bolprocessor-devel — Bol Processor developer discussion list

[bp-dev] Time for finishing Csound score

[Bernard B. <ber...@gm...>]

Finally, the "check Trace file" option doesn't work because a file may 
be visible before it is closed — even if it is never closed… It was 
confusing because sometimes bp has enough time to finish the score and 
sometimes csound throws its plate off the table…

So I modified the console to create this "done.txt" file which the 
interface waits for and then deletes.
Creating files as flags in a process is an old strategy I was using when 
programming on archive servers. Probably my young colleagues would smile 
at it! ;-) At least it is not system-dependent.

Bernard

-------

In fact, no need to create "done.txt": I made a loop to wait until the 
Trace file has been created.

This is acceptable only because the PHP interface always calls option 
"--traceout". We could indeed force "bp" to do a trace whenever it needs 
to create output files : Csound, MIDI or images…

For the moment I wouldn't change anything because if you call the 
console via the Terminal, then you will wait physically for sufficient 
time before running Csound. So this would only be a problem for creating 
shell scripts or designing a different interface. But we agree that the 
current PHP interface is used as a "reference" for writing the required 
code, don't we? ;-)

Bernard

-------

Dear Anthony,

The whole chain of producing sound via BP and Csound works, as you can 
see on the attached image. You will notice that it is set for the WAVE 
format because AIFF is not recognized by the HTML audio tag.

However there is a small problem. Csound should wait until BP has 
finished writing the Csound score, otherwise it produces a shorter sound 
file based on part of the score. We can see in the trace that BP was 
interrupted by the running of Csound.

For the time being I have put a sleep(4) instruction to keep PHP waiting 
for 4 seconds after sending its ccommand to the console. With 2 seconds 
it doesn't work, and evidently this delay will depend on the size of the 
score, the speed of the application etc.

Therefore my suggestion is that we instruct the console to create a 
small file, for instance "done.txt" so that the interface waits for the 
file to appear, deletes it and then procedd to Csound (and to the 
display of images).

Indeed there maybe other methods for checking that "bp" is still active 
but I am afraid of system dependency…

I am going to try this on the current "graphics-for-BP3" is it hasn't 
yet been merged. ;-)

Bernard

[bp-dev] Time for finishing Csound score

[Bernard B. <ber...@gm...>]

In fact, no need to create "done.txt": I made a loop to wait until the 
Trace file has been created.

This is acceptable only because the PHP interface always calls option 
"--traceout". We could indeed force "bp" to do a trace whenever it needs 
to create output files : Csound, MIDI or images…

For the moment I wouldn't change anything because if you call the 
console via the Terminal, then you will wait physically for sufficient 
time before running Csound. So this would only be a problem for creating 
shell scripts or designing a different interface. But we agree that the 
current PHP interface is used as a "reference" for writing the required 
code, don't we? ;-)

Bernard

-------

Dear Anthony,

The whole chain of producing sound via BP and Csound works, as you can 
see on the attached image. You will notice that it is set for the WAVE 
format because AIFF is not recognized by the HTML audio tag.

However there is a small problem. Csound should wait until BP has 
finished writing the Csound score, otherwise it produces a shorter sound 
file based on part of the score. We can see in the trace that BP was 
interrupted by the running of Csound.

For the time being I have put a sleep(4) instruction to keep PHP waiting 
for 4 seconds after sending its ccommand to the console. With 2 seconds 
it doesn't work, and evidently this delay will depend on the size of the 
score, the speed of the application etc.

Therefore my suggestion is that we instruct the console to create a 
small file, for instance "done.txt" so that the interface waits for the 
file to appear, deletes it and then procedd to Csound (and to the 
display of images).

Indeed there maybe other methods for checking that "bp" is still active 
but I am afraid of system dependency…

I am going to try this on the current "graphics-for-BP3" is it hasn't 
yet been merged. ;-)

Bernard

[bp-dev] Time for finishing Csound score

[Bernard B. <ber...@gm...>]

Dear Anthony,

The whole chain of producing sound via BP and Csound works, as you can 
see on the attached image. You will notice that it is set for the WAVE 
format because AIFF is not recognized by the HTML audio tag.

However there is a small problem. Csound should wait until BP has 
finished writing the Csound score, otherwise it produces a shorter sound 
file based on part of the score. We can see in the trace that BP was 
interrupted by the running of Csound.

For the time being I have put a sleep(4) instruction to keep PHP waiting 
for 4 seconds after sending its ccommand to the console. With 2 seconds 
it doesn't work, and evidently this delay will depend on the size of the 
score, the speed of the application etc.

Therefore my suggestion is that we instruct the console to create a 
small file, for instance "done.txt" so that the interface waits for the 
file to appear, deletes it and then procedd to Csound (and to the 
display of images).

Indeed there maybe other methods for checking that "bp" is still active 
but I am afraid of system dependency…

I am going to try this on the current "graphics-for-BP3" is it hasn't 
yet been merged. ;-)

Bernard

Re: [bp-dev] Problem running Csound

[Bernard B. <ber...@gm...>]

Yes, everything works inccluding specifying the destination of the AIFF 
file, which for the moment is close to that of the score file. So, we 
have "vina3.sco" and "vina3.aif" side by side.

Currently, all settings are on the top of file "_basic_tasks.php". But 
these are settings for making the PHP interface work. The interface 
verifies which system is used and it can do things differently on MacOS, 
Windows and Linux. I have added the path to Csound:
$csound_path = "/usr/local/bin/";

It will be easy to put some vaariables in a separate config file if more 
customization is required.

Bernard

Anthony Kozar wrote on 28/10/2020 09:23:
> Ah! That explains why no audio file is produced. :)
>
> Have you thought about introducing a configuration file into the PHP 
> interface?  That could provide a place for the user to supply 
> information like the pathnames of their csound and bp executables, the 
> BP help file, the folder to use as the top level data location, etc.
>
> Anthony
>
>
> On 10/28/20, 4:03 AM, Bernard Bel wrote:
>
>> The Apache error log effectively yields:
>>
>>      ./myscript.sh: line 3: csound: command not found
>>
>> This means that "csound" should be included in some "path"… I'll look at
>> it later today!
>>
>> Bernard
>
>
> _______________________________________________
> bolprocessor-devel mailing list
> bol...@li...
> https://lists.sourceforge.net/lists/listinfo/bolprocessor-devel

Re: [bp-dev] Problem running Csound WORKS!

[Bernard B. <ber...@gm...>]

You are right! This command worked:

$command = "/usr/local/bin/csound newvina.orc ../my_output/out.sco"

It did produce the AIFF file. The rest wiill be a piece of cake :-)

Bernard

--------

I had checked php.ini to make sure that safe mode is off.

Regardiing the syntax of the command, it is supposed to be correct since 
it works when called from the console. Anyway, I am now trying to call 
"csound -- version" from a script called "myscript.sh".

