Since we can look at the source code, we might be able to find the bug. Can you cut the problem down to the bare minimum case that makes the problem show up? If so, post that.
Henry Rich On 5/14/2011 3:06 PM, Arthur Anger wrote: > The immediate execution session below shows that stored values, or > stored-value > retrieval, can be corrupted. In these examples, a change of retrieval > method--1!:2 vs. implicit display--almost always causes corruption to > occur. Repeating a particular display method seldom causes corruption. > (The last > output is particularly puzzling, suggesting that 1!:2 isn't sure what value to > report.) > > This is merely one handle for tracking down the root cause; the original > encounter with corrupt values occurred during an execution with no displays. > --Art Anger > ------------------ > bdysegs > 43179272 0 > 0 0 > bdysegs > 43179272 0 > 0 0 > bdysegs > 43179272 0 > 0 0 > smoutput bdysegs > 1886680431 29813 > 0 0 > smoutput bdysegs > 1886680431 29813 > 0 0 > bdysegs > 13571784 0 > 0 0 > smoutput bdysegs > 1886680431 29813 > 0 0 > bdysegs > 43179272 0 > 0 0 > bdysegs > 43179272 0 > 0 0 > bdysegs > 43179272 0 > 0 0 > smoutput bdysegs > 1886680431 29813 > 0 0 > bdysegs > 13609096 0 > 0 0 > smoutput > ?????????????????????? > ? ? ???????????????? > ? ? ??????????? ? ?? > ? ? ???1?!:?2?? ? ?? > ? ? ???????????&?2?? > ?0 0?$???????????????? > ?????????????????????? > 0 0$ 1!:2&2 bdysegs > 9665096 9805192 > 9665096 9525128 > 0 0$ 1!:2&2 bdysegs > 9665096 9805192 > 9665096 9525128 > 1!:2&2 bdysegs > 9666440 13651208 > 13581256 9666440 > 9666440 9666440 > 13581256 9666440 > --------------------- > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
