Hi Luke, thanks for the report. It's a typical buffer overrun (caused by a loooong line of data), making quick-and-dirty code choke. I hope to fix it shortly, please stand by! greetings, Thomas
Am 09.05.2007 um 04:24 schrieb Luke Iannini (pd): > Hi Thomas (and list), > I have run into what appears to be a limitation with [pool], namely > that it crashes with something around ~780 characters loaded from a > file as values in a key. > > I ran into this loading a particularly long list in Memento. I've > attached a patch and a file from which to load to experience the > crash. The file is just a little over the character count needed to > crash it (it seems to be between 780 and 800?). It does not seem to > be a function of the number of elements in the list, but rather the > length of the text in those elements. Shorten a few of the values in > the list and it loads right up. > > I'm also pretty sure this is a per-key limitation, as memento's lddir > first loads files into /tmp before copying them into the proper key. > I've used that with a file that was just under the crashing length and > that worked fine (loaded into two keys). > > I have tested this on both my PPC Mac and my Intel Mac; maybe others > can check it out too? > > (also to clarify, the problem occurs when loading data from a file; > data of any length (that I've tried) makes it into the pool without > issue, and saves without issue) > > Thanks! > Luke > <PoolCrash.pd> > <CoscNearlyAll> Thomas Grill http://grrrr.org _______________________________________________ PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list