The Apache error log effectively yields:

    ./myscript.sh: line 3: csound: command not found

This means that "csound" should be included in some "path"… I'll look at 
it later today!

Bernard


Anthony Kozar wrote on 28/10/2020 03:03:

    If the Csound command is producing an error under the PHP interface, then the error output would be in the server's error log, right?  That might explain why you aren't seeing anything other than the echo output.


    Anthony

    ct 27, 2020 6:11 PM, Anthony Kozar wrote: > It's hard to be sure without being able to investigate.  Have you tried using the full path to the Csound binary?  Probably /usr/local/bin/csound. > > I always specify the output file and format too: e.g.  -A -o out.aiff > > Anthony > > On Oct 27, 2020 4:09 PM, Bernard Bel wrote: > > > Hello, > > > > I am now trying to launch Csound immediately after the creation of a > > score (on a Mac). Everything works except… Csound! > > > > The command sent from the PHP page runs as follows: > > > > $command =bsp;"csound newvina.orc ../my_output/out.sco"; > > exec($command,$result_csound); > > > > The AIFF file is not created and no message is returned. > > > > If I run the same command "csound newvina.orc ../my_output/out.sco" from > > the shell, it works fine. > > > > I have tried a few tricks such as "sudo" to convince Csound that it > > should accept being driven by exec(). > > > > Then I wrote a shell script: > > > > ----- > > #! /bin/bash > > echo 'The first message' > > csound newvina.orc ../my_output/out.sco > > echo 'The second message' > > ----- > > > > and the command was changed for: > > > > $command = "./myscript.sh"; > > > > It returns the results of "echo": > > > > The first message > > The second message > > > > but again Csound doesn't run. > > > > The script runs if launched  "./myscript.sh" from the shell. > > > > I thought this might be caused by the size of messsages Csound is > > producing, so I replaced the command with "csound --version" but again > > it failed. > > > > This might have something to do with owner/permissions but I can't find > > a way of breaking the wall ;-) > > > > All suggestions welcome! > > > > Bernard > > > > _______________________________________________ > > bolprocessor-devel mailing list > > bol...@li... > > https://lists.sourceforge.net/lists/listinfo/bolprocessor-devel

Re: [bp-dev] Problem running Csound

[Anthony K. <mai...@an...>]

Ah!  That explains why no audio file is produced. :)

Have you thought about introducing a configuration file into the PHP 
interface?  That could provide a place for the user to supply 
information like the pathnames of their csound and bp executables, the 
BP help file, the folder to use as the top level data location, etc.

Anthony


On 10/28/20, 4:03 AM, Bernard Bel wrote:

> The Apache error log effectively yields:
>
>      ./myscript.sh: line 3: csound: command not found
>
> This means that "csound" should be included in some "path"… I'll look at
> it later today!
>
> Bernard

Re: [bp-dev] Problem running Csound

[Anthony K. <mai...@an...>]

I just ran a test here(*) and it appears that all Csound messages are 
output to standard error.  So, I don't think that you will get any text 
back when running it from PHP.

However, there is an option --logfile=FNAME that writes all of the 
output messages to a file.

I still don't understand why you are not getting an output audio file 
though.  The web server probably runs as a different user so its PATH or 
other environment variables could be different.  The server error log 
will hopefully help...

BTW, in the Csound version I have installed, --version is not a valid 
option although the output is pretty close to what you would expect from 
such an option:

0dBFS level = 32768.0
Csound version 5.14 (float samples) Oct 10 2011
libsndfile-1.0.21
unknown long option: '--version'

- Anthony

(*) I used Csound 5.14 which is pretty old now.  I haven't installed any 
newer versions.


On 10/27/20, 10:03 PM, Anthony Kozar wrote:

> If the Csound command is producing an error under the PHP interface,
> then the error output would be in the server's error log, right? That
> might explain why you aren't seeing anything other than the echo output.

Re: [bp-dev] Problem running Csound

[Bernard B. <ber...@gm...>]

I had checked php.ini to make sure that safe mode is off.

Regardiing the syntax of the command, it is supposed to be correct since 
it works when called from the console. Anyway, I am now trying to call 
"csound -- version" from a script called "myscript.sh".

The Apache error log effectively yields:

    ./myscript.sh: line 3: csound: command not found

This means that "csound" should be included in some "path"… I'll look at 
it later today!

Bernard


Anthony Kozar wrote on 28/10/2020 03:03:
> If the Csound command is producing an error under the PHP interface, then the error output would be in the server's error log, right?  That might explain why you aren't seeing anything other than the echo output.
>
> Anthony
>
> ct 27, 2020 6:11 PM, Anthony Kozar wrote: > It's hard to be sure without being able to investigate.  Have you tried using the full path to the Csound binary?  Probably /usr/local/bin/csound. > > I always specify the output file and format too: e.g.  -A -o out.aiff > > Anthony > > On Oct 27, 2020 4:09 PM, Bernard Bel wrote: > > > Hello, > > > > I am now trying to launch Csound immediately after the creation of a > > score (on a Mac). Everything works except… Csound! > > > > The command sent from the PHP page runs as follows: > > > > $command = "csound newvina.orc ../my_output/out.sco"; > > exec($command,$result_csound); > > > > The AIFF file is not created and no message is returned. > > > > If I run the same command "csound newvina.orc ../my_output/out.sco" from > > the shell, it works fine. > > > > I have tried a few tricks such as "sudo" to convince Csound that it > > should accept being driven by exec(). > > > > Then I wrote a shell script: > > > > ----- > > #! /bin/bash > > echo 'The first message' > > csound newvina.orc ../my_output/out.sco > > echo 'The second message' > > ----- > > > > and the command was changed for: > > > > $command = "./myscript.sh"; > > > > It returns the results of "echo": > > > > The first message > > The second message > > > > but again Csound doesn't run. > > > > The script runs if launched  "./myscript.sh" from the shell. > > > > I thought this might be caused by the size of messsages Csound is > > producing, so I replaced the command with "csound --version" but again > > it failed. > > > > This might have something to do with owner/permissions but I can't find > > a way of breaking the wall ;-) > > > > All suggestions welcome! > > > > Bernard > > > > _______________________________________________ > > bolprocessor-devel mailing list > > bol...@li... > > https://lists.sourceforge.net/lists/listinfo/bolprocessor-devel

Re: [bp-dev] Problem running Csound

[Anthony K. <mai...@an...>]

If the Csound command is producing an error under the PHP interface, then the error output would be in the server's error log, right?  That might explain why you aren't seeing anything other than the echo output.

