Mailman 3 May 2011 - yt-dev

Generic data loader
by Matthew Turk 27 May '11

27 May '11

Hi all, In support of the Paraview pipeline between yt/pv, I wrote up a new frontend called "stream." Right now it's not unitful, but it is designed to be lightweight and to accept and transport arbitrary data around as fully-functional yt hierarchy. This could be a mechanism for prototyping new data loaders (or refactoring existing ones, frankly). There are likely still bugs in it, but a proof-of-concept is here: http://matthewturk.bitbucket.org/html/ba3fd37b-842c-4641-b21e-2d3f5268eefe-… To be able to project, you need to set the units with spf.units["cm"] = spf.units["unitary"] = 1.0, because I'm not yet sure how to handle that exactly. All this requires is a few parameters, the mesh, and a manner for generating or loading data. One could even imagine using this to construct arbitrary data sets in memory -- for instance, the field loader could return an analytic value for Density that matches something. In my view, this should ultimately be able to handle conversion to other data formats, and thus the generation of initial conditions or restart outputs. Anyway, comments welcome! -Matt

1 0

Time-based releases
by Matthew Turk 26 May '11

26 May '11

(moved to yt-dev) Hi all, This is a great reason to push on a more regular release schedule. Do you think it might be worthwhile to aim for time-based releases? We've had a spring release, maybe aim for a Summer release, and then we could try to just do rolling releases like that? [+-][01] on [quarterly|semiannual|annual] releases? -Matt On Thu, May 26, 2011 at 1:35 PM, Anthony Harness <anthony.harness(a)colorado.edu> wrote: > Britton, > > Adding those lines works for the stable install, which I guess is maintained > by Hari. My problem was running the unstable install maintained by Stephen, > which he informed me does not work in parallel. As Stephen mentioned before, > if Hari could update his development install, I could use that instead. > > Thanks, > Anthony > > _______________________________________________ > yt-users mailing list > yt-users(a)lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org > >

6 9

Activate / deactivate
by Matthew Turk 24 May '11

24 May '11

Hi all, Casey Stark added a new mechanism for activating and deactivating yt. I've merged this in, modified only slightly to be more explicit in the install script about which version is being loaded, and pushed this to the main repo. Casey, feel free to add any comments you might have on this. It also reduces the huge amount of output the install script spits out. If you have the chance to try out installing yt on a new machine, or an old machine, it'd be great if you could check this functionality out. I think it could be very useful for developing as well as keeping stacks of software separate. No more manually loading and unloading environment variables. Best, Matt

1 0

Fix for Mac readline
by Casey W. Stark 24 May '11

24 May '11

I noticed that on a default install (just running install_script.sh), yt issues warnings about not having access to readline: WARNING: Readline services not available on this platform. WARNING: The auto-indent feature requires the readline library Inside the yt env, I ran this: $ pip install -U ipython $ pip install readline and everything is good now. I'm guessing this just means some simple mods to install_script.sh, but this works for now. Best, Casey

2 1

Reason fixes
by Matthew Turk 23 May '11

23 May '11

Hi all, I've added a task queue-type mechanism to Reason. This fixed for me some of the issues with multiple projections showing up, as well as this weird lingering sometimes-it-doesn't-work thing. Anyway, Reason is working great for me, so testing from everybody out there would be super handy. Just make sure you're on the latest version everywhere, and it should be good to go. https://bitbucket.org/yt_analysis/yt/changeset/b8a4c0e7cd40/ When Britton was here a while back, the two of us and Cameron chatted and came up with three things we wanted implemented before it could go out: 1. Sometimes slices/projections silently failed. This is fixed for me, but your mileage may vary. 2. We need an "Open File" dialog. 3. A button to paste the contents of a single cell. I am thinking this would be a hideable toolbar that shows up on hover in the upper right of a cell, and then you can download, paste, or copy to the current cell the contents of the cell in question. Anything else? -Matt PS This is the last thing for a 2.2 release. After 2.2 goes out, we'll start using the Bitbucket bug tracker for everything.

