Date: Tue, 17 Oct 2006 02:05:44 +0200 To: Miller Puckette mpuckett@imusic1.ucsd.edu From: Steffen Leve Poulsen slagmark@worldonline.dk Subject: re: [PD] Version 0.41 on XP (patch that crashes)
I looked through the tutorial for Datastructures... ..tried to make my own patch. craxpx10 ... I use a YamahaWx7 windcontroller, a noname USB.MIDI interface a ebehigre FCA202 at 96.000 with -asio -nogui. When its works i sounds very good. The whole application has become more steady, with, more snap.
hi this crashes on XP vh/ Steffen
Thanks for the excellent patch... it does indeed crash on XP and not on linux. The culprits are the "append", "getsize" "setsize" and "element" help windows which used the obsolete "template" object - I replaced them with "struct" objects and things ran ( patches per second) for 1/2 hour or so.
However, I doubt that's the only problem out there, I think there's a more serious one...
Where are you getting at? ------------------------------------>
cheers Miller
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.4/475 - Release Date: 13-10-2006
-- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.4/475 - Release Date: 13-10-2006
However, I doubt that's the only problem out there, I think there's a more serious one...
Where are you getting at? ------------------------------------>
cheers Miller
Just that I think there's at least one outstanding bug that causes Pd to crash on XP, that doesn't have anything to do with data structures, and which pops up mysteriously in the middle of sessions. I got one such crash but went back and repeated just what I had done before and could never reproduce it...
cheers Miller
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.4/475 - Release Date: 13-10-2006
-- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.4/475 - Release Date: 13-10-2006
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
I believe that [openpanel] as Chris Clepper stated, was probably the culprit in my own problems.
However, it was as you said - not repeatable per se. It occurred more than once, but not predictably at the same points. It did seem more likely to occur, when sound was already playing, but that's about all I can say.
~David
On 10/16/06, Miller Puckette mpuckett@imusic1.ucsd.edu wrote:
However, I doubt that's the only problem out there, I think there's a more serious one...
Where are you getting at? ------------------------------------>
cheers Miller
Just that I think there's at least one outstanding bug that causes Pd to crash on XP, that doesn't have anything to do with data structures, and which pops up mysteriously in the middle of sessions. I got one such crash but went back and repeated just what I had done before and could never reproduce it...
cheers Miller
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.4/475 - Release Date: 13-10-2006
-- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.4/475 - Release Date: 13-10-2006
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Creating obj's autostyle shakes things up. I will investigate further, but here is my <o8vious beginngng>. "clear ; " or " clear " is often involved when crashes happens. crash.pd
#N canvas 327 166 715 324 12; #X msg 195 87 ; pd-crash.pd clear; #X msg 66 100 clear; #X obj 66 130 s pd-crash.pd; #X msg 195 127 ; pd-crash.pd clear ;; #X text 127 51 crash; #X obj 71 250 s pd-crash.pd; #X msg 71 220 clear ;; #X text 195 225 <-no crash the first time created , and the patch was cleared. But after that........; #X connect 1 0 2 0;
At 04:38 17-10-2006, Miller Puckett wrote:
Just that I think there's at least one outstanding bug that causes Pd to crash on XP, that doesn't have anything to do with data structures, and which pops up mysteriously in the middle of sessions. I got one such crash but went back and repeated just what I had done before and could never reproduce it...
cheers Miller
Hallo, Steffen Leve Poulsen hat gesagt: // Steffen Leve Poulsen wrote:
Creating obj's autostyle shakes things up. I will investigate further, but here is my <o8vious beginngng>. "clear ; " or " clear " is often involved when crashes happens. crash.pd
This will crash on all platforms, and it's actually a known "bug", which happens, if you try to delete yourself. The current fix is easy: Don't try to delete yourself! The longterm fix seems to be harder.
Attached is a variation of the idiom.
Frank Barknecht _ ______footils.org_ __goto10.org__
juhu I think i found the problem!
Some of the tutorial patches has \; or even \; int he source. It is posibble to load the patches, but eventually a crash wil occur. Maybe you have to load a couple of p's..do some editing, as usual.
Ok I'm of the debug for now, need to shake my body.
Hope it helps
med venlig hilsen Steffen
This python script can/might be helpfull for search and replace.
# regexplace: regular expression search and replace # Stefano Spinucci # 2006-02-07 (rev 4)
# thanks to roadrunner.py # http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52553 # for some ideas and some code
# tested with python 2.3.5
import sys, os, re, string
# pupulate and return 'fileslist[]' with all files inside 'dir' matching 'regx' def make_files_list(dir, regx):
# if dir is not a directory, exit with error
if not os.path.isdir(dir):
sys.exit(dir + ' is not a valid dir to walk !!!')