Anthony

ct 27, 2020 6:11 PM, Anthony Kozar wrote: > It's hard to be sure without being able to investigate.  Have you tried using the full path to the Csound binary?  Probably /usr/local/bin/csound. > > I always specify the output file and format too: e.g.  -A -o out.aiff > > Anthony > > On Oct 27, 2020 4:09 PM, Bernard Bel wrote: > > > Hello, > > > > I am now trying to launch Csound immediately after the creation of a > > score (on a Mac). Everything works except… Csound! > > > > The command sent from the PHP page runs as follows: > > > > $command = "csound newvina.orc ../my_output/out.sco"; > > exec($command,$result_csound); > > > > The AIFF file is not created and no message is returned. > > > > If I run the same command "csound newvina.orc ../my_output/out.sco" from > > the shell, it works fine. > > > > I have tried a few tricks such as "sudo" to convince Csound that it > > should accept being driven by exec(). > > > > Then I wrote a shell script: > > > > ----- > > #! /bin/bash > > echo 'The first message' > > csound newvina.orc ../my_output/out.sco > > echo 'The second message' > > ----- > > > > and the command was changed for: > > > > $command = "./myscript.sh"; > > > > It returns the results of "echo": > > > > The first message > > The second message > > > > but again Csound doesn't run. > > > > The script runs if launched  "./myscript.sh" from the shell. > > > > I thought this might be caused by the size of messsages Csound is > > producing, so I replaced the command with "csound --version" but again > > it failed. > > > > This might have something to do with owner/permissions but I can't find > > a way of breaking the wall ;-) > > > > All suggestions welcome! > > > > Bernard > > > > _______________________________________________ > > bolprocessor-devel mailing list > > bol...@li... > > https://lists.sourceforge.net/lists/listinfo/bolprocessor-devel

[bp-dev] Problem running Csound

[Bernard B. <ber...@gm...>]

Hello,

I am now trying to launch Csound immediately after the creation of a 
score (on a Mac). Everything works except… Csound!

The command sent from the PHP page runs as follows:

$command = "csound newvina.orc ../my_output/out.sco";
exec($command,$result_csound);

The AIFF file is not created and no message is returned.

If I run the same command "csound newvina.orc ../my_output/out.sco" from 
the shell, it works fine.

I have tried a few tricks such as "sudo" to convince Csound that it 
should accept being driven by exec().

Then I wrote a shell script:

-----
#! /bin/bash
echo 'The first message'
csound newvina.orc ../my_output/out.sco
echo 'The second message'
-----

and the command was changed for:

$command = "./myscript.sh";

It returns the results of "echo":

The first message
The second message

but again Csound doesn't run.

The script runs if launched  "./myscript.sh" from the shell.

I thought this might be caused by the size of messsages Csound is 
producing, so I replaced the command with "csound --version" but again 
it failed.

This might have something to do with owner/permissions but I can't find 
a way of breaking the wall ;-)

All suggestions welcome!

Bernard

[bp-dev] Immediate play of MIDI files

[Bernard B. <ber...@gm...>]

Dear lal,

I have just pushed a revision of the PHP interface 
(https://github.com/bolprocessor/php-frontend) in which a link is 
produced immediatly after creating a MIDI file.

This means that if the MIDI file output option is selected, you can 
immediately listen to the result — within the limits of the MIDIjs 
player. For checking structures created by a grammar this is almost as 
good as real MIDI!

In fact we could also use PHP to call Csound as soon as a Csound score 
has been produced. Then we would get a sound file that could appear in a 
HTML5 Audio tag! I can do it on a Mac but the script first needs to 
check that Csound is active on the machine.

Bernard

[bp-dev] News article draft

[Bernard B. <ber...@gm...>]

Dear all,

I am enclosing the draft of an article that we could publish as "latest 
news" on https://bolprocessor.org/category/news/

Please don't hesitate to change the content and the form — make it sound 
more "native English"! ;-)

I am also planning to disseminate it via the LinkedIn social network. I 
have more than 400 professional connections, mostly scholars in 
linguistics with focus on speech prosody. Issues of intonation and 
rhythm in language and music are important for some of them — read for 
instance Maartje Schreuder's thesis:
https://www.academia.edu/5221520/Prosodic_processes_in_language_and_music?email_work_card=title

I have taken the liberty of redesigning a Bol Processor logo for BP3 
(see attached picture). Nothing more than a (less pixelised) remake of 
the old one. My son (a graphic designer) would certainly complain that 
"it sucks" but I don't feel like investing in a graphic chart for the 
future application and its related communication… You may take it as a 
concern for continuity and a choice of priority tasks ;-)

Your ideas are welcome, as usual! :-)

Bernard

[bp-dev] Ready for merging branches :-)

[Bernard B. <ber...@gm...>]

Dear Anthony,

The octave problem is solved. Now, B#4 is C5 and not C4…

Feel free to merge branches. I won't do anything on the code until we 
envisage a new branch.

By the way you will notice that I keep adding/fixing code with my 
personal (old-fashioned) way of placing tabulations. I have tried to 
adopt the more recent style but I cannot concentrate on algorithms. 
Later we could envisage to normalize tabs, which some code editors 
(Xcode?) might do automatically…

