascal source with FPC, to obtain a Java native library. The goal is that a Java class use that library. It works for OS (Win, Linux, FreeBSD) + Java installed on Intel 32/64. For ARM 32/64 with Linux and FreeBSD it

Re: [MSEide-MSEgui-talk] MSEcoupon ready for testing

>> Also, using high definition of voice-synth (disponible when compiling the >> library) does not have any difference with the "standard" one. > I don't understand, please explain. There is a option at compiling to choose the audio driver. And if you choose portaudio, you may choose float

Re: [MSEide-MSEgui-talk] MSEcoupon ready for testing

> The eSpeakNG executable is not well suited for the purpose. What are you missing ? I did use also espeak library and did not find exported methods useful that executable can not manage. The only thing that I noted is that all was slower (yes, trust me), problems for user to install libraries

[MSEide-MSEgui-talk] MSE and ./configure

Hello. There is a **very** high number of different Unix distros, widget-set, window-manager, etc,... Same for Windows versions. So, could it not be possible to have a king of ./configure (same idea that used for compiling C programs) loaded at begin or before the app, that will check all the

Re: [MSEide-MSEgui-talk] Fwd: Updated mseconsts_es.pas

Hello Martin. Attached updated mesconsts_fr.pas mseconsts_fr.pas Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check

Re: [MSEide-MSEgui-talk] Problem with last commit.

Hello Martin. With last commit, nothing compile anymore ;-( sak_mse.pas(64,10) Error: No matching implementation for interface method "doapplicationactivated;" found sak_mse.pas(64,10) Error: No matching implementation for interface method "doapplicationdeactivated;" found

Re: [MSEide-MSEgui-talk] Problem with last commit.

> Thereason is that I currently make a tassistivserver component for > applications which are specifically designed for visually impaired people. Ha, earlier tassistivserver was not able to help visually impaired people ? Could you explain, please, what does the new tassistivserver that earlier

Re: [MSEide-MSEgui-talk] Problem with last commit.

> iassistiveserver has been extended, simply add according dummy functions to > your class, clientmouseevent() has been renamed to doclientmouseevent() > for > consistency, it has been mentioned in VERSION.TXT. Ouf, ok, fixed and it works like before (you scared me). Re-by the way, I am

Re: [MSEide-MSEgui-talk] How to use form-scrollbars ?

> container.frame.scrollpos_x:= -3*parentformwidth. wavescroll_strump.mp4 ;-) Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] How to use form-scrollbars ?

Hello Martin. Goal. Here the tips: 1) Recompile MSEide / ideU with last MSEgui commits. 2) In object inspector, select the TSlider and set ow1_noclampinview := true. 3) msescrollpage_strumOK.mp4 Wow,

[MSEide-MSEgui-talk] How to use form-scrollbars ?

Hello Martin. I did try to change position of scrollbar of a form by code. Without luck. Using theform.container.frame.sbhorz.pageup works only once. Howto do if the button of a TSlider goes outside the form, how to , by code, do a page-up of the scrollbar. This to have the button of the

[MSEide-MSEgui-talk] TTimer.options := [to_single] ?

<https://github.com/fredvs/strumpract> ), you can test it that way: In waveform.pas, change the option of ttimer1 to: TTimer.options := []; > on my rpi ---> problems, cannot zoom anymore + lot of ressource used. I do not want to annoy you with this because the problem seems to

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Hello Sieghard. Thanks for answer. OK, all is fixed now, compies + link + run 32 bit applications on a 64 bit multi-arch system. This patch for the fpc compiler did the trick (this for fpc 3.1.1): fpc_checkelf2.gz In

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Hello. > Sorry, I didn't notice that you use Manjaro, Huh, after some bad experience with my ex-favorite Linux OS, I just have switched to a not-Debion OS: Arch-Manjaro. > I thought it's about PolyDev. With polYdev it is a other problem: FreeBSD multi-arch design is not compatible with their

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Hello everybody. > Sorry, I didn't notice that you use Manjaro Huh, after some bad experience with my ex-favorite Linux OS, I just migrate to a not-Debian OS : Arch-Manjaro. >, I thought it's about PolyDev. No with polYdev multi-arch Linux-emulated works like charm (the problem is only with

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Hello Martin. > Because in linker script are absolute paths, see my mail from 2018-07-23: Huh, OK, I did read it but the code for the linker script is in t_linux.pas line 508 (see my post of Jul 24, 2018; 2:41pm) StartSection('INPUT('); if linklibc and (libctype<>uclibc) then begin

Re: [MSEide-MSEgui-talk] fpc and memory leak.

