I think the left image may be better, need to also change the icons in the file explorer to be consistent.
If it matters, I prefer the image on the right where the external files (not in the project folder) are shown in blue. And I don't think you need to change the file explorer.
Ok, yes, in that case there is no need to change the file explorer.
Any chance you're going to the race at Imola this weekend?
Hi macca, all compile errors are gone.
But the generated Code is still buggy.
Here are 2 more Errors I found:
{Spin2_v51}
DAT
counter word 5
PUB start() | ^byte A, i
debug(sdec(counter)) 'OK, counter = 5
repeat counter with i
debug(sdec(i)) 'i counts not 0..4
debug("Start")
if [A] := 3 'crash the Prog.
debug(sdec([A]))
debug("End") 'never get here
In Spin Tools the repeat with loop works with a constant or an initialized variable, just not with a pre-initialized dat value. It does work in PNut v51.
I don't know what the other code is supposed to do so I didn't try testing.
@wummi said:
Hi macca, all compile errors are gone.
But the generated Code is still buggy.
Here are 2 more Errors I found:
{Spin2_v51}
DAT
counter word 5
PUB start() | ^byte A, i
debug(sdec(counter)) 'OK, counter = 5
repeat counter with i
debug(sdec(i)) 'i counts not 0..4
debug("Start")
if [A] := 3 'crash the Prog.
debug(sdec([A]))
debug("End") 'never get here
Thank you for all your reports, I really appreciate it.
@JonnyMac said:
In Spin Tools the repeat with loop works with a constant or an initialized variable, just not with a pre-initialized dat value. It does work in PNut v51.
Ah.. I was wondering why it doesn't work sice there is a test for it, haven't noticed it was a dat variable, need to modify the test.
All I did was extract the zip file and ran the exe. Then the error came up. That was on both versions.
I will work on it again and let you know.
Thanks.
Martin
@pilot0315 said:
Any ideas as to how to fix the error?
All I did was extract the zip file and ran the exe. Then the error came up. That was on both versions.
I will work on it again and let you know.
I already wrote how to fix, delete the .spin-tools file (after closing the program or it will be saved again), but if the error shows even with a clean setup there is something else I don't understand.
Please send a screenshot of the IDE window with the error dialog, and also the .spin-tools file, I really don't understand what happens.
Appears that the gui works but I get this error when launching.
Ok, unfortunately the screenshots aren't of much help, the tab that causes the error seems hidden just after pingtest.spin i guess.
If you switch to each tab what you see ? One of them should be empty, that's the failing file, close the tab and should be fixed (still don't know how that was opened).
You haven't attached the .spin-tools file, it is located in the home directory not the directory where the spin-tools archive was extracted.
But if you delete that file I'm pretty sure the problem goes away.
Also that Ariba x_band code can be downloaded ? I see two x band motion detector entries in Obex but doesn't seems to contain the same files.
What I sent you is how it opens up when launched. Three programs and the error message.
I will try again with your suggestion and let you know. It does not ask to attach anything. I will look for the.spin-tools file but you told me to delete it. I will start over from scratch.
Martin
@pilot0315 said:
What I sent you is how it opens up when launched. Three programs and the error message.
I will try again with your suggestion and let you know. It does not ask to attach anything. I will look for the.spin-tools file but you told me to delete it. I will start over from scratch.
Ok, english is not my mother language so maybe I don't always use the correct terminology and explain myself, let's see if this is more clear:
When you start Spin Tools IDE it reads the .spin-tools file from the user's home directory, for example C:\Users\marco\.spin-tools, this file constains the IDE's preference settings including the source files that were loaded in the editor's tabs the last time the program was closed. I asked you to attach that file to the post to see what's inside that causes the issue, I also asked to delete that file but apparently you didn't because the program still opens with some files loaded, one of which causes the error. If you just delete the spin-tools directory and unpack a freshly downloaded package it doesn't solve anything because the preference file is still there. The file is recreated each time the program is closed, so you need to close the IDE before acting on it.
