I resubscribed with my real email address, which will hopefully fix the posting lag issue.
I've just completed a real world test of the version independant deserialization with a client that spews an enormous number of v1.1.3 events in a very short period of time. All went without any noticable malfunction, though I would still like to get startTime and timeStamp in sync to allow elapsed time. On Sun, 18 Nov 2001, Robert Bushman wrote: > I'm pretty happy with the new version posted at > http://traxel.com/tools/LoggingEvent.java > > It's now reading 1.1.3 and 1.2, and deserialization > has been refactored to a satisfactory level. > > Known Issues: > > - It has not been rigorously tested. > > - startTime is static, so when you serialize, > then compare startTime to timeStamp you get > a meaningless number (startTime is coming > from the server, timeStamp is coming from > the client). This could be solved by making > startTime an instance parameter, and initializing > it to application start time in the constructor > (have a class parameter systemStartTime). > Then readObject could simply overwrite the > instance parameter with the serialized value. > > This is, of course, assuming that the objective > is to be able to tell how long into it's life > the application was when the log was generated. > > > -- > To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> > For additional commands, e-mail: <mailto:[EMAIL PROTECTED]> > > ------------------------------------------------------------------- Who owns your data? Proprietary protocols and file formats are dangerous. ------------------------------------------------------------------- -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