Thanks Marcos. > When the process finish, the OS clean all memory that belongs it. Sorry, I do not understand. Do you mean that the OS clean all memory leak after fpc is closed? What OS ? Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

[MSEide-MSEgui-talk] fpc and memory leak.

Hello. Referencing to this: http://lists.freepascal.org/pipermail/fpc-devel/2018-July/039369.html So, if I understand ok, after few compiling, we have to reboot the system because fpc has eaten lot of memory because of their memory leaks ? Huh, will MSElang have the same behaviour ? Fre;D

Re: [MSEide-MSEgui-talk] fpc and memory leak.

@ Marcos and Douglas: thanks for your light. > However, if your application is just a console app that do the job and > dies, as FPC, then it will be the same. Huh, AFAIK MSE and fpGUI apps are console apps too ? Anyway, in my old time, playing with programing gives you a red card when your

Re: [MSEide-MSEgui-talk] fpc and memory leak.

>That are their thinking... All that story gives me, one more time, a taste of disrespect on the part fpc core developers. And,sorry, but for me respect is paramount. Thanks for your positive view Marcos. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] Multi-arch problem...

> Maybe librarysearchpath has an entry for /lib where on your system a 64 bit crtn.o is located? It is exactly wath I was thinking but... I did not found this in FPC code. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Re-hello. > Maybe librarysearchpath has an entry for /lib where on your system a 64 > bit crtn.o is located? Re-about this, even if librarysearchpath has an entry for /lib (that I did not found), normally, like for all other libraries, it seems to me that there is a check if the file has the

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Hello. > Sorry, I didn't notice that you use Manjaro, Huh, after some bad experience with my ex-favorite Linux OS, I just have switched to a not-Debian OS: Arch-Manjaro. > I thought it's about PolyDev. With polYdev it is a other problem: FreeBSD multi-arch design is not compatible with their

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Ha, the mails work again. ;) For all the passionates about ELF multi-arch, here how to check if a file is a ELF and what kind. From Wikipedia, ELF format: ... Read the first four bytes. If they are equal to \x7fELF, it's an ELF file. The ELF magic number is 0x7f454C46 (0x7f + "ELF") The

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Hello Sieghard. Thanks for your time. Huh, yes, using 32 bit apps on multi-arch 64 bit system can be ***very*** painful. At the moment I can successfully run fpc32 (on a 64 bit os) and compile 32 bit applications. Yesterday, I was even able to run those 32 applications on the multi-arch os !

Re: [MSEide-MSEgui-talk] MSElang news.

> Currently MSElang is for 32 bit Linux only. Ha, ok, a good test for Manjaro 64 bit multi-arch. > it needs LLVM version 3.8.0. Here, with: $ sudo pacman -S lib32-llvm It install llvm-6.0.1-1. Maybe llvm-6.0.1 is compatible with llvm-3.8.0. I will try with this first. Thanks. Fre;D

Re: [MSEide-MSEgui-talk] MSElang news.

Hello Martin. Ooops, out-of-the-box... $ /home/fred/mselang-master/mselang/compiler/mlc --llvm MSElang Compiler version 0.0 Copyright (c) 2013-2018 by Martin Schreiber Fatal: No input file defined Ok, I will try with some input Huh, mlc is compiled with fpc, could it be compiled later

Re: [MSEide-MSEgui-talk] MSElang news.

>It is not ready for testing. OK, I stop. By the way, congrats for what you have done. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community on one of the

Re: [MSEide-MSEgui-talk] MSElang news.

Hello Martin. I did try to compile mselang.prj And get this error message: Compiling mselang.pas ... Compiling globtypes.pas Compiling ./compiler/__mla__internaltypes.pas __mla__internaltypes.pas(40,12) Error: Identifier not found "card32" __mla__internaltypes.pas(40,18) Error: Error in type

Re: [MSEide-MSEgui-talk] stringgrid / fixcols

> I cannot figure out how to set a value into the "fixcols" data cell. I use this: thegrid.fixcols[-1].captions[5] := inttostr(5); Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Re-hello. OK, I get a work-around. in /msegui/lib/common/fpccompatibility/mclasses.pas At line 6573 + 6574, comment this: // raise EReadError.CreateFmt(SPropertyException, // [Name, DotSep, Path, e.Message]); Without those lines, the app 64 bit run OK (no crash). But for the app 32

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Aaargh. It seems to be a Manjaro+MSE problem. Compiling a 64 bit app is ok but running it: An unhandled exception occurred at $0060B7BA : EReadError : Error reading configfo.windowopacity: Unknown property: "windowopacity" $0060B7BA line 6571 of

Re: [MSEide-MSEgui-talk] Multi-arch problem...