So you have two options:
Close the IDE and delete the C:\Users\marco\.spin-tools file (or whatever is your user name, note there is a dot at the beginning of the file name), on reopen it should start with the default settings and no tabs open.
Close all tabs from the IDE, close the program, on reopen it should not load anything.
If these fails there is something else going on and I think I can't help.
Well I deleted all instances of the spin tools program.
Unpacked the _64 again. Launched Pinide and it opened up without an error message.
?????????????Oh well. I will see what happens. The magic Gremlin did some P F Magic.
Thanks for your time.
Martin
Another bunch of bugfixes (hope the most critical are finally done) and a couple of enhancements.
Added an initial set of source formatter preferences. More will be added in the future releases.
Does anyone uses or expects to use the IDE with Windows versions older than 10 (like Windows 7 or 8) ?
The current version works on Windows 7 (hopefully also on 8 but never tested) however the latest updates to SWT breaks the compatibility and supports only Windows 10/11.
I'm not sure if finding a workaround is worth the effort, if nobody wants to use this IDE on older Windows versions then I'll give up with the compatibility and support only what SWT supports.
Changes:
Fixed strings in debug backtick commands
Fold constants in debug statements
Skip repeat 0 loop
Math operations fixes, added QLOG, QEXP, FABS, FSQRT
Added alternate icons to object browser to indicate library objects
Fixed pointer operations in conditional expressions
Fixed dat variables in repeat with loop
Fixed inline pasm code size calculation
Fixed multiple struct definitions on same line
Allow long() function to accept floating point values
Fixed CLKMODE field pointer
Fixed cogspin with method variable pointer
Added source formatter preferences
Added line number to overview ruler popup messages
+1 for win10+. while there are some who keep old machines for economical reasons, your support of Linux makes that less of an argument. imho, most old versions of OSs are kept specifically to run old software that is critical for some reason.
I don't know what DEBUG window @macca is considering next, but I do hope it is the terminal window. Ray Allen started an object for it, and I made one, too. The only thing missing is the abiltity to arbitrarily change the foreground and background color of any character (right now the window is limited to using four pre-defined color sets). I have made a request to Chip to make the color and backcolor commands active while the window is live (not just at instantiation). I also asked for zstr and numeric output as in the generic debug window.
This is an example I did with my debug terminal object and my I2C scanner.
@JonnyMac said:
I don't know what DEBUG window @macca is considering next, but I do hope it is the terminal window.
The Terminal debug is already implemented, however I discovered a couple of bugs that prevents it to work with your code (there is also a weird compiler error on some debug statements that I don't quite understand), I'll do a patch release hopefully in the weekend.
Ray Allen started an object for it, and I made one, too. The only thing missing is the abiltity to arbitrarily change the foreground and background color of any character (right now the window is limited to using four pre-defined color sets). I have made a request to Chip to make the color and backcolor commands active while the window is live (not just at instantiation). I also asked for zstr and numeric output as in the generic debug window.
I can add the color and backcolor commands to the update loop, I don't think it will cause any issue.
@macca That looks interesting... The debug terminal thing doesn't work with flexprop, all debug seems to go to same window, so must be some magic to make it work in Spin Tools IDE...
The debug terminal thing doesn't work with flexprop...
I think the debug windows only exist in PNut, Propeller Tool (though not up to date), and Spin Tools (though not all windows). I can confirm the debug terminal window does work in Windows 10 Pro (I think Marco's screen grab was from Linux). Thanks again, Marco.
Well... I seem to be having trouble with backcolor after instantiation. I tried two variations on style, but neither works.
If I predefine the color sets, it does work (same as PNut).
Just for clarity: my debug terminal object creates a window with the symbol name 'dt' -- knowing this allows manual debug commands to be mixed with calls to the object methods.
@JonnyMac said:
Well... I seem to be having trouble with backcolor after instantiation. I tried two variations on style, but neither works.
The keywords works like the setup keywords, color updates the 4 foreground and background pairs:
debug(`term dt title 'Debug Terminal' size 80 25 color cyan black)
debug(`dt 'Cyan on Black', 13)
debug(`dt color white red)
debug(`dt 'White on Red', 13)
Also backcolor seems to be used only on setup and to clear the window (with the clear command).
