maximum symbol length?


    Apr 05 2006 | 6:55 pm
    I know about the much discussed maximum list length of 256, but is there an official hard-coded maximum symbol length?
    Printing a symbol 1032 characters long crashes for me, where a 1031 character symbol does not crash.
    Example patch:
    My system: OS X alumnium powerbook, OS 10.4.6, Max 4.5.7.
    -charlie

    • Apr 05 2006 | 8:06 pm
      On 5-Apr-2006, at 20:55, Charlie wrote: > I know about the much discussed maximum list length of 256, but is > there > an official hard-coded maximum symbol length?
      Someone asked about this just a few weeks ago. I don't know that there was an official answer. It would be useful if there were one.
      -- P.
      -------------- http://www.bek.no/~pcastine/Litter/ ------------- Peter Castine +--> Litter Power & Litter Bundle for Jitter
      iCE: Sequencing, Recording & |home | chez nous| Interface Building for |bei uns | i nostri| Max/MSP Extremely cool http://www.castine.de http://www.dspaudio.com/
    • Apr 05 2006 | 8:08 pm
      Happy to say, I can't reproduce... both prints fine here.
      MaxMSP 4.57 G5 1.67 mHz
    • Apr 05 2006 | 8:10 pm
      if I add a lot more characters, I get crashes too :(
    • Apr 05 2006 | 9:30 pm
      > Printing a symbol 1032 characters long crashes for me, where a 1031 > character symbol does not crash.
      The forums add spaces to the long symbols, and the email from the list seems to be hard-wrapped at 1000 characters... so you might have to remove the spaces to the patch I sent to get a crash. Or, just add a lot more characters.
      This may be just a "print" related bug, rather than something more fundamental to max...
      -charlie
    • Apr 05 2006 | 9:51 pm
      i heard from well informed sources (ddz) that the maximum length for a symbol is 1999 (or was it 2000?) chars...
      /*j
    • Apr 06 2006 | 10:19 am
      On 5-Apr-2006, at 23:51, /*j wrote: > i heard from well informed sources (ddz) that the maximum length > for a symbol is 1999 (or was it 2000?) chars...
      From what you wrote, probably the buffer is 2000 chars, one of which is the null terminator byte. So effectively 1999 charlen.
      It might be a Good Thing if there were a #define somewhere in the SDK headers so that all objects used the same bufsize. That said, text parsing is usually handled by binbufs, so should be transparent to external objects.
      Another Good Thing might be if Max wouldn't croak on Excessively Enthusiastic Typing. Truncating strings is one thing, crashing another.
      And Charlie's crash was at a mere 1032 chars. Strange. Maybe the print object has a different sized buffer? Unusual choice.
      -- P
      -------------- http://www.bek.no/~pcastine/Litter/ ------------- Peter Castine +--> Litter Power & Litter Bundle for Jitter
      iCE: Sequencing, Recording & |home | chez nous| Interface Building for |bei uns | i nostri| Max/MSP Extremely cool http://www.castine.de http://www.dspaudio.com/