Recent Posts

Topic: Community Support / WebJames help

I’ve just started using WebJames (to provide the web GUI for my new heating control system). I can get the simple things to work OK, but I’m having difficulty finding out how to do one or two things.

How do I (or, indeed, can I) specify a cascading stylesheet? I’ve tried this line:

<link rel="stylesheet" type="text/css" href="styles.css">

but all I get is an error message, on the BBxM, that Zap is already running – even when the browser is on a different machine.

At present, styles.css is of type text, but changing it to type CSS simply changes the error message: App to load that filetype not found.

I’ve used that line when generating HTML that wasn’t served by WebJames, and it appears to be understood OK.

WebJames’ log file indicates that the path to the file is correct, i.e. the error is not caused by failing to find the file.

Topic: Code review / wchar in ansilib

Perhaps it is better to use Unicode library?

Most likely, yes.

Wide character encodings were designed to deal with things like Asian characters before the adoption of Unicode – such as Shift-JIS and the like. Ironically wide encodings don’t specify any actual format, only that characters are represented by multiple byte encodings… The size of which is compiler specific.
If you’re thinking a pre-Unicode world was a mess, you’d be right.

Wide character support could, I suppose, be pressed into use for UTF-16, however if I was going to use Unicode on RISC OS, I would use the one most compatible with existing code – namely UTF-8 where a string is still a string in C¹, all you need to understand and support is a break from the idea that 1 byte = 1 character. Luckily the UTF-8 encoding is both simple to understand and gives it’s byte length (normally 1 to 4 bytes, but can be 1-6) as a part of itself.

The challenge, of course, is getting the Wimp to do something sensible with the text. While the machine can be switched to UTF-8 and it all magically works, the side effect is that everything else (assuming Latin1) breaks. As a Frenchie you’d see all your accents disappear.
How much I wish the Wimp had a flag that you could tell it “this application is UTF-8” and it would just go ahead and do something sensible regardless of the configured alphabet. RISC OS can so it – the NetSurf browser and my hacks to Ovation prove this, but the caveat is that you must paint the text yourself, which is somewhere between useless and impossible when we’re talking about window icons and error boxes.
It shouldn’t be hard for the Wimp to often a duplicate don’t handle, the second in UTF-8, and switch as the app requires, but since most users speak English and English looks the same in ASCII and UTF-8…..

And, no, I don’t expect fruity things like switching out fonts to support inline Cyrillic, Greek, kana, etc – that probably requires changes in FontManager for what might be a very narrow use case mais d’avoir des mots dans différents encodages qui semblent les mêmes sur l’écran and not with missing accents (other languages use twiddly bits over characters and they aren’t there as a fashion accessory like the umlaut in the names of certain metal bands) just because the alphabet is different… That is something the Wimp should do.²

¹ UTF-16, on the other hand, represents most English text as a character followed by a null, which would need continual special handling due to C treating a null as a string terminator.

² But, then, I’m well known as the sleeping monster that gets poked whenever something mentions Unicode and RISC OS in the same sentence. ;-) I probably would have ranted more had I a keyboard instead of a tiny screen to write this on. ;-)

Topic: Community Support / Nanoarm on a Pi 3

In fact that was all I needed to identify a subtle differnce between the hard-coded device id and my device id: 1267_0000 vs 1267_0001.

Many thanks, all. I now have a working robot arm and some code I can hack around with and learn from!

Topic: Code review / wchar in ansilib

Hello,
I have DDE27 and I am testing UTF-8 (I am French)
In Clib there is wchar.h but the librarie does’nt contain wchar.c, wctype !
I look up in the last BCM2835Dev, idem.
Please is it possible to get this library, or just wchar.c and wctype.c which are missing.

Perhaps it is better to use Unicode library?

Thanks
Jean-Michel

Topic: Community Support / Nanoarm on a Pi 3

…and…

*Show USB*

…to see what the device identification variable is actually set to.

Topic: Community Support / Nanoarm on a Pi 3

Thanks for the suggestions, guys. I’ll try out !USBInfo first as that seems a sensible place to start.

Topic: Community Support / Nanoarm on a Pi 3

I guess something else to check might be that if there are different revisions of the robot arm, that magic string “$Device_00_00_00_1267_0000_-1*” might need to be something different. Maybe someone can suggest how you’d work out what it should be?