> tcustommseform.windowopacity has been introduced recently. I assume you try to > load a form produced by current MSEide with an old MSEgui mseforms unit. Yes, it must be the problem. The mseforms unit was created with a OS that use windowopacity. And it seems that Manjaro xfce does not use

Re: [MSEide-MSEgui-talk] Multi-arch problem...

> Please ensure that you use the mseform.pas file from current git master > branch everywhere and ensure that the used mseform.ppu will be recompiled. Of course I did use the last msegui from Gitlab and deleted all the .ppu and .o before compile. Anyway now it is fixed (and I do not know how to

Re: [MSEide-MSEgui-talk] GitLab problems.

Hello Martin. Indeeed, today all 3 links are working. Strange, yesterday there was a message for those 3 links saying that Gitlab was transfering all their data-server using now some Google servers. Sorry for the noise. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] GitLab problems.

Ha, I did find the site where I was redirected: https://about.gitlab.com/2018/06/25/moving-to-gcp/ -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community on one of the

Re: [MSEide-MSEgui-talk] Multi-arch problem...

> -Xd (do not use the standard library path) link.res without Xd parameter: (and of course with correct -e and Fl) (that is what I did) -> SEARCH_DIR("/lib/") SEARCH_DIR("/usr/lib/") SEARCH_DIR("/usr/lib32/") SEARCH_DIR("/usr/lib/gcc/x86_64-pc-linux-gnu/8.1.1/32/") ... INPUT(

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Re-hello. Here a advice that has absolutely no importance. OK, -Xd does the trick because it does not give /lib nor /usr/lib as searh path. But I maintain that there is still a problem with crti.o and crtn.o It is not normal that all other needed libraries are using the correct 32 bit path

[MSEide-MSEgui-talk] GitLab problems.

Hello Martin. What appends with GitLab ? I cannot connect to MSE GitLab site: https://gitlab.com/mseide-msegui/mseide-msegui It is redirected and say it is clouded by Google now... What is the new MSE-Gitlab URL ? Thanks. Fre;D -- Sent from:

[MSEide-MSEgui-talk] Multi-arch problem...

Hello Martin. I try to make my new Linux (Manjaro xfce) 64 bit multi arch work. OK, compiling 32 bit MSE application with fpc32 is OK on the 64 bit OS. But, while running (with gdb) the 32 bit app on the 64 bit system, there is that error + crash: (gdb) run Starting program:

Re: [MSEide-MSEgui-talk] Multi-arch problem...

Hello Seighard. > - considering, that these two _object_ files seem to be multi architacture > in systems with multi architecture support. Not sure to understand. Do you mean that it exist object_ files that have both ELF32+64 signatures ? And then only one file could be used in a multi-arch

Re: [MSEide-MSEgui-talk] Using sta files.

Hello Martin. Thanks for light. I will study/test it. Write you later. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community on one of the world's most engaging

Re: [MSEide-MSEgui-talk] Using sta files.

Hello Martin. Works like charm with only 2 lines of code! msesta.mp4 Many thanks. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] Useless searches all over the Internet

Hello Graeme. Here, it works. (using Opera + GitLab). Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the

Re: [MSEide-MSEgui-talk] Using sta files.

Hello. And here using sta file to create custom playlist... msestagrid.mp4 By the way, do you have some idea (+code) how to move a row, with the mouse, inside the tstringgrid ? Thanks. Fre;D -- Sent from:

Re: [MSEide-MSEgui-talk] Using sta files.

> The grid must have a fixed column. Ha, OK: msegridmoverow.mp4 Excellent ! Thanks. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

[MSEide-MSEgui-talk] Using sta files.

