Comments for michaelgalloy.com

Comment on VISualize 2012 by michaelgalloy.com — IDL User Group 2012 Fall Meeting

michaelgalloy.com — IDL User Group 2012 Fall Meeting — Tue, 22 May 2012 23:46:46 +0000

[...] addition to VISualize coming up soon in June, Exelis VIS is holding a fall User Group meeting as [...]

Comment on User functions in ENVI 5 by Mort Canty

Mort Canty — Mon, 21 May 2012 09:36:53 +0000

I have mixed feelings, too. Another pile of functions/procedures to learn. As a minimum, we’ll have to arrange that the extensions appear both in the ENVI 5 and ENVI Classic menus. I took part in the beta, and I got the impression that a lot of users would, like me, shun the new single window GUI and stick to the classic. I use two monitors and want to spread the displays over both. Personal opinion: the new GUI sucks.

Comment on Processing with Javascript by michaelgalloy.com — Data visualization tools

michaelgalloy.com — Data visualization tools — Fri, 18 May 2012 19:35:57 +0000

[...] have mentioned many of these tools like ColorBrewer (many times), Processing, Processing.js, and Many Eyes. No Comments [...]

Comment on Prompt hiding for cut-and-paste by Mark Piper

Mark Piper — Fri, 18 May 2012 03:04:24 +0000

This is really cool, Mike; thanks!

Comment on Prompt hiding for cut-and-paste by Michael Galloy

Michael Galloy — Thu, 17 May 2012 22:49:29 +0000

OK, styling is a bit better now; this is in the trunk of IDLdoc now and will be in the next release.

Comment on Prompt hiding for cut-and-paste by Michael Galloy

Michael Galloy — Thu, 17 May 2012 02:08:43 +0000

Here's an example of IDLdoc using this idea.

Comment on Scientific iPhone backgrounds created in IDL by Michael Galloy

Michael Galloy — Mon, 19 Mar 2012 21:15:15 +0000

Mark Piper has an Android phone of some sort and he showed me the background on his phone (the home screen version has a band for Apple's page indicator dots, so stick with the lock screen version).

I will make a post about GPULib soon. The short answer is: I have more funding (that isn't quite in yet), expect a new release this spring.

Comment on Scientific iPhone backgrounds created in IDL by Mort Canty

Mort Canty — Mon, 19 Mar 2012 19:19:57 +0000

Hi Mike,

Very nice. Anything for us Apple-hating androids?

But "seriously", how is GPULib doing these days?

Best wishes,

Mort

Comment on IDLdoc 3.4.3 released by Michael Galloy

Michael Galloy — Wed, 07 Mar 2012 14:47:01 +0000

I have an effort in the dist_tools project to auto generate IDL bindings from something similar to a C header file. I am currently using it to generate IDL bindings for MAGMA routines (GPU accelerated LAPACK).

Comment on IDLdoc 3.4.3 released by Jim P

Jim P — Wed, 07 Mar 2012 06:35:18 +0000

The template got me most of the way there and I just ended up writing a post processor to remove the excess characters. From the code, it looks like I'd use the "prefix" property of a doctreeargument instead of the suffix. The suffix appears to always return a null.

I've found a couple new uses for the doc template in the last week. I wonder how hard it would be to use this to generate SWIG bindings. An exercise left for the reader...

Thanks, Mike!

Comment on IDLdoc 3.4.3 released by Michael Galloy

Michael Galloy — Fri, 02 Mar 2012 04:33:35 +0000

No, there is no good way to do this in the template language. What I did in places like profile.tt was added a suffix variable to each parameter/keyword object. This placed the burden on the parameter to know if it was the last one, setting it's suffix to "," if it wasn't, to "" if it was. Then my template loop was like:

[% FOREACH p IN args][% SCOPE p %]...[% suffix %][% END %][% END %]

Comment on IDLdoc 3.4.3 released by Jim P

Jim P — Thu, 01 Mar 2012 23:22:45 +0000

Hey Mike, I've been messing with templates to have IDLdoc generate a different sort of output file. Let's say I have a loop such as

[[% FOREACH p IN parameters %][% SCOPE p %] {...}, [% END %]]

Is there a magic template escape I can use to prevent the last "," from appearing? I want my output to look like [ {...}, {...} ] instead of [ {...}, {...}, ]

Thanks, Jim P.

Comment on POV-Ray destination v0.2 by michaelgalloy.com — Object graphics to the web

michaelgalloy.com — Object graphics to the web — Mon, 06 Feb 2012 20:16:15 +0000

[...] I have done similar things for various other new destination classes for outputs such as POV-Ray and anaglyphs. No Comments [...]

Comment on IDL 8.0: Operator overloading by Implicit Get/SetProperty calls in user-defined classes, part II | The IDL Data Point

Thu, 02 Feb 2012 15:51:56 +0000

[...] we can go further! What I’ll do next overlaps with the description and example of operator overloading that Mike Galloy gave some time ago, but it feels natural to extend the [...]

Comment on Tech-X at AGU by Michael Galloy

Michael Galloy — Tue, 29 Nov 2011 15:01:41 +0000

You said you weren’t going to AGU this year, but I still had hope that there would be something.

Comment on Tech-X at AGU by Mark Piper

Mark Piper — Tue, 29 Nov 2011 03:19:07 +0000

Unfortunately, VIS will not host an IDL user group meeting this year.

(Y'know, you could've asked the VIS employee you see a couple times a week, the one who'd likely be running the user group meeting...)

Comment on Logging by Tony Leggett

Tony Leggett — Sat, 26 Nov 2011 10:00:43 +0000

sorry my error, Mike..

Comment on Logging by Tony Leggett

Tony Leggett — Thu, 24 Nov 2011 09:08:17 +0000

I must have downloaded an old version of mg_log from svn.

Comment on Logging by Michael Galloy

Michael Galloy — Wed, 23 Nov 2011 17:40:41 +0000

I'm not sure what you mean by a "filename parameter to a log file"? The filename for a log file can be specified using the FILENAME keyword to MG_LOG.

Comment on Logging by Tony Leggett

Tony Leggett — Wed, 23 Nov 2011 07:22:42 +0000

Looks to be a useful tool for IDLDE testing. The filename parameter to a log file is not implemented in the version 1.0 is it?