The answer is probably to check the USB devices showing connected by using !USBInfo and if the robot arm shows as a different devcie ID then try a mod of the code to use the ID that is shown by USBInfo

Topic: Community Support / Nanoarm on a Pi 3

I’ve not done any coding with USB, but I would try just printing out size% and device$ directly to see what’s in them instead of just VAL.

Jon Abbott posted some USB code here https://www.riscosopen.org/forum/forums/11/topics/9026#posts-63498 to do with reading a keyboard but it might help as it looks like it checks the status a bit differently.

I guess something else to check might be that if there are different revisions of the robot arm, that magic string “$Device_00_00_00_1267_0000_-1*” might need to be something different. Maybe someone can suggest how you’d work out what it should be?

Topic: Community Support / Nanoarm on a Pi 3

Ah, I see the cut and paste lost some of the content. Let’s try that last bit again:

device$=""
DIM device_buffer% 255
SYS “XOS_ReadVarVal”,"usb"+"$Device_00_00_00_1267_0000_-1*", device_buffer%,255,0,3 TO,,size%
FOR index%=0 TO size%-1: device$+=CHR$(device_buffer%?index%): NEXT index%
IF VAL(device$)=0 PRINTTAB(10,20)‘" **** Robot arm not found ****"’: END

Topic: Community Support / Nanoarm on a Pi 3

Wow, textile is weird…

Topic: Community Support / Nanoarm on a Pi 3

If the code you listed is correct, something is very wrong.
VAL is a BASIC Function, returning the numeric value of a textual string.
The line IF VAL=0 PRINT…
should give an error Unknown or missing Variable – it cannot display Robot arm Not found

Topic: General / We need Acorn back...

Why isn’t this in Aldershot?

Because it cropped up on a thread that was (arguably) respectable elsewhere, and trying to shift a conversation out of a thread into a new thread in a different folder is liable simply to kill it?

Topic: General / We need Acorn back...

Here’s one to ponder: Fall vs Autumn.

And… Why isn’t this in Aldershot?

Topic: General / We need Acorn back...

BTW – I’ve always learned, and used -ise, with the belief that -ize is a strange Americanism, like making “through” be a four letter word.

If it was used in Britain until recently, I wonder if it went out of fashion or if it was a North/South thing?

Topic: General / We need Acorn back...

Arguments about it being a “difference of opinion” fall short of anything approaching intelligence

That’s so often the case, though, isn’t it?

Topic: Community Support / Nanoarm on a Pi 3

Has anyone got a Maplin robotic arm working on a Pi 3 using the BASIC code in the examples directory of the Pico distribution?

I have a Maplin A37JN robot arm that works on another system but if I connect the arm to a USB port on my Pi and double click the BASIC program from the desktop I get “**** Robot arm not found ****”

Is the fact I’m running this from the desktop and not via a Pico boot significant?

Even those without Nanoarm experience may be able to shed some light on this code that’s “failing”…

DEF PROCinit_usb
name$=“OWI 535 ‘Robot Arm Edge’”
device_number%=0: timeout%=100: device$=""
DIM device_buffer% 255,send_buffer% 8,send_byte(8),receive_buffer% 8,byte%(8)
send_bmRbRwV%=&40 OR (&06<<8) OR (&100<<16): send_wIwL%=3<<16
receive_bmRbRwV%=&80 OR (&06<<8) OR (&A00<<16)
receive_wIwL%=2<<16
SYS “XOS_ReadVarVal”,"usb"+"$Device_00_00_00_1267_0000_-1*",device_buffer%,255,0,3 TO,,size%
FOR index%=0 TO size%-1: device$=CHR$(device_buffer%?index%): NEXT index%
IF VAL=0 PRINTTAB‘" **** Robot arm not found ****"’: END
usb$=“USB”device$
ENDPROC

As I noobily understand it, the only relevant bits to initialisation are:

device$=""
DIM device_buffer% 255
SYS “XOS_ReadVarVal”,"usb"+"$Device_00_00_00_1267_0000_-1*",device_buffer%,255,0,3 TO,,size%
FOR index%=0 TO size%-1: device$+=CHR$(device_buffer%?index%): NEXT index%
IF VAL=0 PRINTTAB‘" **** Robot arm not found ****"’: END

From a pseudocode point of view I think I see it’s trying to read a variable relating to USB via a SWI, the result should go into memory pointed at by device_buffer% and then it should iterate through that area of memory byte-by-byte, storing the ASCII equivalent of each one in device$.

But I get nada.

Any observations welcome. Not least on the very-magic-looking string "usb"+"$Device_00_00_00_1267_0000_-1*". (And why is it concatenated?!)

