LeafLabs Garden

A place to share, learn, and grow...

1. 

   StephenFromNYC
   Moderator

   Friday February 18, 2011

   Hello,

   I would have announced the good news about my successful LabVIEW 2009 - Maple communication code sooner ( http://forums.leaflabs.com/topic.php?id=656 ), but a new Maple IDE bug almost caused me to give up two hours ago. The problem was not in the SerialUSB.println() function, but in SerialUSB.read().

   The following code works on Maple IDE 0.0.9, but not with the latest Maple IDE snapshot ( described in http://forums.leaflabs.com/topic.php?id=651#post-3635 )

   I cannot get the null character to display correctly, even using the backticks. I put a comment on the line where it should appear.

   void setup()
   {
   }

   void loop()
   {
   int count=0;
   char chararray[64];

   if (SerialUSB.available())
   {
   count=0;

   while (SerialUSB.available() > 0)
   {
   chararray[count++]=SerialUSB.read();
   }

   chararray[count]='\'; /* null character to end the string */

   SerialUSB.println(chararray);
   }
   }

   If you send "ABC" (without the quotes) using the serial monitor Maple IDE 0.0.9 correctly echos back "ABC", but the snapshot echos "AAA".

   Is this an easy fix?

   Thanks!

   Stephen from NYC

   P.S. - I am using Windows XP SP3 on a Pentium III laptop

   Posted 4 years ago #
2. 

   crenn
   Moderator

   I can confirm this behaviour, I'll take a look into it now and see if I can fix it.

   Posted 4 years ago #
3. 

   crenn
   Moderator

   I can side step the problem by doing this:

   void loop(){
int count=0;
char chararray[64];
if (SerialUSB.available()){
count=SerialUSB.read(chararray,SerialUSB.available());
chararray[count]='\'; /* null character to end the string */
SerialUSB.println(chararray);
}
}

   EDIT: Seems like there is a problem with the code.

   This will fix it:

   void loop(){
int count=0;
char chararray[64];
if (SerialUSB.available()){
count=SerialUSB.available();
SerialUSB.read(chararray,count);
chararray[count]=''; /* null character to end the string */
SerialUSB.println(chararray);
}
}

   Posted 4 years ago #
4. 

   StephenFromNYC
   Moderator

   Hello crenn,

   Thanks for the help.

   In your last comment, when you said "EDIT: Seems like there is a problem with the code" are you saying there is a problem with the code you posted above or a problem with the Maple IDE code?

   This is a just a stab in the dark, but I just had an idea. If the SerialUSB.read() problem is in the Maple code is it possibly relate to SerialUSB.peek()? I do know if SerialUSB.peek() is implemented in the current snapshot (February 2011), but if read() is erroneously acting like peek() then the read() index would not be incremented.

   Also, are you also have problems showing code which displays the null character (backslash zero)?

   I created a new topic asking how to use the backtick character.

   http://forums.leaflabs.com/topic.php?id=658

   I found a hack to show a null character. I do not know if it is possible to edit an old post, but to avoid confusion for future reads I will not try. Here is an edited copy of the code which I posted to start this topic which hopefully will show the backslash zero null character in code which shows a possible bug in SerialUSB.read():

   void setup()
   {
   }

   void loop()
   {
   int count=0;
   char chararray[64];

   if (SerialUSB.available())
   {
   count=0;

   while (SerialUSB.available() > 0)
   {
   chararray[count++]=SerialUSB.read();
   }

   chararray[count]=''; /* null character to end the string */

   SerialUSB.println(chararray);
   }
   }

   The null character now correctly appears in the code.

   Thanks!

   Stephen from NYC

   Posted 4 years ago #
5. 

   gbulmer
   Moderator

   I don't see any null character in your post

   try '\O' (used the letter O, instead of 0, yes, rubbish, but better than nothing)

   Posted 4 years ago #
6. 

   crenn
   Moderator

   It was a problem with the code I posted above due to the code not working how I was expecting it to work.

   Posted 4 years ago #
7. 

   mbolivar
   LeafLabs

   Hi Stephen,

   Yes, as I mentioned when I posted the snapshot, there appears to be a regression in the IDE's built-in serial monitor:

   http://forums.leaflabs.com/topic.php?id=651

   There've been some issues lately with the built-in serial monitor; if you get bad behavior, try using something like PuTTY, HyperTerminal, screen, or minicom. Working on fixing this as well.

   The behavior is the one you discovered: If you try to echo back some bytes, you'll appear to get copies of the first byte you sent.

   We've tested it out on other terminal programs (e.g. screen, minicom) and everything works fine. This behavior is not a problem with the SerialUSB implementation. We're going to get this issue ironed out before the next release, but for now, since every OS comes with a serial monitor built-in, we recommend not using the IDE's serial monitor as a workaround.

   Thanks for the report!

   Posted 4 years ago #
8. 

   mbolivar
   LeafLabs

   (And sorry for not being more clear about what to expect!)

   Posted 4 years ago #
9. 

   crenn
   Moderator

   The thing that interests me is the behaviour changed, when using the slightly different function, to what I would expect it to be. However, I'll go and check out this shortly and see what happens.

   Also, I strongly believe that Windows 7 (The only OS I've been using for this project, so far) has no longer got a serial monitor built-in. However, I do have 3rd party programs that I use (although usually only used with either different serial hardware or to complete a serial connection with a bluetooth module).

   Posted 4 years ago #
10. 

    mbolivar
    LeafLabs

    crenn,

    Sigh, you're right. We should have checked. In the meantime, here's a workaround using PuTTY:

    1) Download PuTTY: http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html
    2) Run putty.exe
    3) There will be a radio menu titled "Connection Type". Choose "Serial", then pick the COM port corresponding to your Maple, and you should be all set.

    Posted 4 years ago #
11. 

    crenn
    Moderator

    It's fine, as I use PuTTY on a daily basis to connect back to my home network to grab files I've forgotten and also go on IRC with a single session. I've also used it when using different serial hardware and the bluetooth module.

    Posted 4 years ago #
12. 

    mbolivar
    LeafLabs

    crenn,

    Have you been successful avoiding the bad behavior using PuTTY?

    Also, I didn't mean to imply anything about you in particular; the workaround was posted for everyone's benefit :).

    Posted 4 years ago #
13. 

    gbulmer
    Moderator

    mbolivar -
    

    Have you been successful avoiding the bad behavior using PuTTY?

    For the benefit of the non-PuTTY users, what is the bad behaviour, and how can it be avoided?

    Posted 4 years ago #
14. 

    snigelen
    Member

    Here's an example of the bad behviour
    

    If you send "ABC" (without the quotes) using the serial monitor Maple IDE 0.0.9 correctly echos back "ABC", but the snapshot echos "AAA".

    I guess it can be avoided by not using the snapshot, or if you use it, don't use the built in serial console in the Maple IDE.

    Posted 4 years ago #
15. 

    crenn
    Moderator

    mbolivar said:

    crenn,
    Have you been successful avoiding the bad behavior using PuTTY?

    The problem still exists in PuTTY, it's just masked. Because for PuTTY, as you type, it sends the characters as you type them. So it's only collecting a single character. However, if you paste text into the PuTTY window, it sends this text in very quick succession so that the code sees multiple bytes are available, which triggers the bug.

    gbulmer said:

    For the benefit of the non-PuTTY users, what is the bad behaviour, and how can it be avoided?

    See the above posts for the description of the bad behaviour, and also how to avoid it.

    Posted 4 years ago #

RSS feed for this topic

Reply »

You must log in to post.