Welcome, Guest. Please login or register.
Amiga Kit Amiga Store Iridium Banner AMIStore App Store A600 Memory

AuthorTopic: The Os 3.1.4 Thread  (Read 80863 times)

0 Members and 3 Guests are viewing this topic.

Offline my_pc_is_amiga

Re: The Os 3.1.4 Thread
« Reply #660 on: August 28, 2019, 01:39:51 AM »
This will change for 3.2 where I touched the dos.library to redirect PrintFault() through stderr (rather than stdout) and where most application programs in C: where also modified to use stderr consistently for error output.

Thanks for the peak preview of 3.2 :).   If there are some changes being done for the C: programs, then there could be a slight improvement for Eval error detection...

Eval '? LFORMAT="%X*N";  this is not producing an error as it should be.  The %X is not correct and should be %X2 or some number after %x.  Instead in 3.1/3.1.4, eval prints something but doesn't make sense (*N doesn't get correctly parsed).  In 4.x, the error is produced saying in affect that LFORMAT is not in correct format.
 

Offline Gulliver

Re: The Os 3.1.4 Thread
« Reply #661 on: August 28, 2019, 03:48:12 AM »
@Thomas Richter

It seems that my_pc_is_amiga has scored a point!

 ;)

I've posted too many things and so not sure which one you are referring to :)

Thomas was awarding virtual points to any user guessing some of the new features that 3.2  brings. You guessed one: a working stderr.

And it seems you are now near another, this one is very small. ;-)

It is up to Thomas to "spill the beans".
 

Offline my_pc_is_amiga

Re: The Os 3.1.4 Thread
« Reply #662 on: August 28, 2019, 05:46:23 AM »

Thomas was awarding virtual points to any user guessing some of the new features that 3.2  brings. You guessed one: a working stderr.

And it seems you are now near another, this one is very small. ;-)

It is up to Thomas to "spill the beans".

Mmm...AmigaGuide text search (though I don't think that one is small).   Seems like there was some kind of search in v34.6   And the other "wish" is a fix for the high res. pointer slanting on native amiga screens (and that one surely isn't an easy fix).   This has been there since 3.0...
« Last Edit: August 28, 2019, 06:08:29 AM by my_pc_is_amiga »
 

Offline Thomas Richter

Re: The Os 3.1.4 Thread
« Reply #663 on: August 28, 2019, 06:20:19 AM »
Mmm...AmigaGuide text search (though I don't think that one is small). 
It is not a big one. We have a text datatype with search functionality, but did not use it since it was dependend on reaction which we did not have.

And the other "wish" is a fix for the high res. pointer slanting on native amiga screens (and that one surely isn't an easy fix).   This has been there since 3.0...
I've no idea what this means.
 

Offline kolla

Re: The Os 3.1.4 Thread
« Reply #664 on: August 28, 2019, 08:06:51 AM »
Thomas was awarding virtual points to any user guessing some of the new features that 3.2  brings. You guessed one: a working stderr.

If pointing out well known limitations and bugs in OS 3.1(.4(.1)) count as guessing new features for 3.2, then there are plenty of "virtual points" to be earned. Thomas has himself documented the issue with lacking support for stderr rather extensively in his Shell.guide for Shell v45 on aminet - this guide, with updates for v46 and future v47, should also have come as part of OS release.

http://aminet.net/package/util/boot/ShellUpdate
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC
 

Offline Gulliver

Re: The Os 3.1.4 Thread
« Reply #665 on: August 28, 2019, 11:49:16 AM »
  And the other "wish" is a fix for the high res. pointer slanting on native amiga screens (and that one surely isn't an easy fix).   This has been there since 3.0...

Could you please try to describe with more detail this issue you find?
 

Offline Thomas Richter

Re: The Os 3.1.4 Thread
« Reply #666 on: August 28, 2019, 03:54:51 PM »
If pointing out well known limitations and bugs in OS 3.1(.4(.1)) count as guessing new features for 3.2, then there are plenty of "virtual points" to be earned. Thomas has himself documented the issue with lacking support for stderr rather extensively in his Shell.guide for Shell v45 on aminet - this guide, with updates for v46 and future v47, should also have come as part of OS release.

http://aminet.net/package/util/boot/ShellUpdate
There will be an update of this guide as well. Actually, there already is one which describes the new commands, the new arguments of old commands, and the recent and less recent changes to the shell. So yes, we're well aware of the need to document the upcoming changes 3.2 will bring, including a new SDK that is also under preparation. It is just a slow going process.

