<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Yes, the user will always be able to
      change the "fields" keyword in load() to override the defaults.<br>
      <br>
      A user-defined default field vector sounds like something that
      needs careful thought, since it's common to run simulations with
      and without metals, MHD, etc, in the same project, in which case
      different field arrays are needed in rapid succession. Of course,
      it's always possible for users to create modules for themselves
      and add them to their $PYTHONPATH, so in this case, I think the
      simplest solution would be to write a module with something like
      the following:<br>
      <br>
      myfields = ["blah1","blah2",...]<br>
      def myload(fileloc):<br>
          from yt.mods import load<br>
          return load(fileloc, fields=myfields)<br>
      <br>
      then import that module to get a personal default loading
      interface.<br>
      <br>
      On 26/11/13 19:52, Romain Teyssier wrote:<br>
    </div>
    <blockquote
      cite="mid:A3C58892-D443-4CAE-9D9F-AD6A7CA260D3@gmail.com"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html;
        charset=ISO-8859-1">
      I like Sam’s suggestion (option a).
      <div>Based on how many variables we have in the file, we propose a
        default variable name vector,</div>
      <div>like the ones proposed by Sam, but the user should always
        have the possibility to overwrite the default </div>
      <div>because he is the only one to really know which variables he
        used.</div>
      <div><br>
      </div>
      <div>One cool option would be for the user to be able define a
        default variable name vector himself.</div>
      <div><br>
      </div>
      <div>Cheers,</div>
      <div>Romain</div>
      <div><br>
      </div>
      <div><br>
        <div>
          <div>On 26 Nov 2013, at 16:44, Sam Geen <<a
              moz-do-not-send="true"
              href="mailto:samgeen@astro.ox.ac.uk">samgeen@astro.ox.ac.uk</a>>
            wrote:</div>
          <br class="Apple-interchange-newline">
          <blockquote type="cite">
            <div bgcolor="#FFFFFF" text="#000000">
              <div class="moz-cite-prefix">Quick question for RAMSES
                users - do you prefer:<br>
                a) YT figures out what variables are in the file from
                NVAR, e.g. NVAR=5 gives density,vx,vy,vz,pressure,
                NVAR=6 adds metallicity, NVAR=10/11 assumes MHD<br>
                b) YT assumes density,vx,vy,vz,pressure only *unless*
                you do:<br>
                load("mydata/info_blahblah.txt",metals=True)<br>
                (with similar flags for mhd, etc?)<br>
                <br>
                In all cases if you set "fields" as a parameter (as
                normal), YT will use that, of course.<br>
                <br>
                For RT YT just has to find a info_rt_blahblah.txt file,
                so no user input is needed. If anyone can give info on
                how RAMSES-ATON outputs work, it would be appreciated.<br>
                <br>
                On 25/11/13 20:56, Nathan Goldbaum wrote:<br>
              </div>
              <blockquote
cite="mid:CAJXewOnr-_gsLrq+UMGanVqPzayO6hU1CX8HOcqZO1mJunNE7A@mail.gmail.com"
                type="cite">
                <div dir="ltr">Just for completeness, the yt developer
                  guide is here: <a moz-do-not-send="true"
                    href="http://yt-project.org/doc/developing/developing.html">http://yt-project.org/doc/developing/developing.html</a></div>
                <div class="gmail_extra"><br>
                  <br>
                  <div class="gmail_quote">On Mon, Nov 25, 2013 at 11:38
                    AM, Matthew Turk <span dir="ltr"><<a
                        moz-do-not-send="true"
                        href="mailto:matthewturk@gmail.com"
                        target="_blank">matthewturk@gmail.com</a>></span>
                    wrote:<br>
                    <blockquote class="gmail_quote" style="margin:0 0 0
                      .8ex;border-left:1px #ccc solid;padding-left:1ex">
                      Yup, there is, but stop by IRC if you run into
                      trouble.  It should be<br>
                      as easy as making sure your ~/.hgrc has your
                      name/email in [ui], for<br>
                      instance:<br>
                      <br>
                      [ui]<br>
                      username = Matthew Turk <<a
                        moz-do-not-send="true"
                        href="mailto:matthewturk@gmail.com">matthewturk@gmail.com</a>><br>
                      <br>
                      then:<br>
                      <br>
                      cd /path/to/yt/hg/repo<br>
                      <br>
                      -edit edit edit-<br>
                      hg commit<br>
                      <br>
                      On the BitBucket site you can fork our repository
                      (do not make a<br>
                      branch though!) here: <a moz-do-not-send="true"
                        href="https://bitbucket.org/yt_analysis/yt-3.0/fork"
                        target="_blank">https://bitbucket.org/yt_analysis/yt-3.0/fork</a><br>
                      <br>
                      and you can then just push to your repo:<br>
                      <br>
                      hg push -f <a moz-do-not-send="true"
                        href="http://bitbucket.org/SamsUserName/yt-3.0"
