Hi Tim,
After a few evenings of struggle, I think I found where the problem
comes from. I was thrown off by the fact that the same application
works, successfully deployed for over 2 years now. Here is the issue:
the null/empty pair in Jscheme is represented by the static final
Pair.EMPTY which has both first and rest assigned to itself. The method
'isEmpty' simply compares 'this == Pair.EMPTY' instead of 'this == first
== rest" (doesn't look like it's possible in Jscheme to construct such a
pair). Now, the problem is when I serializes a Pair into a DB blob and
then deserialize it. The deserialized object looks exactly like
Pair.EMPTY, but it's not one and the same Java reference.
So, I found out that previous version of Jscheme had a Pair.readResolve
method that specifically dealed with the empty pair. Is there any
particular reason why this method was removed from later versions?
Thanks,
Boris
| -----Original Message-----
| From: Timothy J Hickey [mailto:tjh...@br...]
| Sent: Wednesday, December 07, 2005 7:10 AM
| To: Borislav Iordanov
| Subject: Re: [Jscheme-user] Help interpreting a stack trace
|
|
| Hi Boris,
|
| It looks like the stack overflow occurs inside the
| iterate call, but its hard to tell. If you send me the source
| code I might be able to help you find it.... Typically this
| will happen with a non-tail-call recursion as the Java stack
| is not very deep...
|
| Best,
| ---Tim---
|
| On Dec 7, 2005, at 12:14 AM, Borislav Iordanov wrote:
|
| > Hello,
| >
| > I'm trying to track down a stack overflow in a Jscheme code I wrote
| > some
| > time ago. Can anyone help me with the meaning of the following stack
| > trace:
| >
| > (heat-it image-graphics (get-polygons-for-coordinate (.get
| radials i
| > )(.get stages j ))(get-signal-for-region i j insitu-result
| )max-signal
| > )
| > j = 3
| > stages = [Stage 4, Stage 3, Stage 2, Stage 1]
| > applicable = [Stage 1]
| > stage-loop = {jsint.Closure ??[3] (j stages applicable)}
| > i = 0
| > radials = [Columella, Cortex, Endo, Endo+Cortex,
| > Epidermis(atrichoblasts), Lat.root cap, Pericycle, Phloem, St...
| > radial-loop = {jsint.Closure ??[2] (i radials)}
| > max-signal = 527.924F
| > image-graphics =
| >
| sun.java2d.SunGraphics2D[font=java.awt.Font[family=Dialog,name
| =Dialog,s
| > t
| > yle=plain,size=12],color=jav...
| > image-encoder = sun.awt.image.codec.JPEGImageEncoderImpl@39452f
| > insitu-result =
| org.pamweb.ui.ticl.model.MySQLTableModel$Row@1ed13da
| > session =
| org.apache.catalina.session.StandardSessionFacade@1b25a82
| > request = org.apache.coyote.tomcat4.CoyoteRequestFacade@541b02
| > out = org.apache.coyote.tomcat4.CoyoteOutputStream@1b94ea2
| >
| > ====================================
| >
| > (iterate radial-polygons {jsint.Closure ??[1]
| (radial-point-list)} )
| > Error in BacktraceException.printStackTrace:
| > java.lang.StackOverflowError
| > Error in BacktraceException.printStackTrace:
| > java.lang.StackOverflowError
| > Error in BacktraceException.printStackTrace:
| > java.lang.StackOverflowError
| >
| > Specifically, should I assume that the stack overflow
| occurs somewhere
| > with the (iterate radial-polygons ...) call?
| >
| > Thanks,
| > Boris
| >
| >
| >
| > -------------------------------------------------------
| > This SF.net email is sponsored by: Splunk Inc. Do you grep
| through log
| > files
| > for problems? Stop! Download the new AJAX search engine that makes
| > searching your log files as easy as surfing the web.
| DOWNLOAD SPLUNK!
| > http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
| > _______________________________________________
| > Jscheme-user mailing list
| > Jsc...@li...
| > https://lists.sourceforge.net/lists/listinfo/jscheme-user
|
|