Warm regards

Bernard

Anthony Kozar wrote on 26/10/2020 22:23:
> Dear Bernard,
>
> The progress that you have been making is so wonderful!  I'm looking 
> forward to trying out the revised Csound implementation.  The audio 
> files that you sent of Kumar's vina example are quite inspiring. :)
>
> I'm glad that you are posting some Csound examples to the web site. I 
> would be happy to work on creating some additional ones.  Although I 
> haven't done much with Csound in the last 10 years, I don't think the 
> parts relevant to BP have changed very much.  So my in-depth knowledge 
> of the system should still be helpful.
>
> Please do write a news update for the site.  There is so much to 
> report!  Indeed, I believe that with the full support for MIDI files, 
> Csound, and sound objects that BP3 should be very usable for 
> non-real-time composition at this point.  Attracting some attention at 
> this point would not be premature.
>
> I'm sorry that I have contributed very little recently.  These past 
> few weeks I have spent a lot of time reading and listening to the 
> news.  I am feeling anxious about the U.S. election on Nov. 3rd.  Even 
> though I voted with a mail-in ballot several weeks ago, I am eager 
> each day to hear any news that might hint at a "good result" for those 
> of us who are desperate for change.  My anxiety may not be relieved 
> next week though as the results of the election may take longer to 
> determine this year (or be contested!).  :(
>
> I'm going to try again to merge your work into the BP3-develop branch. 
> I didn't finish looking over all of the changes last time.  I noticed 
> a few small things that probably should be modified.  But I may just 
> do the merge as long as the latest code compiles and runs (for me) and 
> reexamine it later.
>
> I'm not sure that I'm going to be much use to the project for awhile 
> yet.  My focus has been so scattered since September started.  If our 
> election contest drags out, I may not be able to concentrate until 
> after the holiday season is over.  I think I will have to be OK with 
> releasing BP 3.0.0 after Jan. 1st.  Does that sound alright?
>
> Anthony
>
>
> On 10/26/20, 12:23 PM, Bernard Bel wrote:
>> Dear all,
>>
>> After documenting the basic usage of Csound for controlling performance
>> parameters on "simple notes" (https://bolprocessor.org/csound-checkup/),
>> I figured out that long ago I had created a documented set of examples
>> dealing with "Csound objects".
>>
>> This was much more complicated… First it was necessary to implement deep
>> changes in the console for a simple reason: in BP2, Csound scores
>> attached to sound-objects would be loaded from the "-mi" sound-object
>> prototypes file and copied to an editable window. Later, BP2 would pick
>> up the text of that window to compile the score… Now we need to directly
>> compile scores received from the "-mi" file and store them in memory.
>> This means that if a score has been found while loading sound-object
>> prototypes, BP3 should interrupt the loading to fetch the "-cs" Csound
>> instruments file and apply the proper transformations. Well, this wasn't
>> easy but finally it works! :-)
>>
>> I have taken all the old examples and reprocessed them to make sure that
>> they produce exactly the same Csound scores with BP3. They do, except a
>> detail: BP2 would sometimes create a table with an incorrect 'x'
>> interval at the end. I mean, for instance the table was declared 256
>> points but the sum of 'x' intervals was larger than 256. This has been
>> fixed. Probably Csound did not worry about this, or the change in sound
>> rendering wasn't audible.
>>
>> All new examples are exposed and documented here:
>>
>> https://bolprocessor.org/csound-objects/
>>
>> It's a presumably complete tutorial to understand algorithms behind the
>> implementation. At least, my memory was refreshed! I checked every
>> example again and again because if we look for "clients" in the Csound
>> community there should be no suspicion of faulty calculations.
>>
>> All examples can be checked with grammar “‑gr.tryCsoundObjects”. I also
>> recommend to update the PHP interface, mostly because I programmed a
>> safe check of values stored in the settings file. I you try to send a A4
>> frequency that is not a positive integer there is no certainty to get a
>> decent Csound ouput!
>>
>> This is the first step: proving Csound geeks that the tool we designed
>> for them is working. Next I wish that we can replace some examples with
>> real musical phrases — as Kumar had done with Sarasvati vina
>> (https://bolprocessor.org/sarasvati-vina/). Tiny "unmusical" examples
>> are necessary for showing where numbers come from, but musical examples,
>> even brief, will serve for convincing skeptics about the interest of
>> dealing with these numbers.
>>
>> I renew my suggestion to publish a News article on work progress. ;-)
>>
>> Bernard
>
>
> _______________________________________________
> bolprocessor-devel mailing list
> bol...@li...
> https://lists.sourceforge.net/lists/listinfo/bolprocessor-devel

Re: [bp-dev] Csound documentation completed

[Bernard B. <ber...@gm...>]

Dear Anthony,

I had guessed your silence had a link with the election issue! It's a 
matter of concern for the world beyond its obvious incidence on the life 
of US citizens. We are also following many debates and documentaries on 
the campaign. There are indeed many other sources of concern, including 
in Europe which is often mirroring the worst aspects of cultural changes 
in North America. To stay away from the vortex of anxiety and fear we 
concentrate on unrelated yet equally important things. This has given me 
energy to focus on programming work.

Please take your time. As you know I never think in terms of deadlines!

Today I would like to solve a little problem which nobody ever figured 
out with BP2. If you enter a simple note "B#4" it is interpreted as "C4" 
instead of "C5". Same issue with "Cb5" that will be replaced with "B5" 
instead of "B4".  Let us admit that using B# or Cb is quite uncommon but 
it may happen in a program that is not bound to any musical style.

It won't take more than one day solving this problem — as I dreamed 
about it in the early morning. ;-) So you may wait for another day 
before merging the branches… Never mind, I'll keep a trace of 
corrections in case you've already done it.

All the best, and best wishes for the near future!

Bernard

Anthony Kozar wrote on 26/10/2020 22:23:
> Dear Bernard,
>
> The progress that you have been making is so wonderful!  I'm looking 
> forward to trying out the revised Csound implementation.  The audio 
> files that you sent of Kumar's vina example are quite inspiring. :)
>
> I'm glad that you are posting some Csound examples to the web site.  I 
> would be happy to work on creating some additional ones.  Although I 
> haven't done much with Csound in the last 10 years, I don't think the 
> parts relevant to BP have changed very much.  So my in-depth knowledge 
> of the system should still be helpful.
>
> Please do write a news update for the site.  There is so much to 
> report!  Indeed, I believe that with the full support for MIDI files, 
> Csound, and sound objects that BP3 should be very usable for 
> non-real-time composition at this point.  Attracting some attention at 
> this point would not be premature.
>
> I'm sorry that I have contributed very little recently.  These past 
> few weeks I have spent a lot of time reading and listening to the 
> news.  I am feeling anxious about the U.S. election on Nov. 3rd.  Even 
> though I voted with a mail-in ballot several weeks ago, I am eager 
> each day to hear any news that might hint at a "good result" for those 
> of us who are desperate for change.  My anxiety may not be relieved 
> next week though as the results of the election may take longer to 
> determine this year (or be contested!).  :(
>
> I'm going to try again to merge your work into the BP3-develop branch. 
> I didn't finish looking over all of the changes last time.  I noticed 
> a few small things that probably should be modified.  But I may just 
> do the merge as long as the latest code compiles and runs (for me) and 
> reexamine it later.
>
> I'm not sure that I'm going to be much use to the project for awhile 
> yet.  My focus has been so scattered since September started.  If our 
> election contest drags out, I may not be able to concentrate until 
> after the holiday season is over.  I think I will have to be OK with 
> releasing BP 3.0.0 after Jan. 1st.  Does that sound alright?
>
> Anthony
>
>
> On 10/26/20, 12:23 PM, Bernard Bel wrote:
>> Dear all,
>>
>> After documenting the basic usage of Csound for controlling performance
>> parameters on "simple notes" (https://bolprocessor.org/csound-checkup/),
>> I figured out that long ago I had created a documented set of examples
>> dealing with "Csound objects".
>>
>> This was much more complicated… First it was necessary to implement deep
>> changes in the console for a simple reason: in BP2, Csound scores
>> attached to sound-objects would be loaded from the "-mi" sound-object
>> prototypes file and copied to an editable window. Later, BP2 would pick
>> up the text of that window to compile the score… Now we need to directly
>> compile scores received from the "-mi" file and store them in memory.
>> This means that if a score has been found while loading sound-object
>> prototypes, BP3 should interrupt the loading to fetch the "-cs" Csound
>> instruments file and apply the proper transformations. Well, this wasn't
>> easy but finally it works! :-)
>>
>> I have taken all the old examples and reprocessed them to make sure that
>> they produce exactly the same Csound scores with BP3. They do, except a
>> detail: BP2 would sometimes create a table with an incorrect 'x'
>> interval at the end. I mean, for instance the table was declared 256
>> points but the sum of 'x' intervals was larger than 256. This has been
>> fixed. Probably Csound did not worry about this, or the change in sound
>> rendering wasn't audible.
>>
>> All new examples are exposed and documented here:
>>
>> https://bolprocessor.org/csound-objects/
>>
>> It's a presumably complete tutorial to understand algorithms behind the
>> implementation. At least, my memory was refreshed! I checked every
>> example again and again because if we look for "clients" in the Csound
>> community there should be no suspicion of faulty calculations.
>>
>> All examples can be checked with grammar “‑gr.tryCsoundObjects”. I also
>> recommend to update the PHP interface, mostly because I programmed a
>> safe check of values stored in the settings file. I you try to send a A4
>> frequency that is not a positive integer there is no certainty to get a
>> decent Csound ouput!
>>
>> This is the first step: proving Csound geeks that the tool we designed
>> for them is working. Next I wish that we can replace some examples with
>> real musical phrases — as Kumar had done with Sarasvati vina
>> (https://bolprocessor.org/sarasvati-vina/). Tiny "unmusical" examples
>> are necessary for showing where numbers come from, but musical examples,
>> even brief, will serve for convincing skeptics about the interest of
>> dealing with these numbers.
>>
>> I renew my suggestion to publish a News article on work progress. ;-)
>>
>> Bernard
>
>
> _______________________________________________
> bolprocessor-devel mailing list
> bol...@li...
> https://lists.sourceforge.net/lists/listinfo/bolprocessor-devel