                        target="_blank">http://bitbucket.org/SamsUserName/yt-3.0</a><br>
                      <br>
                      Then pull request away!<br>
                      <br>
                      -Matt<br>
                      <div class="HOEnZb">
                        <div class="h5"><br>
                          On Mon, Nov 25, 2013 at 2:35 PM, Sam Geen <<a
                            moz-do-not-send="true"
                            href="mailto:samgeen@astro.ox.ac.uk">samgeen@astro.ox.ac.uk</a>>

                          wrote:<br>
                          > OK, will try to figure out all that
                          tomorrow. I guess there's a guide to<br>
                          > issuing pull requests on the yt site?<br>
                          ><br>
                          ><br>
                          > On 25/11/13 20:30, Matthew Turk wrote:<br>
                          >><br>
                          >> Hi Sam,<br>
                          >><br>
                          >> This looks great.  Any chance you
                          could add this to the<br>
                          >> yt/frontends/ramses/definitions.py
                          file as a function, issue a pull<br>
                          >> request, and we'll get your name in
                          the commit log?  I'll then modify<br>
                          >> it to fit in and we'll use it as
                          fallback when fields are specified.<br>
                          >><br>
                          >> -Matt<br>
                          >><br>
                          >> On Mon, Nov 25, 2013 at 1:39 PM, Sam
                          Geen <<a moz-do-not-send="true"
                            href="mailto:samgeen@astro.ox.ac.uk">samgeen@astro.ox.ac.uk</a>>

                          wrote:<br>
                          >>><br>
                          >>> I've hacked up a simple example
                          of what I mean (attached).<br>
                          >>><br>
                          >>><br>
                          >>> On 25/11/13 17:50, Nathan
                          Goldbaum wrote:<br>
                          >>><br>
                          >>> Matt's opened a PR for this here:<br>
                          >>><br>
                          >>><br>
                          >>> <a moz-do-not-send="true"
href="https://bitbucket.org/yt_analysis/yt-3.0/pull-request/140/adding-field_aliases-to-ramses-datasets/diff"
                            target="_blank">https://bitbucket.org/yt_analysis/yt-3.0/pull-request/140/adding-field_aliases-to-ramses-datasets/diff</a><br>
                          >>><br>
                          >>> Sam, if you could enumerate how
                          to map from the boolean flags you're<br>
                          >>> suggesting to a final field list,
                          I think that would help immensely.<br>
                          >>><br>
                          >>><br>
                          >>> On Mon, Nov 25, 2013 at 8:42 AM,
                          Sam Geen <<a moz-do-not-send="true"
                            href="mailto:samgeen@astro.ox.ac.uk">samgeen@astro.ox.ac.uk</a>>

                          wrote:<br>
                          >>>><br>
                          >>>> Yes and no. NVAR must be big
                          enough, but it can be bigger than necessary<br>
                          >>>> (I think); it's a Makefile
                          setting when compiling Ramses. You can<br>
                          >>>> guarantee<br>
                          >>>> that the first 4 variables
                          are the same, but after that I suspect the<br>
                          >>>> best<br>
                          >>>> approach is to use boolean
                          flags to specify whether the run contains<br>
                          >>>> MHD,<br>
                          >>>> RHD, metals, etc. Normally
                          it's a good guess that 5 is pressure and 6<br>
                          >>>> (if it<br>
                          >>>> exists) is metals, but as I
                          said it seems like MHD puts the B-field<br>
                          >>>> before<br>
                          >>>> pressure and metallicity.<br>
                          >>>><br>
                          >>>><br>
                          >>>> On 25/11/13 17:36, Nathan
                          Goldbaum wrote:<br>
                          >>>><br>
                          >>>> Does that mean that it should
                          be possible to infer the field list from<br>
                          >>>> NVAR alone?<br>
                          >>>><br>
                          >>>><br>
                          >>>> On Mon, Nov 25, 2013 at 8:29
                          AM, Sam Geen <<a moz-do-not-send="true"
                            href="mailto:samgeen@astro.ox.ac.uk">samgeen@astro.ox.ac.uk</a>><br>
                          >>>> wrote:<br>
                          >>>>><br>
                          >>>>> For MHD, the variables
                          stored are the B-field strengths on the faces
                          of<br>
                          >>>>> each cell - the fields in
                          order are: (1) density, (2-4) [x,y,z]<br>
                          >>>>> velocity,<br>
                          >>>>> (5-7) [x,y,z] B-field
                          left, (8-10) [x,y,z] B-field right, (11)
