Welcome to this ***VMS*** SYSTEM /local/whatever is a convenience for changing directory You cannot download a file by specifying /local/whatever/filename - you can use [local.whatever]filename but easier to cd /local/whatever get filename some particular things here... ------------------------------ (Latest addition - a LaTeX spell checker - see split.doc) divpost*.trm is a modified postcript driver for gnuplot, allowing greek symbols and super/sub-scripts on labels, etc. It is a drop-in replacement for post.trm SORRY - THIS IS VERY DISORGANISED AT PRESENT - I am waiting for 3.6 ... /local/gnuplt34/term/enhpost.trm is the one I am presently using IT APPEARS THAT ENHPOST WILL BE ADPOTED IN GNUPLOT 3.6 divpost32.trm is for gnuplot 3.2 divpost324.trm is the above, modified to run under 3.4 divpost34.trm is a rewrite for 3.4 - get divpost324 if you have problems... newpost.trm is yet another postscript driver! This one changes the encoding vector for the default font - brings in various european characters... - different char set from 'eurofont.ps' - see below... this uses ISOLatin char set. eurofont.ps uses pc character set (?) [ I didn't make this one, so dont know if it works...] [ might not work with multiplot patch ?] enhpost.patch is a rewrite, using rather more clever postscript than I could have done. Also contains some other gnuplot patches.By Matt Heffron. enhpost.trm contains just the terminal driver. This one needs modifications to term.c as well... THERE IS A SMALL BUG IN ENHPOST - doesn't affect output... see enhpost.fix - I haven't fully tested the fix yet... a patched enhpost is in /local/gnuplt34/term ENHPOST IS ALSO AVAILABLE at ftp.dartmouth.edu:pub/gnuplot/enhpost.patches eurofont.ps is some postscript code which will allow divpost.trm to access the full pc character set - angstroms, umauts, etc If you want to get a printout of the symbol character set, with character numbers in octal, for use in enhpost, see /local/gs261/prfont.ps Add a line to the bottom like /Symbol DoFont then send the file to your postscript printer [and cross your fingers] - base was hex by default - I set to octal, but its simple to change. A hack to improve felxibility of gnuplot data-file plotting is in /local/gnuplt34/divhack.patch This allows full expressions in the using... clause of a plot command. $n is used to access the value in column 'n' eg plot 'data' using ($2+$3):($2*$3):4 contour{34|35}.patch is a small mod to the table terminal type, which allows the contours calculated by splot to be written to file set nosurface set contour set term tab set out 'contour.dat' splot whatever set out # contours are now in contour.dat plot 'contour.dat' using whatever, etc - plot affords more control over the contour plot than set view and splot polar.patch is a completely untested attempt to fix a bug with polar datafile splots in 3.5 - ignores radius specifed in 3rd column Only try this if you need polar splots. It's fixed properly in 3.6 Let me know if the patch breaks anything else. A replacement linux terminal driver to try to plug an enormous security hole. I needed it because gnuplot running as root couldn't access remote files by nfs. linux.trm - replacement linux driver gnuplot_linux.c - image spawned by linux.trm - needs sticky bit fonts.h - used by above - ripped from bitmap.h - apparently this isn't needed - or won't be soon..? (Anything else here was probably put here temporarily for particular people to download - there might be details in 11readme.txt ?) roadmap ------- The tree starting [local...] is where I put all the _working_ system (VMS) extensions - you're welcome to browse, but nothing is set up to make it easy to download... in fact, there are often undocumented local modifications, so this is not really a good place to be downloading stuff from. [ Changes are sometimes flagged with 'div was here' or such things in source. Look out for files called readme.div or similar I have been going round all my c utilitities, adding "mbc=60" to all fopen's - you wouln't believe how much direct i/o it saves (eg 1000 down to 20 for a large file !) ] many people seem to worry about this being a vms system... but cd /local will still get you there :-) NB the anon-ftp server wont let you go to / then to local - you'll have to go straight to /local or [local] - actually, I've made a hard-link from /anonymous/local to /local so 'cd local' should get there - hope that helps... Having to reorg disks.... /local is still DCL procedures and vax binaries. Alpha binaries are scratch$disk:[local] Development sources are now more and more under scratch$disk:[local...] What might you be interested in..? /local/scan_q.com runs in background to pass jobs submitted to sys$postscript through ghostscript to print on sys$print. I actually no longer use this, and use remprtsmb to send jobs with decnet task-to-task to a dcl program running ghostscript [below]. It tries to peek back at the queue for accounting info, but this doesn't work on non-cluster nodes. /local/gslaser.com runs ghostscript to convert .ps files and print. accepts various flags, such as /batch and /delete Also works as a decnet object. see /local/help/postscript.hlp /local/cd.com allows vms directory to be set with dos or unix syntax /local/hp2sixel is a hp pcl (ie laserjet) to DEC sixel _or_ postscript converter. See readme.div in that directory... Most binaries are in [local] - VMS 5.4 and DECWindows, if it matters. - might change this, as we now have an alpha - vax .exe's might go in [local.vax] and alpha in [local.axp] Default application startup files go in [local.defaults] Locally written .hlp files are in [local.help] I keep `interesting' snippets from infovax in [local.infovax] Some interesting documents, newsgroup FAQ's, etc go in [local.docs] Source, and things like fonts for ghostscript, are below this in the tree [but I'd rather you got this sort of thing from the official distributions - just seen someone 'mget' the whole gs source, file by file !] - source for packages is now going on scratch$disk:[local...] Sometimes I forget to allow world read on files - please send reminders and other problems to system@conmat.phys.soton.ac.uk ( Note that some files are not public domain, so some things might be deliberately protected. Please do not interpret access to mean availability - I'm not terribly careful, but OTOH we dont have much commercial software anyway... )