This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
Hi Jay- Thanks to Tom Whittaker's help, I think I have a solution that sets the first word to 0 for ADDE images. It will be in tomorrow's build. Let me know if that solves your problem. Thanks to Dave for addressing the IMGLIST issue. Don > > A user in CIMSS and I have been trying out the "Making data source > > local" option to create area files. I've been using the latest IDV > > nightly (from August 3rd), and the area files I'm producing are not > > usable in the -V Local ADDE Manager or in -X. > > > > I've talked with TomW and DaveS and they've indicated the problem to > > me... that a zero needs to be written into the first word of the area > > file. TomW mentioned that you had released a change on 1 July which > > does this, and it was included in the last visad.jar file. So, I'm > > wondering if this change was lost, not implemented at the moment, or > > broken, etc. Here is the latest from TomW: > > As you point out, the problem is that the first word is not 0. There is > conflicting documentation on this. If it's truly a bad image, why can one do > an IMGLIST on it? > > That being the case, I have not yet implemented the change I made in > visad.jar in the IDV. That requires some overhead which I'm not sure I want > to incur in the IDV. Currently, we just write the ADDE stream directly to > disk and the first word is not 0 in an ADDE Image object. Using the VisAD > fix, I'd have to create an AreaFile out of each stream and then use the new > method to write to disk. My attempts to just set word 0 after writing out > the ADDE stream fail on Windows because it locks the file and doesn't allow a > subsequent write. It probably works on Linux, but I haven't had a chance to > test it there. > > Dave Santek's workaround is to do an LWU POKE 0 0 on each file. However, the > file names have to be short or you have to create links to each file with a > short name since LWU doesn't handle long names/paths. But, that will work. > > Ticket Details =================== Ticket ID: QHB-724761 Department: Support IDV Priority: Normal Status: Open