Hello Martin. Is it possible to use sta files to save/load a layout at runtime. The result of using a sta file is perfect while closing-running a program. But is it possible to save at runtime the layout (with a custom name for the sta file) and then reload it when needed ? If all this is

Re: [MSEide-MSEgui-talk] Some change in mseconst.pas...

Hello Martin. In the close confirmation dialog of MSEide, the Button "All" is still there (in place of "Yes all"). See picture: ;-( Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] MSElang news.

Perfect. And for the look, what about the size of a app compiled with LLVM (vs FPC) ? Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community on one of the world's

Re: [MSEide-MSEgui-talk] MSE and ./configure

> I'll try to make a separate pcaudiolib.dll. Yep, super. ;-) And, by the way, you have a new fan (me) for your https://gitlab.com/mseide-msegui/pcaudiolib project. Fre;D PS: pcaudiolib-ng ? ;-) -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

[MSEide-MSEgui-talk] MSE pcaudio.

library. So, it is a **excellent** news that Martin forks this interesting part of espeak-ng and updated the code to make Pcaudio works even on old Windows machines. I did a Pascal wrapper for this library with a fpc sine-wave demo: https://github.com/fredvs/pcaudiolib_pas Martin did his own MSE

Re: [MSEide-MSEgui-talk] MSE and ./configure

> Done, Perfect. I will create a new topic about MSE pcaudio. Many thanks. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community on one of the world's most

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> Are you sure that the program still is OK with your changes? Ok, re-re-re checked. With forced: ch1on.value := true; ch2on.value := true; twochannel:= true; channels := 2; and audio_object_write(faudioobj,p1,bytelen div channels) And with sleep(100) after wavethread.terminate() :

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> The problem with the wrong buffer length is in pcaudiolib/source/src/alsa.c Wow, you are strong. Indeed it is the problem. > It probably should be > #define FORMAT(srcfmt, dstfmt, size) case srcfmt: pcm_format = dstfmt; > self->sample_size = size*channels; break; Yep, it is this. >

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Re-re-re Hello. About the crash of the demo, this solves all on my system: procedure tmainfo.soundoff() at line 220: wavethread.terminate(); wavethread.active:= false; // add this here, not 4 lines later. Ok, I stop to annoy you with this. Fre;D -- Sent from:

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> Is this a workaround or a bug fix? Ho, it is a workaround to test pcaudiolib with your demo. Now for the bug-fix, I will propose something like (if wavethread has freeonterminate): wavethread.terminate(); while wavethread <> nil do sleep(1); Fre;D -- Sent from:

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> It is not pulseaudio but alsa. Ok. I will attack this tomorrow. Many thanks Martin. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community on one of the world's

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> I would vote for a "NumChunck" parameter in audio_object_write() Maybe it will be easier to add it in audio_object_open(). And to prepare the future, add some "TypePut" (Input or Output) parameter too. ;-) Ok, I stop. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> so the pcaudiolib Alsa driver is active. Ha, that is more logic for me (even if I would prefer that pcaudiolib load a "pure" Alsa, but ok, capito, it must be done by PulseAudio). > Agreed? Now 200 % perfectly. ;-) About pcaudiolib with Alsa driver active With a samplerate = 44100 and a

Re: [MSEide-MSEgui-talk] MSE pcaudio.

(the latency time is not accessible in pcaudiolib!) Oooops. ;-( By the way, if some parameter should be added in some method, I would vote for a "NumChunck" parameter iin audio_object_write(pcaudioobj, buffer, bufferlength, numchunck)); With PortAudio, numchunck is the number of samples from

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> I assume if there is no PulseAudio installed, OK, asap, I will un-install PulseAudio and see what appends. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Re-hello Martin. If I may Doing this in procedure tmainfo.onsetev() line 243: channels:= 2; // force 2 channels. (with channels as global variable) // if ch1on.value and ch2on.value then begin // channels:= 2; // end; And this for procedure tmainfo.waveexe() line 127: //

Re: [MSEide-MSEgui-talk] MSE pcaudio.

>About the crash, ok, I will investigate. >(But my nose say it is not a pcaudiolib problem) ... Ok, I trusted my nose and ... in procedure tmainfo.soundoff(), line 215: if oned.value then begin wavethread.terminate(); sleep(100); // Adding this. if faudioobj <> nil then

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> As you like. ;-) Huh, you do not trust me ? main.zip -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Re-hello Martin. Aaargh, always a bémol somewhere... audioobj := create_audio_device_object('default', 'thename', 'theinfo'); > Do not get back to default (pulse), this must be done: audioobj := create_audio_device_object(nil, 'thename', 'theinfo'); PS: This is a very micro-mini bémol.

