Steve, Thanks for this improvement.
You're welcome. Though, there may be a small bug with it that I have been unable to determine the exact cause and solution for. Sometimes it has crashed on stopProgressDialog() in the Stimuli processInput function of RawSpiceOut.java. Sometimes a workaround is to run it multiple times until the Waveforms window comes up and restarting Eclipse sometimes may have helped. I haven't tried making a jar file to see if there is a relation to use of the Eclipse IDE environment or not. I'm not really happy that you have to place a dummy file somewhere to tell Electric what is going on. It should really be a preference. So I changed it. Yeah, I was also thinking about a checkbox in "Spice/CDL" Preferences but due to my lack of experience with the Electric code base I wasn't sure how complex the programming for it would be. So I went with the simple dummy file approach looking for advice, but thanks for implementing the checkbox. New code is in the repository. Now, in the "Spice/CDL" Preferences, there is a checkbox "Use newer LTSpice XVII" which you must check to make your code active. No empty files are necessary. Thanks for putting the new code in the repository, the new option in "Spice/CDL" looks nice: [image: New_Spice-CDL_window.png] Also, I have attached the Ant build file that assembles a working version of Electric with all the bells and whistles. Thanks for sending the buildPublicWithExtras.xml file for Ant from the packaging folder [ http://svn.savannah.gnu.org/viewvc/electric/trunk/electric/packaging/ ]. I will read more about Ant in the Electric manual https://www.staticfreesoft.com/jmanual/mchap01-04-03.html https://www.staticfreesoft.com/jmanual/mchap01-04-02.html and on the Internet. Of note, in the "Building Electric from Source Code with Eclipse.pdf" previously sent, I have discovered that it is the installation of Scala IDE 4.7.x that currently breaks the "New launch configuration" in the Run Configuration. So the "Add Scala IDE to Eclipse" section can be skipped. Gavin On Monday, December 21, 2020 at 3:27:55 AM UTC-7 Gavin wrote: > Developer(s) and Steve, > > 1. I'm looking at the Electric Eclipse usage instructions at: > https://www.staticfreesoft.com/jmanual/mchap01-04-05.html > > Do you have steps that you can share on how you create the > electric-9.08d.jar and electricBinary-9.08d.jar from the current trunk > source code with Eclipse as I'm not seeing that in the instructions? > > 2. Can you test and incorporate the attached RawSpiceOut.java into the > savannah repository for Electric, if you like how it works for plotting > LTspice XVII raw data with Write Spice Deck, refer to the attached "Plot > LTSPICE XVII raw with Electric.pdf"? > > Thanks, > > Gavin > > On Sunday, November 22, 2020 at 10:42:05 PM UTC-7 Gavin wrote: > >> Prof. Baker, >> >> Thanks for informing of the existence of the 'readLtsRaw.m' Matlab >> script. I currently don't have access to Matlab, but with a few >> modifications to 'readLtsRaw.m', it looks like it will work with Octave as >> shown in the attached "Plot LTSPICE XVII raw with Octave.pdf". I may be >> able to use what I learned from this in the future to try to modify the >> Electric JAVA source. >> >> Gavin >> >> On Sunday, November 22, 2020 at 2:36:23 PM UTC-7 rjacobbaker wrote: >> >>> One of my students just sent me the following related to problems with a >>> MATLAB script we use also not being able to read the output of the newer >>> version of LTspice. This (newer version of LTspice represents characters >>> with 2 bytes instead of 1) is likely the issue with Electric so if someone >>> wants to use the newer version of LTspice with Electric the below might be >>> useful (if you make the programming Java change). >>> >>> If there is ever a new version of Electric, that is v. 9.08, it would be >>> good to get the LTspice code updated so that newer version of LTspice work >>> with Electric. Again see: >>> http://cmosedu.com/cmos1/ltspice/ltspice_electric.htm >>> >>> Best wishes, Jake. >>> >>> ------------------------------------------------------- >>> Old versions of LTSpice apparently use one byte to represent characters. >>> The old 'readLtsRaw.m' script would therefore use this code to read from >>> the header: >>> >>> header(linenum,charnum)= fread(fid, 1, '*uint8'); >>> >>> The result is that when the new versions of LTSpice use two characters >>> (UTF16?) every second character is just reading a zero and the condition >>> checks for end of line >>> and end of header fail. Changing the code to read two bytes fixes the >>> issue. the corrected line is: >>> >>> header(linenum,charnum)= fread(fid, 1, '*uint16'); >>> >>> Updated 11/22/2020 >>> >>> -- >>> Faculty do research to enhance their teaching in the lab and classroom. >>> http://cmosedu.com/jbaker/jbaker.htm >>> >> -- You received this message because you are subscribed to the Google Groups "Electric VLSI Editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to electricvlsi+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/9bff3651-0272-4dc3-9c7b-38f8a6cbaf6cn%40googlegroups.com.