Re: [bp-dev] Csound documentation completed

[Anthony K. <mai...@an...>]

Dear Bernard,

The progress that you have been making is so wonderful!  I'm looking 
forward to trying out the revised Csound implementation.  The audio 
files that you sent of Kumar's vina example are quite inspiring. :)

I'm glad that you are posting some Csound examples to the web site.  I 
would be happy to work on creating some additional ones.  Although I 
haven't done much with Csound in the last 10 years, I don't think the 
parts relevant to BP have changed very much.  So my in-depth knowledge 
of the system should still be helpful.

Please do write a news update for the site.  There is so much to report! 
  Indeed, I believe that with the full support for MIDI files, Csound, 
and sound objects that BP3 should be very usable for non-real-time 
composition at this point.  Attracting some attention at this point 
would not be premature.

I'm sorry that I have contributed very little recently.  These past few 
weeks I have spent a lot of time reading and listening to the news.  I 
am feeling anxious about the U.S. election on Nov. 3rd.  Even though I 
voted with a mail-in ballot several weeks ago, I am eager each day to 
hear any news that might hint at a "good result" for those of us who are 
desperate for change.  My anxiety may not be relieved next week though 
as the results of the election may take longer to determine this year 
(or be contested!).  :(

I'm going to try again to merge your work into the BP3-develop branch. 
I didn't finish looking over all of the changes last time.  I noticed a 
few small things that probably should be modified.  But I may just do 
the merge as long as the latest code compiles and runs (for me) and 
reexamine it later.

I'm not sure that I'm going to be much use to the project for awhile 
yet.  My focus has been so scattered since September started.  If our 
election contest drags out, I may not be able to concentrate until after 
the holiday season is over.  I think I will have to be OK with releasing 
BP 3.0.0 after Jan. 1st.  Does that sound alright?

Anthony


On 10/26/20, 12:23 PM, Bernard Bel wrote:
> Dear all,
>
> After documenting the basic usage of Csound for controlling performance
> parameters on "simple notes" (https://bolprocessor.org/csound-checkup/),
> I figured out that long ago I had created a documented set of examples
> dealing with "Csound objects".
>
> This was much more complicated… First it was necessary to implement deep
> changes in the console for a simple reason: in BP2, Csound scores
> attached to sound-objects would be loaded from the "-mi" sound-object
> prototypes file and copied to an editable window. Later, BP2 would pick
> up the text of that window to compile the score… Now we need to directly
> compile scores received from the "-mi" file and store them in memory.
> This means that if a score has been found while loading sound-object
> prototypes, BP3 should interrupt the loading to fetch the "-cs" Csound
> instruments file and apply the proper transformations. Well, this wasn't
> easy but finally it works! :-)
>
> I have taken all the old examples and reprocessed them to make sure that
> they produce exactly the same Csound scores with BP3. They do, except a
> detail: BP2 would sometimes create a table with an incorrect 'x'
> interval at the end. I mean, for instance the table was declared 256
> points but the sum of 'x' intervals was larger than 256. This has been
> fixed. Probably Csound did not worry about this, or the change in sound
> rendering wasn't audible.
>
> All new examples are exposed and documented here:
>
> https://bolprocessor.org/csound-objects/
>
> It's a presumably complete tutorial to understand algorithms behind the
> implementation. At least, my memory was refreshed! I checked every
> example again and again because if we look for "clients" in the Csound
> community there should be no suspicion of faulty calculations.
>
> All examples can be checked with grammar “‑gr.tryCsoundObjects”. I also
> recommend to update the PHP interface, mostly because I programmed a
> safe check of values stored in the settings file. I you try to send a A4
> frequency that is not a positive integer there is no certainty to get a
> decent Csound ouput!
>
> This is the first step: proving Csound geeks that the tool we designed
> for them is working. Next I wish that we can replace some examples with
> real musical phrases — as Kumar had done with Sarasvati vina
> (https://bolprocessor.org/sarasvati-vina/). Tiny "unmusical" examples
> are necessary for showing where numbers come from, but musical examples,
> even brief, will serve for convincing skeptics about the interest of
> dealing with these numbers.
>
> I renew my suggestion to publish a News article on work progress. ;-)
>
> Bernard

[bp-dev] Work on memory leakage

[Bernard B. <ber...@gm...>]

Dear all,

I have spent more than two days trying to solve the problem of a 
presumed "memory leakage" displayed at the end of a session on BP3.

I've always been very cautious to obtain the "0 bytes leaked" diagnosis 
in previous versions because it proved that no memory space had been 
left unattended. The method is simple: each time a handle is created or 
resized its size increase is added to a "MemoryUsed" global variable, or 
substracted when the handle is disposed of.

This has worked for decades (!) but suddenly became faulty when loading 
sound-object prototypes. Worse, it goes bad even if no sound-object 
prototypes file is invoked, probably because BP2/BP3 creates by default 
2 sound-objects for silences '-' and empty events '_' (prolongations).

At first view it seemed easy to compensate for the creation of these 
sound-objects, but apparently the difficulty is that this creation takes 
place before calling Inits(), and we start counting memory 
creation/deletion after Inits().

I have tried dozens of solutions to get back to my obsessive "0 bytes 
leaked" but could not succeed. It really drove me mad ;-) However, in 
the end I admit that the "leak" remains very small and it can even be 
anticipated when no "-mi" file is loaded.