Topic: General / We need Acorn back...

Well. Sort of. See http://clive.semmens.org.uk/Opinion.php?Ize (and a few other infelicities…)

Interesting; growing up in NZ I learned that -ise is the correct (“British”) way and that -ize is the yucky American way. Both Microsoft and Apple’s spellcheckers reject -ize when the country is set to NZ.

As for the claim that -ise is American, it wouldn’t surprise me if it’s soccer all over again!

Topic: General / We need Acorn back...

8~)

Never heard of “ztimulated emizzion of radiasion”?

Topic: General / We need Acorn back...

See http://clive.semmens.org.uk/Opinion.php?Ize (and a few other infelicities…)

Of course there’s always, one of my favourites, the totally incorrect “lazer” to play with :)
Arguments about it being a “difference of opinion” fall short of anything approaching intelligence since LASER is actually an acronym derived from “light amplification by stimulated emission of radiation”.

Topic: General / Beta testing : First Person Shooter and Defender Clone

Thanks for offering to help David Feugey, Rick Murray and Rob Heaton.

Much appreciated.

Back to some late night coding & graphics/video transcoding……..

Topic: General / We need Acorn back...

Windows 8 came out, adding support for proper English

Well. Sort of. See http://clive.semmens.org.uk/Opinion.php?Ize (and a few other infelicities…)

Topic: General / We need Acorn back...

That’s what I always thought, but the OED seems to disagree

Here’s the definition from my copy:

pound sign
noun

• the sign £, representing a pound sterling.
• North American the sign #, representing a pound as a unit of weight or mass, or used as a symbol on a phone keypad or computer keyboard.

I remember back in the mid-90s, we’d just got a voice-capable modem at home and I was trying out the software that came with it. “Record a greeting then press pound”. At that point I thought that American phones actually had £ in place of #!

Meanwhile some of you would probably feel a bit lost here in NZ; apparently our old rotary phones had the numbers in the opposite order from usual, and we currently use the US layout for computer keyboards. You get £ under RISC OS by pressing Shift-Alt-3, and under MacOS by pressing Option-3. As for Windows, I have no idea!

I also remember when Windows 8 came out, adding support for proper English. There were two distinct install DVDs; an American one and an international one. So I put the international DVD in, booted up, and had to do something at the command line. I try to type “C:\” and it pops up with “C:#” instead!

Topic: Bugs / VDU 23,7 behaviour change

Yes, scrolling text is painfully slow on the Pi

I noticed. Thankfully we have a zippy alternative – TaskWindows.

Topic: General / We need Acorn back...

There never was any justification for calling # a £ symbol. A lb symbol if you’re a historian, perhaps.

The historical justification shown on the wikipedia page is like many such things – written by a person with no knowledge of the handwriting of the era. I’ve noted such people transcribe Audrey as Oidrey because they aren’t used to reading that letter formation and assume an uppercase “a” would look like a handwritten “A”

WRT the # £:
Look closely at the image of the handwritten item from Isaac Newton, flatten both upper loops into simple ascenders and remove the bottom right to top left and left-right flourish. What do you have? lb
No relation to the “hash” character anymore than the glyph created when I write my initials makes me a close relative of the (deceased) artist formerly known as Prince

Topic: Bugs / VDU 23,7 behaviour change

Yes. Hardware frame buffers (VIDC, OMAP…) can roll around. The Pi goes via the GPU so behaviour is a bit different.

Ah, so in layman’s terms the reason hardware scrolling is disabled is because it might wrap and GPU’s don’t support wrapped buffers. Now that I know it’s disabled, I probably just need force the hardware scrolling bit when my GraphicsV driver takes over…modifies code…yep, that works.

On platforms which lack VIDC-style hardware scrolling, when asked to scroll the screen the OS typically falls back to the behaviour it uses when the screen has been redirected to a sprite, which is to copy/move pixels around

That explains the black bar that appears then.

For a while I’ve been thinking that it would be good to allow the OS to support limited hardware scrolling on modern systems. E.g. when scrolling the screen up, do a hardware scroll until you hit the limit of screen memory, and then reset back to the start of screen memory and copy the relevant pixels up. When you’re scrolling large amounts of text (e.g. *help) this should make things a lot better. But no promises on when I’ll do that, or exactly what restrictions may be in place!

Yes, scrolling text is painfully slow on the Pi – it’s like using my A310…very nostalgic. I’m sure there are more pressing things to worry about than hardware scrolling support though.