> since I am leaving on vacation for the holidays tomorrow, and I was > partly responsible for starting this thread, I thought I would sum up > my position and make a graceful exit.
Since I am *primarily* responsible and since I am also going to be travelling ... I should do the same :-) I have not posted anything the last couple of days. In part because I have been busy and in large part because I don't have any vested interest one way or the other. I have never written any real RasMol scripts; I don't know what the usual coding standards are and I personally don't have any legacy scripts to support. Here are my thoughts and observations on various postings that were made: ^ conflicts with regular expressions ------------------------------------ This is not an issue. If there were some need to introduce regular expression pattern matching into the scripting language, then we would do so in an unambiguous context. icode on/off ------------ When I first read this proposal I was attracted to it. Because to me the expressions look much cleaner when the insertion code immediately follows the sequence number. (My initial implementation in Jmol operated in this way, and I was quite distressed when someone pointed out that this conflicted with the relaxed/non-colon chain specification syntax.) However, I don't like the idea of having this piece of 'state' as a 'mode setting' in the background. When reading an expression, it is another piece of state that one has to remember ... or forget. It would be better if the expression evaluation behavior was more context-free. The issue of 'subroutines' could certainly be addressed by putting the icode setting at the top of each file, but it is something else that would have to be done. #! version number header ------------------------ I don't think this is appropriate at this time. It implies that there will be other scripting language enhancements. I am not willing to start that discussion until Jmol is released and has 99.44% compatibility with existing Chime scripts. Roger Sayle's 2 cents --------------------- RAS indicated his preference for the ^ syntax. For me, that carries a lot of weight. Tim Driscoll's opinion ---------------------- Tim is actively helping me with Jmol ... *that* carries more weight that Roger :-) Numeric insertion codes ----------------------- This is *the* deciding factor. The fact is, there are entries in the PDB that have numeric insertion codes. (As a software engineer, my favorite is 1BIT :-) Therefore, some escape character is needed to separate the sequence number from the insertion code. I was not aware that these numeric insertion codes existed when I posted the request-for-comments ... the PDB File Format Spec says "A-Z,a-z". I must admit that I have had some second thoughts about having raised this issue. In hindsight, I wonder whether or not I should have stayed away from the 'insertion code' issue. The current goal for Jmol is RasMol/Chime compatibility ... to provide an upgrade path for existing Chime web applications. I chose to bring up the insertion code issue because it was easy for me to fix in the Jmol code and because someone asked me for it. And I thought that the syntax extension would be a good thing to put before the community. But the discussion on this topic has uncovered the fact that we have no formal mechanism to resolve differences of opinion and reach closure on an issue. In summary, I posted a "Request For Comments" and I received some very good and thoughtful comments ... but now I'm not sure what to do. I have the responsibility to follow up on what I started. I suppose that I need to post a final 'ruling', but that seems rather harsh and I'm not particularly comfortable doing it. So I'll put it off and do it next week :-) Miguel ------------------------------------------------------- This SF.net email is sponsored by: IBM Linux Tutorials. Become an expert in LINUX or just sharpen your skills. Sign up for IBM's Free Linux Tutorials. Learn everything from the bash shell to sys admin. Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click _______________________________________________ Jmol-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jmol-users