Most important, it happens only once in the session. The problem of 
memory leakage would be that of a continuous work of the console, for 
instance non-stop improvizing music in real-time MIDI or reacting to 
MIDI input in a collective improvisation setup. But as we see it now, 
the leakage won't cumulate. We will check that it doesn't when 
implementing repetitive tasks. So, the topic is not yet solved but it 
may be kept pending with no incidence on future work. :-)

I have added maybe a hundred memory notices and flags to the console's 
code, all activated by a single global variable "check_memory_use". I 
suggest that we keep them for further checking when implementing new 
features.

I also feel sorry for confusing between "ShowMessage()", "Alert1()" and 
"BPPrintMessage()" that seem to produce the same output visible on the 
PHP interface. I used "BPPrintMessage()" for most new notices. This may 
be fixed if necessary, rethinking about what should go where. Currently 
the console is very "talkative" and it might be worth having a "verbose" 
option if messages need to be switched off. Even, maybe, different 
levels of talk from silence to debugging… I confess having spent no time 
figuring out a better combination…

I also many questions writtent by Anthony on the source code. I have not 
taken time to look for answers but they will certainly be addressed 
"when time permits"!

Bernard

[0 bytes leaked]

[bp-dev] Csound documentation completed

[Bernard B. <ber...@gm...>]

Dear all,

After documenting the basic usage of Csound for controlling performance 
parameters on "simple notes" (https://bolprocessor.org/csound-checkup/), 
I figured out that long ago I had created a documented set of examples 
dealing with "Csound objects".

This was much more complicated… First it was necessary to implement deep 
changes in the console for a simple reason: in BP2, Csound scores 
attached to sound-objects would be loaded from the "-mi" sound-object 
prototypes file and copied to an editable window. Later, BP2 would pick 
up the text of that window to compile the score… Now we need to directly 
compile scores received from the "-mi" file and store them in memory. 
This means that if a score has been found while loading sound-object 
prototypes, BP3 should interrupt the loading to fetch the "-cs" Csound 
instruments file and apply the proper transformations. Well, this wasn't 
easy but finally it works! :-)

I have taken all the old examples and reprocessed them to make sure that 
they produce exactly the same Csound scores with BP3. They do, except a 
detail: BP2 would sometimes create a table with an incorrect 'x' 
interval at the end. I mean, for instance the table was declared 256 
points but the sum of 'x' intervals was larger than 256. This has been 
fixed. Probably Csound did not worry about this, or the change in sound 
rendering wasn't audible.

All new examples are exposed and documented here:

https://bolprocessor.org/csound-objects/

It's a presumably complete tutorial to understand algorithms behind the 
implementation. At least, my memory was refreshed! I checked every 
example again and again because if we look for "clients" in the Csound 
community there should be no suspicion of faulty calculations.

All examples can be checked with grammar “‑gr.tryCsoundObjects”. I also 
recommend to update the PHP interface, mostly because I programmed a 
safe check of values stored in the settings file. I you try to send a A4 
frequency that is not a positive integer there is no certainty to get a 
decent Csound ouput!

This is the first step: proving Csound geeks that the tool we designed 
for them is working. Next I wish that we can replace some examples with 
real musical phrases — as Kumar had done with Sarasvati vina 
(https://bolprocessor.org/sarasvati-vina/). Tiny "unmusical" examples 
are necessary for showing where numbers come from, but musical examples, 
even brief, will serve for convincing skeptics about the interest of 
dealing with these numbers.

I renew my suggestion to publish a News article on work progress. ;-)

Bernard

[bp-dev] Csound complete cheeckup

[Bernard B. <ber...@gm...>]

Dear all,

Being puzzled (and annoyed) at a discrepency between BP2 and BP3 
regarding the Sarasvati vina example, I returned to a set of examples 
which I had designed in those happy days to check every detail of 
algorithms implemented for the production of Csound scores. The data set 
was on "-da.tryCsound" and I renamed it "-da.checkAllCsound".

I ran all examples on both BP2 and BP3, luckily getting exactly the same 
outputs. Since these examples are informative — at least they reminded 
me of the approach! — I converted them to a tutorial:

https://bolprocessor.org/csound-checkup/

So, at the moment the only unsolved trouble is the vina example when 
performed on BP2.  When time permits I will try to understand why it 
goes wrong. Launching a DiffMerge between the scores indicate that 
differences lie mainly in the contruction of tables.

Bernard

[bp-dev] Csound small discrepency between BP2 and BP3

[Bernard B. <ber...@gm...>]

Dear all,

Yesterday you listened to Vina3.mp3 made from -gr.vina3 on BP3.

Today I compared it with the same production on BP2.9.8 and I saw/heard 
a small difference.

If you compare the sound recordings, they sound similar except that 
around t = 10 seconds the BP2 version includes a nasty sound ressembling 
the frictions that guitar players produce while performing. ;-) This 
would be impossible on a Sarasvati vina — and afficionados would be 
horrified! So, evidently we have an artifact which can be verified 
comparing the (enclosed) score files.

I have not been able to understand why this happened. I am not too 
worried about it because the correct version is on BP3! Still, it 
requires some investigation into the way interpolation tables are 
constructed (the CsoundMaths.c file, pfff…) because it might be the 
trace of a bug. Csound experts are welcome to find the culprit!

Some explanations in chapter 17 of https://bolprocessor.org/misc/docs/ 
will probably be helpful. I am also going to check all documented 
examples in -da.tryCsound.

Bernard

[bp-dev] Csound + News

[Bernard B. <ber...@gm...>]

Dear all,

Following my discovery of a long-forgotten and impressive example of 
Csound/Bol Processor design by Srikumar Karaikudi (in 1995), I 
documented it on a separate article:

https://bolprocessor.org/sarasvati-vina/

I expect that we'll soon be able to expose constructions of equal 
complexity inspired by various musical cultures. This should be a key to 
join the Csound community and invite more adventurers!

At this stage I believe that it would be appropriate to publish a News 
article. The most recent one dates back to July this year:

https://bolprocessor.org/towards-a-multiplatform-bol-processor/

We need to show that we're still alive and the implementation of BP3 is 
in good shape. Announce notably that MIDIfile export, graphics, 
sound-object prototyping and the full support of Csound have been 
achieved. Include links to the temporary BP3 documentation 
(https://bolprocessor.org/check-bp3/) and maybe other articles like the 
former one.

This News article could also be disseminated on the BP open discussion 
forum and shared on social networks. Again, there is a very small 
probability that a software designer joins the project before the 
official release of BP3, but we should tell that the doors are open and 
convince people that our project is not dormant…

Shall I write a draft and submit it here?

Bernard

[bp-dev] Csound in full swing!