Maybe I don't understand the syntax of the PNut debug terminal window colors. I may be chatting with Chip today so I'll ask. the "color on color" syntax is much nicer than what I was doing which was trying to match the canonical syntax. I will suggest your variation to Chip.
@JonnyMac said:
Maybe I don't understand the syntax of the PNut debug terminal window colors. I may be chatting with Chip today so I'll ask. the "color on color" syntax is much nicer than what I was doing which was trying to match the canonical syntax. I will suggest your variation to Chip.
It is not my variation, it is how the color keyword work with PNut (and is what you wrote in the setup string):
byte "color "
CS0 byte "cyan black "
CS1 byte "green black "
CS2 byte "black green "
CS3 byte "white red "
Each argument is a foreground / background pair for the four sets of colors.
My only variation is to allow the commands to work after the initial setup.
Comments
Can't wait!!!! (Though from a coach with Pizza!)
Ok, yes, in that case there is no need to change the file explorer.
Nope.
Hi macca, all compile errors are gone.
But the generated Code is still buggy.
Here are 2 more Errors I found:
{Spin2_v51} DAT counter word 5 PUB start() | ^byte A, i debug(sdec(counter)) 'OK, counter = 5 repeat counter with i debug(sdec(i)) 'i counts not 0..4 debug("Start") if [A] := 3 'crash the Prog. debug(sdec([A])) debug("End") 'never get hereIn Spin Tools the repeat with loop works with a constant or an initialized variable, just not with a pre-initialized dat value. It does work in PNut v51.
I don't know what the other code is supposed to do so I didn't try testing.
Thank you for all your reports, I really appreciate it.
Ah.. I was wondering why it doesn't work sice there is a test for it, haven't noticed it was a dat variable, need to modify the test.
I think the "repeat loop with dat variable" thing was at one point a bug in flexspin -1bc mode as well :P Fun times.
@macca
Howdy,
Any ideas as to how to fix the error?
Thanks
Martin
@macca
All I did was extract the zip file and ran the exe. Then the error came up. That was on both versions.
I will work on it again and let you know.
Thanks.
Martin
I already wrote how to fix, delete the .spin-tools file (after closing the program or it will be saved again), but if the error shows even with a clean setup there is something else I don't understand.
Please send a screenshot of the IDE window with the error dialog, and also the .spin-tools file, I really don't understand what happens.
@macca
Here are four pictures.
1. The extracted files with the highlighted selection.
2. The error message.
3. The details.
4. Ariba x_band code in operation
Appears that the gui works but I get this error when launching.
Thanks for your time.
Martin
Ok, unfortunately the screenshots aren't of much help, the tab that causes the error seems hidden just after pingtest.spin i guess.
If you switch to each tab what you see ? One of them should be empty, that's the failing file, close the tab and should be fixed (still don't know how that was opened).
You haven't attached the .spin-tools file, it is located in the home directory not the directory where the spin-tools archive was extracted.
But if you delete that file I'm pretty sure the problem goes away.
Also that Ariba x_band code can be downloaded ? I see two x band motion detector entries in Obex but doesn't seems to contain the same files.
@macca
What I sent you is how it opens up when launched. Three programs and the error message.
I will try again with your suggestion and let you know. It does not ask to attach anything. I will look for the.spin-tools file but you told me to delete it. I will start over from scratch.
Martin
Ok, english is not my mother language so maybe I don't always use the correct terminology and explain myself, let's see if this is more clear:
When you start Spin Tools IDE it reads the .spin-tools file from the user's home directory, for example
C:\Users\marco\.spin-tools, this file constains the IDE's preference settings including the source files that were loaded in the editor's tabs the last time the program was closed. I asked you to attach that file to the post to see what's inside that causes the issue, I also asked to delete that file but apparently you didn't because the program still opens with some files loaded, one of which causes the error. If you just delete the spin-tools directory and unpack a freshly downloaded package it doesn't solve anything because the preference file is still there. The file is recreated each time the program is closed, so you need to close the IDE before acting on it.So you have two options:
C:\Users\marco\.spin-toolsfile (or whatever is your user name, note there is a dot at the beginning of the file name), on reopen it should start with the default settings and no tabs open.If these fails there is something else going on and I think I can't help.