1 0

A fun interface for data
by Matthew Turk 19 May '11

19 May '11

Hi all, Britton, Cameron, Sam and I were chatting this weekend about a "google maps" style interface to 2D data, like slices and projections. We ended up finding a javascript library that handles it, so I wrote a simple little script that runs it. I'd love it if people could download and try it out. If you have yt installed, then you don't need anything other than this repo: https://bitbucket.org/MatthewTurk/yt_map You just need to change the loading of pf at the top to be a dataset on your local machine; I'd recommend one with a couple levels of refinement. (I really have checked that this works out of the box. ;-) Once you've pointed it at some other data, just do: python2.7 run.py and then open up: http://127.0.0.1:8080/ On my laptop, I get great performance, and it's even all hidden by the background-loading of tiles. The overhead for keeping a very large projection in memory can be as high as a couple hundred megs, but I think that's the only impediment to making this accessible on a public facing webserver. One could imagine this being used with multiple tile layers (which is supported by the javascript engine) for each field, adding in markers for halos/galaxies/star forming clumps, and then even textual annotations of those clumps. I think this could be pretty fun to play with. The entire code necessary to run this is in run.py; as you can see, it's super small and very straightforward. The entire thing is ~60 lines. It'll also probably get rolled into reason before the yt 2.2 release; it's really fun to use to explore data! Any feedback would be greatly appreciated, either through forking on BB or on the mailing list. -Matt PS I have a pretty cool dataset that can be explored with this, which I'm happy to turn on if you want to explore/experiment without downloading this to your machine. Contact me off-list if you do.

5 11

Forked depo
by gso＠physics.ucsd.edu 13 May '11

13 May '11

Hi everyone, I've completed the AMR3D object "ellipsoid", and I think it is ready for others to test/play with. I've forked a yt repo and followed the steps on http://yt.enzotools.org/doc/advanced/developing.html under section "How To Submit Changes" however, after I put in the changes, hg commit hg push went to notify the developer at http://hg.enzotools.org/yt/pull Matt suggested that I notify people on the developer list as well, and I'm stuck on how to let others pull the changes I've made to my bitbucket repo. This is probably a hg/bitbucket question, but I'm new to this and thought you guys are probably pro at this. How do I let others access or pull a specific changeset (676c7f17fb97) off of the repo I just forked? Once that is setup, I'll post some scripts to plot the ellipsoid cells in 2D, 3D so people can make sure that it working. From G.S.

4 16

CIA.vc is down indefinitely
by Matthew Turk 12 May '11

12 May '11

Hi all, Looks like we're down one bot in IRC: http://scanlime.org/2011/05/cia-vc-service-is-down-indefinitely/ To be honest, I actually really liked the bot -- and I have spoken to a couple of you that also really liked it. Alas, it was not meant to be. One could imagine bringing this back in a minor way with an IRC bot that lived on a server somewhere, but that's not going to happen for a while. So long, CIA.vc! -Matt

2 2

Forked depo
by gso＠physics.ucsd.edu 10 May '11

10 May '11

Hi everyone, I've completed the AMR3D object "ellipsoid", and I think it is ready for others to test/play with. I've forked a yt repo and followed the steps on http://yt.enzotools.org/doc/advanced/developing.html under section "How To Submit Changes" however, after I put in the changes, hg commit hg push went to notify the developer at http://hg.enzotools.org/yt/pull Matt suggested that I notify people on the developer list as well, and I'm stuck on how to let others pull the changes I've made to my bitbucket repo. This is probably a hg/bitbucket question, but I'm new to this and thought you guys are probably pro at this. How do I let others access or pull a specific changeset (676c7f17fb97) off of the repo I just forked? Once that is setup, I'll post some scripts to plot the ellipsoid cells in 2D, 3D so people can make sure that it working. From G.S.

1 0

Re: [Yt-dev] [yt-users] User Survey Results and "State of Stuff"
by Stephen Skory 09 May '11

09 May '11

Hi all, I've sent this reply to the dev list on purpose. > Additionally, the recipe system is going to be expanded. > Several new recipes have been added, but unfortunately we are now > reaching a scaling issue. If anyone has any suggestions how to scale > to dozens of recipes/snippets, they would be greatly appreciated! I've just spent a few days attempting to learn a couple astronomer-written tools (Xspec, Sherpa) and had a bear of a time due to their generally poor documentation. Incorrect names (seriously, it's ridiculous), poor- to non-existant-cross referencing, and way too complicated examples that rely on specific files that aren't introduced or explained during the example. I think our docs are generally OK on these issues, but it always pays to be aware of what a new user experiences. Perhaps for organizing examples, would a tagging system work? Tags like 'Halo Finding' or 'Volume Rendering' and so on... -- Stephen Skory s(a)skory.us http://stephenskory.com/ 510.621.3687 (google voice)

1 0