# compile the search regexp
cregex=re.compile(regx)
# initialize the file list
fileslist = []
# loop on all files and select files matching 'regx'
for root, dirs, files in os.walk(dir):
for name in files:
if cregex.search(name):
path = os.path.join(root, name)
fileslist.append(path)
# return the file list
return fileslist[:]
# in all files in 'fileslist' search the regexp 'searchregx' and replace # with 'replacestring'; real substitution in files only if 'simulation' = 0; # real substitution may also be step by step (if 'stepbystep' = 1) def replace_in_files(fileslist, searchregx, replacestring, simulation, stepbystep):
# compile regexp
cregex=re.compile(searchregx)
# print message to the user
if simulation == 1:
print '\nReplaced (simulation):\n'
else:
print '\nReplaced:\n'
# loop on all files
for xfile in fileslist:
# initialize the replace flag
replaceflag=0
# open file for read
readlines=open(xfile,'r').readlines()
# intialize the list counter
listindex = -1
# search and replace in current file printing to the user
changed lines for currentline in readlines:
# increment the list counter
listindex = listindex + 1
# if the regexp is found
if cregex.search(currentline):
# make the substitution
f=re.sub(searchregx,replacestring,currentline)
# print the current filename, the old string and the
new string print '\n' + xfile print '- ' + currentline , if currentline[-1:]!='\n': print '\n' , print '+ ' + f , if f[-1:]!='\n': print '\n' ,
# if substitution is real
if simulation == 0:
# if substitution is step by step
if stepbystep == 1:
# ask user if the current line must be replaced
question = raw_input('write(Y), skip (n),
quit (q) ? ') question = string.lower(question)
# if quit
if question=='q':
sys.exit('\ninterrupted by the user !!!')
# if skip
elif question=='n':
pass
# if write
else:
# update the whole file variable ('readlines')
readlines[listindex] = f
replaceflag=1
# if substitution is not step by step
else:
# update the whole file variable ('readlines')
readlines[listindex] = f
replaceflag=1
# if some text was replaced
# overwrite the original file
if replaceflag==1:
# open the file for writting
write_file=open(xfile,'w')
# overwrite the file
for line in readlines:
write_file.write(line)
# close the file
write_file.close()
# main function def main():
# if parameters are wrong, exit with error
if len(sys.argv) < 5:
print '\nUsage:'
print 'python regexplace.py dirname files-regexp
search-regexp replace-string'
sys.exit(1)
# ask user for simulated execution or real substitution
print '\nyou are replacing %s with %s in %s' %(sys.argv[3],
sys.argv[4], sys.argv[2]) question1 = raw_input('continue with real substitution (y/N) ? ') question1 = string.lower(question1)
# if user selected real substitution, ask user if execution must
be step by step if question1=='y': question2 = raw_input('\nsubstitute step by step (Y/n) ? ') question2 = string.lower(question2)
# make the file list
fileslist = make_files_list(sys.argv[1], sys.argv[2])
# if real substitution
if question1=='y':
# if step by step
if question2!='n':
replace_in_files(fileslist, sys.argv[3], sys.argv[4], 0, 1)
# if not step by step
else:
replace_in_files(fileslist, sys.argv[3], sys.argv[4], 0,
# if simulated execution else: replace_in_files(fileslist, sys.argv[3], sys.argv[4], 1, 0)
if __name__ == '__main__': main()
THis is of course an unauthorized way to produce the bug. I figure it's relevant anyway because it happens a lot like this when editing. Steffen .................................... new-close-new.pd
#N canvas 116 33 1168 508 12; #X msg 232 360 ; #N canvas 23 24 345 234 12 ;; #X msg 72 360 ; pd filename new ./ ;; #X msg 682 360 ; pd-new menuclose 1 ;; #X msg 456 360 ; pd-new vis 1 ;; #X msg 569 360 ; pd-new vis 0 ;; #X obj 193 285 select 0 1 2 3; #N canvas 0 0 494 344 iterate 0; #X obj 196 50 inlet; #X obj 231 215 outlet; #X obj 46 154 list append; #X obj 196 94 list split 1; #X obj 290 214 print; #X obj 196 133 list trim; #X obj 196 164 t b f f; #X obj 196 246 del 100; #X connect 0 0 3 0; #X connect 2 0 3 0; #X connect 3 0 5 0; #X connect 3 1 2 1; #X connect 5 0 6 0; #X connect 6 0 7 0; #X connect 6 1 1 0; #X connect 6 2 4 0; #X connect 7 0 2 0; #X restore 193 179 pd iterate; #X msg 275 101 1 1 1 2; #X text 343 102 also this is...?; #X text 500 238 you have to restart pd to get rid of the new window. ; #X msg 280 136 0 1 2 3 4 1 2; #X text 16 16 PD 0.40-1 on XP; #X text 404 137 when pressed as the first = crash; #X msg 193 32 0 1 4 2; #N canvas 0 0 820 227 edit 0; #X obj 167 112 t b f f; #X obj 189 146 outlet; #X text 255 110 Backspace the last f , and not the space before , click on canvas. This should crash; #X connect 0 0 1 0; #X restore 347 53 pd edit; #X text 271 33 <-Press x times , until windows stays open. Then edit inside this. ->; #X connect 5 0 1 0; #X connect 5 1 0 0; #X connect 5 2 3 0; #X connect 5 3 4 0; #X connect 5 4 2 0; #X connect 6 0 5 0; #X connect 7 0 6 0; #X connect 10 0 6 0;
Miller Puckett wrote:
Just that I think there's at least one outstanding bug that causes Pd to crash on XP, that doesn't have anything to do with data structures, and which pops up mysteriously in the middle of sessions. I got one such crash but went back and repeated just what I had done before and could never reproduce it...
cheers Miller