@macca
No issue with the mother language.
I will follow the instructions shortly.
Thanks and let you know.
Martin
@macca
Well I deleted all instances of the spin tools program.
Unpacked the _64 again. Launched Pinide and it opened up without an error message.
?????????????Oh well. I will see what happens. The magic Gremlin did some P F Magic.
Thanks for your time.
Martin
That's good news. You should be cheering.
Released version 0.48.0
Another bunch of bugfixes (hope the most critical are finally done) and a couple of enhancements.
Added an initial set of source formatter preferences. More will be added in the future releases.
Does anyone uses or expects to use the IDE with Windows versions older than 10 (like Windows 7 or 8) ?
The current version works on Windows 7 (hopefully also on 8 but never tested) however the latest updates to SWT breaks the compatibility and supports only Windows 10/11.
I'm not sure if finding a workaround is worth the effort, if nobody wants to use this IDE on older Windows versions then I'll give up with the compatibility and support only what SWT supports.
Changes:
+1 for Win10+ only. You have to optimise your generously gifted time.
In a pinch we can run older versions on an older OS.
+1 for win10+. while there are some who keep old machines for economical reasons, your support of Linux makes that less of an argument. imho, most old versions of OSs are kept specifically to run old software that is critical for some reason.
I don't know what DEBUG window @macca is considering next, but I do hope it is the terminal window. Ray Allen started an object for it, and I made one, too. The only thing missing is the abiltity to arbitrarily change the foreground and background color of any character (right now the window is limited to using four pre-defined color sets). I have made a request to Chip to make the color and backcolor commands active while the window is live (not just at instantiation). I also asked for zstr and numeric output as in the generic debug window.
This is an example I did with my debug terminal object and my I2C scanner.
The Terminal debug is already implemented, however I discovered a couple of bugs that prevents it to work with your code (there is also a weird compiler error on some debug statements that I don't quite understand), I'll do a patch release hopefully in the weekend.
I can add the color and backcolor commands to the update loop, I don't think it will cause any issue.
I'm sorry I didn't know that -- I'll give it a try when you make the patch.
Thank you!
Released version 0.48.1
Bugfix release to fix some issues with the Terminal debug window.
Changes:
Thank you, Marco!
@macca That looks interesting... The debug terminal thing doesn't work with flexprop, all debug seems to go to same window, so must be some magic to make it work in Spin Tools IDE...
I think the debug windows only exist in PNut, Propeller Tool (though not up to date), and Spin Tools (though not all windows). I can confirm the debug terminal window does work in Windows 10 Pro (I think Marco's screen grab was from Linux). Thanks again, Marco.
Well... I seem to be having trouble with backcolor after instantiation. I tried two variations on style, but neither works.
If I predefine the color sets, it does work (same as PNut).
Just for clarity: my debug terminal object creates a window with the symbol name 'dt' -- knowing this allows manual debug commands to be mixed with calls to the object methods.
The keywords works like the setup keywords, color updates the 4 foreground and background pairs:
debug(`term dt title 'Debug Terminal' size 80 25 color cyan black) debug(`dt 'Cyan on Black', 13) debug(`dt color white red) debug(`dt 'White on Red', 13)Also backcolor seems to be used only on setup and to clear the window (with the
clearcommand).Maybe that's not what you mean.
Maybe I don't understand the syntax of the PNut debug terminal window colors. I may be chatting with Chip today so I'll ask. the "color on color" syntax is much nicer than what I was doing which was trying to match the canonical syntax. I will suggest your variation to Chip.
It is not my variation, it is how the color keyword work with PNut (and is what you wrote in the setup string):
Each argument is a foreground / background pair for the four sets of colors.
My only variation is to allow the commands to work after the initial setup.