                          pressure<br>
                          >>>>> (thermal + MHD), then
                          (12+) any passive scalars (metals, etc) - see<br>
                          >>>>> mhd/output_hydro.f90.
                          Note that the MHD module adds 3 variables, so
                          the<br>
                          >>>>> number of variables will
                          be nvar+3.<br>
                          >>>>><br>
                          >>>>> For RHD (Joki Rosdahl's
                          M1 method implementation, at least - someone<br>
                          >>>>> else<br>
                          >>>>> will have to help with
                          the ATON version), there is a separate "rt_"<br>
                          >>>>> output<br>
                          >>>>> file for each CPU (like
                          the hydro/amr/particles), as well as a
                          separate<br>
                          >>>>> info<br>
                          >>>>> file (there is a
                          bitbucket link on the Ramses homepage if you
                          want to<br>
                          >>>>> see<br>
                          >>>>> what the latest version
                          of the code does in detail). The RHD module<br>
                          >>>>> divides<br>
                          >>>>> photons into "nGroups"
                          different flux bins; for each bin, it writes a<br>
                          >>>>> photon<br>
                          >>>>> density and then a flux
                          vector. nGroups is stored in the rt info file;<br>
                          >>>>> nrtvar is written to the
                          rt_ binary file, but this is (I think) a<br>
                          >>>>> Makefile<br>
                          >>>>> setting and not
                          guaranteed to be nGroups/(1+ndim) (the user
                          is, I<br>
                          >>>>> believe,<br>
                          >>>>> normally free to waste
                          memory by allocating more variables than they<br>
                          >>>>> need).<br>
                          >>>>> For details it might be
                          easiest to download the latest code and look
                          at<br>
                          >>>>> the<br>
                          >>>>> code in the folder rt/,
                          or ask Romain or Joki Rosdahl directly.<br>
                          >>>>><br>
                          >>>>> (Feel free to correct me
                          if any of this is wrong!)<br>
                          >>>>><br>
                          >>>>> I agree that setting the
                          default to the standard five (perhaps some<br>
                          >>>>> boolean variables for
                          common configurations such as metals, RHD,
                          MHD,<br>
                          >>>>> etc)<br>
                          >>>>> and then letting users
                          set or add their own fields by hand if they
                          need<br>
                          >>>>> to<br>
                          >>>>> is a good approach.<br>
                          >>>>><br>
                          >>>>><br>
                          >>>>> On 25/11/13 16:23,
                          Matthew Turk wrote:<br>
                          >>>>>><br>
                          >>>>>> Hi all, especially
                          RAMSES users,<br>
                          >>>>>><br>
                          >>>>>> I'm setting up
                          aliases to make it easier to load RAMSES
                          datasets that<br>
                          >>>>>> have different field
                          configurations.<br>
                          >>>>>><br>
                          >>>>>> Right now, to specify
                          a different set of fields, you have to do:<br>
                          >>>>>><br>
                          >>>>>> pf = load( ... ,
                          fields = [ "Density", "x-velocity", ...])<br>
                          >>>>>><br>
                          >>>>>> I am going to change
                          this to add on different field configurations<br>
                          >>>>>> that can be specified
                          with just a string.  For instance:<br>
                          >>>>>><br>
                          >>>>>> pf = load(..., fields
                          = "standard_five")<br>
                          >>>>>><br>
                          >>>>>> I'm wondering what
                          the different configurations are.  I have
                          these<br>
                          >>>>>> two:<br>
                          >>>>>><br>
                          >>>>>> standard_five:
                          Density, x-velocity, y-velocity, z-velocity,
                          Pressure<br>
                          >>>>>> standard_six:
                          Density, x-velocity, y-velocity, z-velocity,
                          Pressure,<br>
                          >>>>>> Metallicity<br>
                          >>>>>><br>
                          >>>>>> But for MHD and RHD,
                          what are the additional fields that should be<br>
                          >>>>>> added, and in what
                          order?  And are there other common