Online outlawal2

Re: The Os 3.1.4 Thread
« Reply #667 on: August 28, 2019, 06:28:43 PM »
Forgive me if this is a silly question or if this has been answered already, but will 3.2 require another new ROM or will it work with 3.1.4?
And thanks for all of your hard work!  Sometimes it doesn't appear that folks appreciate the work you are doing but I can assure you MANY of us do..

 

Offline Thomas Richter

Re: The Os 3.1.4 Thread
« Reply #668 on: August 28, 2019, 06:39:37 PM »
Forgive me if this is a silly question or if this has been answered already, but will 3.2 require another new ROM or will it work with 3.1.4?
And thanks for all of your hard work!  Sometimes it doesn't appear that folks appreciate the work you are doing but I can assure you MANY of us do..
I don't know, as I don't have to decide. This is a management decision. At this point, the project is run in such a way that we keep all options open.
 

Offline Thomas Richter

Re: The Os 3.1.4 Thread
« Reply #669 on: August 28, 2019, 07:34:14 PM »
Eval '? LFORMAT="%X*N";  this is not producing an error as it should be.  The %X is not correct and should be %X2 or some number after %x.  Instead in 3.1/3.1.4, eval prints something but doesn't make sense (*N doesn't get correctly parsed).  In 4.x, the error is produced saying in affect that LFORMAT is not in correct format.
I'm not even sure what it should print in your expectation. "*N" in quotes expands to the line feed. This is expected and desired. If you want an asterisk within quotes, you need to escape the BCPL escape character by an escape character, which is the asterisk itself.


 

Offline kolla

Re: The Os 3.1.4 Thread
« Reply #670 on: August 28, 2019, 10:12:38 PM »
Part of the "issue" here is that in this case, *N does not expand to linefeed (but ***N does).


Edit: But after a little experimenting, this looks more like confusion over what %X is and how it works - lformat "%X1*n" works as expexted, with "%X*n" %X apparently just "swallows" the * and the n, while "%X *n" (space betweem X and *) does not. What's correct behaviour here is probably debatable, but I agree that it would be best to exit with error about bad template.

So, will C:Info and C:Date get LFORMAT too? And maybe C:Assign?
« Last Edit: August 28, 2019, 10:46:43 PM by kolla »
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC
 

Offline kolla

Re: The Os 3.1.4 Thread
« Reply #671 on: August 28, 2019, 10:43:20 PM »
(Delete - double post)
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC
 

Offline Thomas Richter

Re: The Os 3.1.4 Thread
« Reply #672 on: August 29, 2019, 02:24:53 AM »
Part of the "issue" here is that in this case, *N does not expand to linefeed (but ***N does).
Huh? %X is not part of an LFORMAT template. So list just prints that, for every file found. *N is expanded to a line feed as it is in double quotes. Every line feed in backticks is replaced by a blank. Thus, `list LFORMAT="%X N"` returns the pattern "%X" repeated n times, where n is the number of files in the directory. Every %X is separated from every other %X by blanks. Of course, "eval" cannot compute anything from that, so it prints an error.

I'm not sure what is the defective part here. That looks all exactly as it should look.
 

Offline my_pc_is_amiga

Re: The Os 3.1.4 Thread
« Reply #673 on: August 29, 2019, 04:50:03 AM »
Edit: But after a little experimenting, this looks more like confusion over what %X is and how it works - lformat "%X1*n" works as expexted, with "%X*n" %X apparently just "swallows" the * and the n, while "%X *n" (space betweem X and *) does not. What's correct behaviour here is probably debatable, but I agree that it would be best to exit with error about bad template.

OS4 reports an error about a bad LFROMAT while OS3.x does not.   %X is not the correct format, it needs to be %X1, or %X2, etc. According to AmigaDOS manual, "The %X and %O options require a number of digits specification (for example %X8 gives 8 digits of hex output)."  So my expectation was to have EVAL produce an error.   
 

Offline my_pc_is_amiga

Re: The Os 3.1.4 Thread
« Reply #674 on: August 29, 2019, 04:51:32 AM »
And thanks for all of your hard work!  Sometimes it doesn't appear that folks appreciate the work you are doing but I can assure you MANY of us do..

I copy that.   Yes, thanks for all the hard work!