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Hello Martin. Using this (pure alsa): audioobj := create_audio_device_object('sysdefault', nil, nil); Using this (pure pulseaudio): faudioobj:= create_audio_device_object(nil,pchar(''),pchar(''));

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Aaargh bad image for pulse in previous post, here the good one. Hello Martin. Using this (pure alsa): audioobj := create_audio_device_object('sysdefault', nil, nil); Using this (pure pulseaudio): faudioobj:=

Re: [MSEide-MSEgui-talk] MSE and ./configure

> Can you be more specific? OK, I will try. Here, for example the result of configure for espeak-ng: fred@fred-Lenovo-YOGA-300-11IBY ~/espeak-ng-master $ ./configure checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking for a

Re: [MSEide-MSEgui-talk] MSE and ./configure

> But that is completely unrelated to Free Pascal and MSEgui? Usual Free Pascal > projects don't need a complex build environment like usual gcc projects > do. Please Martin... Of course I know that. The idea was to give a script or program that will check the unix system and give a list of

Re: [MSEide-MSEgui-talk] MSE and ./configure

> OK, you don't want a list of MSEgui features but a list of > Unix/Linux/desktop-environment/graphical-environment - features of a given > installation. Yeeep, you get it. ;) > That is a big task and out of scope of MSEgui I fear. I can't help you > here. OK, no problem, I am a warrior

Re: [MSEide-MSEgui-talk] MSE and ./configure

> Your Ubuntu installation or your repository setup seems to be broken. Maybe ;( Did you try with a up-to-date Debian OS ? : sudo apt-get install espeak-ng Does it install espeak-ng ? > Suggestion: build and install it from source. Yes, I did it already. And after (lot of) tips, espeak-ng

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Re-re-re-... hello. For info about memory used: Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> Then one needs to find a workaround Hello Martin. Till now, I do not get good result using only snd_pcm_drop to stop the stream. I m more lucky (with help of PortAudio code) with snd_pcm_drain. int alsa_object_flush(struct audio_object *object) { struct alsa_object *self =

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> Why do need to change it? Because I did forget to delete the status.sta after a cross-compilation. Sorry for the noise. > The problem with "pure" alsa mode is that Ok, I will study deeply your notes. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> does not drop all buffers. Huh, do you have a idea how to fix this ? Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community on one of the world's most engaging

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> You can test it in mono mode if you change the sine frequency. Indeed ;-( What is strange too is that using snd_pcm_drain() gives strange result too. (In theory it should result only the delay to write all the buffer, but it gives a short extra sound too with alsa) Compiling pcaudiolib with

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> Probably there are not so many responsive speech assistive systems on Linux. Yes but all the other applications that use alsa directly, they did not note this ? But maybe it is a recent new bug from a recent earlier commit. (Or that explains why some people say that pulse has a "better"

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> Do you have plan to commit your workaround: > http://mailman.alsa-project.org/pipermail/alsa-devel/2018-March/133327.html With your workaround (re-compile pcaudio.so with your code) uos works perfectly. No more "click" at begin of sound (if a other was loaded before). No crash when other

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Ooops, I did not see your answer. I will study it. Write you later. Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech community on one of the world's most engaging tech

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Hello Martin. Do you have plan to commit your workaround: http://mailman.alsa-project.org/pipermail/alsa-devel/2018-March/133327.html into https://gitlab.com/mseide-msegui/pcaudiolib ? Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> case AUDIO_OBJECT_FORMAT_S32LE: return >CreateWaveFormatEx(WAVE_FORMAT_PCM, 32, rate, channels, wfmt); Yes, I did see this, it is why the **very little bémol** (PortAudio can deal it). But the most important resolutions (IMHO) are int16 and float32 (and both work perfectly). Thanks.

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> Ok. I will attack this tomorrow. Aaargh, I cannot resist. Bon, ok, re-try with libpcaudio.so,0 compiled with PulseAudio on. Ok, this time PulseAudio is used with your demo (and mine) and indeed "div chan" is not needed anymore for stereo channels. So, yes alsa.c is bugged. Not only the

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Huh, sorry in previous mail, for the date, please read: In fact I know PulseAudio from his (buggy) beginning (2004). Fre;D -- Sent from: http://mseide-msegui-talk.13964.n8.nabble.com/ -- Check out the vibrant tech

Re: [MSEide-MSEgui-talk] MSE pcaudio.