                          configurations<br>
                          >>>>>> that should be
                          aliased?<br>
                          >>>>>><br>
                          >>>>>> If we can come up
                          with a set of aliases that match most use
                          cases,<br>
                          >>>>>> that'd work quite
                          well.  I'm going to issue a PR for this
                          shortly, and<br>
                          >>>>>> hopefully it will
                          make everybody's life a bit easier.<br>
                          >>>>>><br>
                          >>>>>> -Matt<br>
                          >>>>>>
                          _______________________________________________<br>
                          >>>>>> yt-users mailing list<br>
                          >>>>>> <a
                            moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          >>>>>> <a
                            moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
                          >>>>><br>
                          >>>>><br>
                          >>>>>
                          _______________________________________________<br>
                          >>>>> yt-users mailing list<br>
                          >>>>> <a
                            moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          >>>>> <a
                            moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
                          >>>><br>
                          >>>><br>
                          >>>><br>
                          >>>><br>
                          >>>>
                          _______________________________________________<br>
                          >>>> yt-users mailing list<br>
                          >>>> <a moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          >>>> <a moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
                          >>>><br>
                          >>>><br>
                          >>>><br>
                          >>>>
                          _______________________________________________<br>
                          >>>> yt-users mailing list<br>
                          >>>> <a moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          >>>> <a moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
                          >>>><br>
                          >>><br>
                          >>><br>
                          >>>
                          _______________________________________________<br>
                          >>> yt-users mailing list<br>
                          >>> <a moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          >>> <a moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
                          >>><br>
                          >>><br>
                          >>><br>
                          >>>
                          _______________________________________________<br>
                          >>> yt-users mailing list<br>
                          >>> <a moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          >>> <a moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
                          >>><br>
                          >>
                          _______________________________________________<br>
                          >> yt-users mailing list<br>
                          >> <a moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          >> <a moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
                          ><br>
                          ><br>
                          >
                          _______________________________________________<br>
                          > yt-users mailing list<br>
                          > <a moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          > <a moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
_______________________________________________<br>
                          yt-users mailing list<br>
                          <a moz-do-not-send="true"
                            href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
                          <a moz-do-not-send="true"
                            href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org"
                            target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
                        </div>
                      </div>
                    </blockquote>
                  </div>
                  <br>
                </div>
                <br>
                <fieldset class="mimeAttachmentHeader"></fieldset>
                <br>
                <pre wrap="">_______________________________________________
yt-users mailing list
<a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a>
</pre>
              </blockquote>
              <br>
            </div>
            _______________________________________________<br>
            yt-users mailing list<br>
            <a moz-do-not-send="true"
              href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
<a class="moz-txt-link-freetext" href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
          </blockquote>
        </div>
        <br>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
yt-users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a>
<a class="moz-txt-link-freetext" href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>