[Bernard B. <ber...@gm...>]

Dear all,

Listen to the enclosed MP3  files. These have been entirely created by 
Bol Processor + Csound… :-)

This was work by my colleague Srikumar Subramanian. In this time (circa 
1995) he was already an accomplished Carnatic musician (he plays the 
Sarasvati vina) and an wizard in maths/physics! He had picked up Csound 
at the stage I started getting lost in figures and calculations. ;-)

I pushed to "bp3-ctests" the "-gr.vina3" grammar, Csound instruments 
file, orchestra file and sampled waveform of Sarasvati vina. I am also 
enclosing scores and orchestra files with this message so you can try to 
reprocess them with Csound, as I did.

I confess that I don't understand anything of these scores! I vaguely 
remember that BP was smart enough to construct tables (the f101, f102 
etc. lines) helping in the interpolation — i.e. the "slide" parameter 
attached to the "vina" instrument — but I have completely forgotten the 
process… :-(

Never mind, it works. I would like to publish more examples of 
sophisticated Csound works engineered by Bol Processor. I remember 
having played with an impressive Csound algorithm producing the sound of 
a guitar in a few lines of codes… This would be cute on Western 
classical examples.

Bernard

[bp-dev] Csound implementation is complete

[Bernard B. <ber...@gm...>]

PS. In the implementation I had forgotten the copy of Csound tables on 
top of the score!

; Csound score
f1 0 256 10 1
f2 0 256 10 1 0.5 0.3 0.4

Now fixed. I will push the correct version as soon as my Internet 
connection works again — presumably when you receive this message. ;-)

Bernard

-------


Dear all,

I am delighted to announce the completion of dealing with Csound!

I programmed the loading of "-cs" Csound instrument files which wasn't 
difficult because of similarity with "-mi" Sound-object prototypes.

But things went bad when errors showed up in the CsoundMaths.c file! The 
regressions weren't working… In fact this was caused by the value of a 
"very large number" indicating that a numeric field is empty. (Otherwise 
it can be negative or positive…)

The "almost infinite" number used on the PHP interface is 2147483647 = 
2^31 - 1. But on the console we were using Infpos which is calculated 
with LONG_MIN, i.e. 9223372036854775807, indeed system dependent.

I feel relieved not going back to these regression calculations!

You can run an example with a command line such as:

../bp produce -gr ../ctests/-gr.tryCsound -se ../ctests/-se.tryCsound 
-cs ../ctests/-cs.tryCsound -d --csoundout ../my_output/out.sco 
--traceout trace.txt --seed 0

The item is:

_ins(3) _volumecont _volume(127) A4 G4 C5 A5 A4 G4 C5 A5 _volume(0)

The resulting score is:

t 0.000 60.000
i3 0.000 1.000 440.00 8191.500 8191.500 24.000 21.688 0.000 0.000 0.000 
0.000 0.000 ; A4
i3 1.000 1.000 392.00 8191.500 8191.500 21.688 18.431 0.000 0.000 0.000 
0.000 0.000 ; G4
i3 2.000 1.000 523.25 8191.500 8191.500 18.431 12.899 0.000 0.000 0.000 
0.000 0.000 ; C5
i3 3.000 1.000 880.00 8191.500 8191.500 12.899 -0.031 0.000 0.000 0.000 
0.000 0.000 ; A5
i3 4.000 1.000 440.00 8191.500 8191.500 -0.031 -1.179 0.000 0.000 0.000 
0.000 0.000 ; A4
i3 5.000 1.000 392.00 8191.500 8191.500 -1.179 -2.794 0.000 0.000 0.000 
0.000 0.000 ; G4
i3 6.000 1.000 523.25 8191.500 8191.500 -2.794 -5.547 0.000 0.000 0.000 
0.000 0.000 ; C5
i3 7.000 1.000 880.00 8191.500 8191.500 -5.547 -24.000 0.000 0.000 0.000 
0.000 0.000 ; A5

It shows the effect of Csound instrument 3 (Harpsichord) which is 
translating MIDI volume control (0 .. 127) to (-24 .. +24) on parameters 
7 and 8 of each event. Because of _volumecont, the volume is also 
changing gradually within each event. That's the great thing MIDI would 
never achieve…

In addition, the mapping from MIDI value to Csound parameters is set by 
a quadratic regression, given the points defined on instrument 
"Harpsichord". Imagine that if the same is done with pitchbend any sort 
of glissando can be designed. Srikumar Subramanian had started playing 
with a melodic movement of the sarasvati veena.

I admit that the whole thing would not make sense with a harpsichord. 
Names of instruments have been attributed arbitrarily ;-)

If you want to complete the experiment, set up the same volume mapping 
on "Flute" (intrument 2), replace _ins(3) with _ins(2) in the grammar, 
and… design a Csound orchestra able to play a flute with these 
parameters! ;-)

Bernard

[bp-dev] Csound implementation is complete

[Bernard B. <ber...@gm...>]

Dear all,

I am delighted to announce the completion of dealing with Csound!

I programmed the loading of "-cs" Csound instrument files which wasn't 
difficult because of similarity with "-mi" Sound-object prototypes.

But things went bad when errors showed up in the CsoundMaths.c file! The 
regressions weren't working… In fact this was caused by the value of a 
"very large number" indicating that a numeric field is empty. (Otherwise 
it can be negative or positive…)

The "almost infinite" number used on the PHP interface is 2147483647 = 
2^31 - 1. But on the console we were using Infpos which is calculated 
with LONG_MIN, i.e. 9223372036854775807, indeed system dependent.

I feel relieved not going back to these regression calculations!

You can run an example with a command line such as:

../bp produce -gr ../ctests/-gr.tryCsound -se ../ctests/-se.tryCsound 
-cs ../ctests/-cs.tryCsound -d --csoundout ../my_output/out.sco 
--traceout trace.txt --seed 0

The item is:

_ins(3) _volumecont _volume(127) A4 G4 C5 A5 A4 G4 C5 A5 _volume(0)

The resulting score is:

t 0.000 60.000
i3 0.000 1.000 440.00 8191.500 8191.500 24.000 21.688 0.000 0.000 0.000 
0.000 0.000 ; A4
i3 1.000 1.000 392.00 8191.500 8191.500 21.688 18.431 0.000 0.000 0.000 
0.000 0.000 ; G4
i3 2.000 1.000 523.25 8191.500 8191.500 18.431 12.899 0.000 0.000 0.000 
0.000 0.000 ; C5
i3 3.000 1.000 880.00 8191.500 8191.500 12.899 -0.031 0.000 0.000 0.000 
0.000 0.000 ; A5
i3 4.000 1.000 440.00 8191.500 8191.500 -0.031 -1.179 0.000 0.000 0.000 
0.000 0.000 ; A4
i3 5.000 1.000 392.00 8191.500 8191.500 -1.179 -2.794 0.000 0.000 0.000 
0.000 0.000 ; G4
i3 6.000 1.000 523.25 8191.500 8191.500 -2.794 -5.547 0.000 0.000 0.000 
0.000 0.000 ; C5
i3 7.000 1.000 880.00 8191.500 8191.500 -5.547 -24.000 0.000 0.000 0.000 
0.000 0.000 ; A5

It shows the effect of Csound instrument 3 (Harpsichord) which is 
translating MIDI volume control (0 .. 127) to (-24 .. +24) on parameters 
7 and 8 of each event. Because of _volumecont, the volume is also 
changing gradually within each event. That's the great thing MIDI would 
never achieve…

In addition, the mapping from MIDI value to Csound parameters is set by 
a quadratic regression, given the points defined on instrument 
"Harpsichord". Imagine that if the same is done with pitchbend any sort 
of glissando can be designed. Srikumar Subramanian had started playing 
with a melodic movement of the sarasvati veena.