>> Please see attachment. > or this attachment. Huh, sorry but even with clean glasses, it seems to me that "alsa_object_write()" is a call to ALSA. Or I miss, one more time, something. ;-( In fact I know PulseAudio from his (buggy) beginning (2014). And yes, maybe my mind was altered

Re: [MSEide-MSEgui-talk] MSE pcaudio.

sine-wave, I get good result with last parameter of audio_object_write(), in Unix: number of bytes div channels. in Windows: number of bytes. audio_object_write(audioobj, pointer(ps), arlen*sizeof(ps[0]) {$IFDEF unix} div channels {$endif}); https://github.com/fredvs/pcaudiolib_pas/blob/m

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> seems to show that on your system Alsa calls Pulseaudio not the opposite. Hum, difficult to find info about this: https://en.wikipedia.org/wiki/PulseAudio Not lot of infos... I am nearly more than sure that alsa does not know PulseAudio and never call it in his code. But those days, so many

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> /usr/src/debug/pulseaudio-9-0/src/pulse/stream.c:1613 which clearly is a Pulseaudio function. > Agreed? OK. > snd_pcm_ioplug_writei()->snd1_pcm_write_areas()->ioplug_priv_transfer_areas(). > > Agreed? ... ... ... ... G, ..., ok. > https://en.wikipedia.org/wiki/PulseAudio > Thus,

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> pulseaudio.png (163K) Download Attachment Just for confirmation... This was a print-screen of pcaudiolib using alsa or pulseaudio ? (Just now I have doubt, I was thinking it was a print-screen of pcaudiolib using pulseaudio). But maybe it was a print-screen of pcaudiolib using alsa. (And

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> and especially here: > http://mailman.alsa-project.org/pipermail/alsa-devel/2018-March/133351.html Huh, big fan, not yet... ;-) What I find strange is: how could it be possible that this "drop" problem (not release last buffer) appears only now. Alsa is a old project (1998), nobody did note

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Ooops, what appends with the previous message-layout ? Ok, I try again: Hello Martin. OK, I did: # sudo apt-get install libpulse-dev ---> OK, installed. Re-configure (ok, Pulse audio recognized) + make all + sudo make install > OK, libpcaudio.so.0 installed. But now, with the

Re: [MSEide-MSEgui-talk] MSE pcaudio.

>libpcaudio.so.0 binaries are here: > https://sourceforge.net/projects/mseuniverse/files/eSpeakNG/ OK, I did download libpcaudio.so.0 from https://sourceforge.net/projects/mseuniverse/files/eSpeakNG/ Then copy it to /usr/local/lib and (to be sure) in root directory of /mse-pcaudio/ Now, after

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Hello Martin. > An example is here: > https://gitlab.com/mseide-msegui/mseuniverse/tree/master/samples/audio/pcaudio Wow, magistral sine-wave demo. > It is in msepcaudio.pas:115. Indeed, after cleaning the glasses, it appears, sorry for the noise. > It is used in PulseAudio. Hah, ok, >>

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> "alsa" pcaudiolib afaik is buggy. Grr. > Don't you have "pulseaudio" installed? Yes, but it seems that pcaudiolib wants something more at compilation. Something like pulseaudio-dev (Gr x2). And I would prefer a "pure" audio-library that can deal with root sound-drivers, like alsa or

Re: [MSEide-MSEgui-talk] MSE pcaudio.

> For me it is always the buffer size in bytes, please see the extended demo: > https://gitlab.com/mseide-msegui/mseuniverse/tree/master/samples/audio/pcaudio Hum, indeed, i have to re-check my code, something strange appended... On my (maybe corrupted) Linux Mint 17 64 bit system,

Re: [MSEide-MSEgui-talk] MSE pcaudio.

Hello Martin.OK, I did:# sudo apt-get install libpulse-dev ---> OK, installed.Re-configure (ok, Pulse audio recognized) + make all + sudo make install > OK, libpcaudio.so.0 installed.But now, with the mse-pcaudio-demo, after 2xclick on "Sound on" : crash with error message:

<    1   2   3   4   5   6   7   8   9   10   >