I admit that the whole thing would not make sense with a harpsichord. 
Names of instruments have been attributed arbitrarily ;-)

If you want to complete the experiment, set up the same volume mapping 
on "Flute" (intrument 2), replace _ins(3) with _ins(2) in the grammar, 
and… design a Csound orchestra able to play a flute with these 
parameters! ;-)

Bernard

Re: [bp-dev] Completion of loading and using sound-object prototypes

[Bernard B. <ber...@gm...>]

Dear all,

I have spent time checking various combinations of graphic display 
(objects or pianoroll) and output formats (MIDI file, Csound files, BP 
data).

If the item contains sound-obects, by default it will be displayed as 
"objects". But you can also force it to show a pianoroll (of the same 
content). This was not possible on BP2.9.8.

Attached are the 2 versions of

a b <<f>> a a <<chik>> a b

taking its sound-object prototypes in -mi.abc1

The pianoroll mode shows what we expect to hear whereas the object mode 
shows the structure. For instance, "a" and "b" are "relocatable", as 
indicated by their triangle pivots, whereas <<f>> and <<chick>> aren't 
because their pivots are complete arrows.

You can see that time is metronomic and striated, but the third and 
fourth occurrences of "a" are not placed on the streak. In fact, "a" has 
a topological "continuity" property forcing it to move back and stick to 
the preceding object.

<<f>> and <<chick>> are "out-time" objects (null duration). On the 
pianoroll they only appear as dots on F2, C3, F3, C4.

On the contrary, if you take a piece such as -gr.Nadaka and force it to 
display "objects", you will get a very complicated image that will 
certainly not reflect the melodic/harmonic structure.

You can play with the same alphabet "-ho.abc1" and its set of 
sound-object prototypes "-mi.abc1" trying for instance to use periodic 
objects "cycle1" and "cycle2" that will repeat their sequence if forced 
to a longer duration.

Lots of fun already :-)

Bernard

Re: [bp-dev] Completion of loading and using sound-object prototypes

[Bernard B. <ber...@gm...>]

Dear Anthony,

I have just pushed the console and interface updates implementing the 
creation of image files directly readable with a standard browser.

I followed only part of your suggestion because I need to check that a 
javascript function would be executed on a pop-up window. Earlier I had 
experienced that functions would be blocked by security settings of the 
web browser. When time permits I will try your "function" solution as it 
might permit more options.

Meanwhile I believe that this version of the branch could be merged. 
Later I may work exclusively on Graphic.c.

I've spent almost the whole day struggling with string operations in C. 
A nightmare after many years enjoying the PHP environment!  Notably this 
function for replacing substrings (in "Strings.c") which I finally 
borrowed from a programmer.

All the best

Bernard

Anthony Kozar wrote on 18/10/2020 04:27:
> I don't know if this email will display correctly for you since the 
> content includes HTML tags and scripts.
>
> On 10/17/20, 7:20 PM, Bernard Bel wrote:
>> In fact the console could also write the temporary image file with its
>> unknown values, then read and copy it to the final file once the values
>> have been calculated. So the prototype.php script won't be required to
>> finalize the file. I will do it quickly if you think it's worth!
>
> No need.  I may have a better solution.  Could you write the 
> JavaScript like this instead?
>
> <script>
> function drawImage1(ctx) {
> // This image was created by Bol Processor BP3
> ctx.scale(0.5,0.5);
> ctx.lineWidth = 1;
> ctx.font = "56px arial";
> ctx.lineWidth = 1;
> ctx.beginPath();
> ...
> }
>
> // WRITE THIS PART AT THE END WHEN THE WIDTH AND HEIGHT ARE KNOWN
> window.onload = function() {
>   const canvas = document.querySelector('.myCanvas');
>   const width = canvas.width = 8736;
>   const height = canvas.height = 404;
>   const ctx = canvas.getContext('2d');
>   drawImage1(ctx);
> };
> </script>
>
> (I was going to suggest that you use the Canvas scale() function to 
> make the graphics smaller but I see that you already found it. :)
>
> Another suggestion that I have is to enclose the canvas in a DIV tag 
> with a fixed width so that it gets its own scroll bar, like this:
>
> <div class="canvasWrapper">
> <canvas class="myCanvas">
> <p>Your browser does not support CANVAS… :-(</p>
> </canvas>
> </div>
>
> And add this to the <style> tag:
>
> .canvasWrapper {
>   width: 100%;
>   overflow-x: scroll;
> }
>
> This makes it so that the canvas scrolls by itself which will be 
> useful if additional content is added to the HTML such as a header 
> labels, links, and/or additional canvas images.
>
> Rather than writing multiple images out to separate files, would you 
> want to combine them  all in one file? Something like this?
>
> <body>
> <h1>Image 1</h1>
> <div class="canvasWrapper">
>  <canvas id="myCanvas-1">
>   <p>Your browser does not support CANVAS… :-(</p>
>  </canvas>
> </div>
> <script>
> function drawImage1(ctx) { ... }
> </script>
>
> <h1>Image 2</h1>
> <div class="canvasWrapper">
>  <canvas id="myCanvas-2">
>   <p>Your browser does not support CANVAS… :-(</p>
>  </canvas>
> </div>
> <script>
> function drawImage2(ctx) { ... }
> </script>
>
> ......
>
> <script>
> window.onload = function() {
>   canvas = document.querySelector('#myCanvas-1');
>   canvas.width = 8736;
>   canvas.height = 404;
>   ctx = canvas.getContext('2d');
>   drawImage1(ctx);
>
>   canvas = document.querySelector('#myCanvas-2');
>   canvas.width = 3000;
>   canvas.height = 300;
>   ctx = canvas.getContext('2d');
>   drawImage2(ctx);
>
>   ...
> };
> </script>
> </body>
>
>
> I've attached a modified HTML image with some of these changes.
> Interested to know what you think of these possibilities.
>
> Anthony
>
>
>
>
> _______________________________________________
> bolprocessor-devel mailing list
> bol...@li...
> https://lists.sourceforge.net/lists/listinfo/bolprocessor-devel