changed environment workflow to user-oriented setup.

Project.toml

@@ -18,6 +18,7 @@ Logging = "56ddb016-857b-54e1-b83d-db4d58db5568"
 NPZ = "15e1cf62-19b3-5cfa-8e77-841668bca605"
 NaNMath = "77ba4419-2d1f-58cd-9bb1-8ffee604a2e3"
 PDMats = "90014a1f-27ba-587c-ab20-58faa44d9150"
+Pkg = "44cfe95a-1eb2-52ea-b672-e2afdf69b78f"
 Printf = "de0858da-6303-5e67-8744-51eddeeeb8d7"
 PyCall = "438e738f-606a-5dbb-bf0a-cddfbfd45ab0"
 PyPlot = "d330b81b-6aea-500a-939a-2ce795aea3ee"

docs/src/installation.md

@@ -2,25 +2,42 @@
 
 ## Julia
 
-First, you should to install the [Julia programming language](http://julialang.org/downloads/) on your machine. 
+First, you should to install the [Julia programming language](http://julialang.org/downloads/) on your machine. Depending on how you choose to install Julia, you may need to take the additional step of adding the Julia executable to your `PATH`.
 
-Depending on how you choose to install Julia, you may need to take the additional step of adding the Julia executable to your `PATH`. [Here](https://en.wikibooks.org/wiki/Introducing_Julia/Getting_started#Running_directly_from_terminal) are some suggestions for OS X if you are experiencing difficulty.
+Julia uses "environments" to manage package dependencies. These are similar in concept to Python's virtual environments, and very useful in practice. If you are unfamiliar, it is highly recommended to read through the Julia environment documentation [here](https://docs.julialang.org/en/v1/stdlib/Pkg/), since they are used to install DiskJockey in the recommended configuration. 
 
-Julia uses "environments" to manage package dependencies. If you are unfamiliar, it is worth reading through the documentation on how Julia's package manager works with environments [here](https://docs.julialang.org/en/v1/stdlib/Pkg/). 
+To get started, create a new folder that you will contain your work for a particular project.
 
-## DiskJockey
+    $ mkdir myproject 
 
-`DiskJockey` is not (yet) registered an official Julia package, so it needs to be installed by using the link to the github repository. First, open up a Julia prompt in the REPL, then type `]` to enter the package manager, then use the `add` command to clone the DiskJockey repository.
+Then, start Julia, and use the interactive package manager `Pkg` (enter via `]`) to create an environment specific to this project. 
 
-    julia> ] 
-    (@v1.4) pkg> add https://github.com/iancze/DiskJockey.git
+    $ cd myproject
+    $ julia 
+    julia> ]
+    pkg> activate . 
+    Activating new environment at `~/myproject/Project.toml`
+    (myproject) pkg>
+
+## DiskJockey Package
+
+Next, we will install the DiskJockey package to this same Julia environment. `DiskJockey` is not (yet) registered an official Julia package, so it needs to be installed by using the link to the github repository. 
+
+    (myproject) pkg> add https://github.com/iancze/DiskJockey.git
 
-This process may take a few minutes as the relevant packages (including RADMC-3D) are downloaded from the web and installed. If you run into errors in this build process, please file an [issue](https://github.com/iancze/DiskJockey/issues) on the github repository so that we may try to fix this. If you already have RADMC-3D installed on your system, this process won't interfere with that installation, `DiskJockey` will use the version of RADMC-3D it downloaded during the build process. If you anticipate editing the source code, rather than installing the package with the `add` command, you can install using the `develop` command
+This process may take a few minutes as the relevant packages (including RADMC-3D) are downloaded from the web and installed. If you run into errors in this build process, please file an [issue](https://github.com/iancze/DiskJockey/issues) on the github repository so that we may try to fix this. If you already have RADMC-3D installed on your system, this process won't interfere with that installation, `DiskJockey` will use the version of RADMC-3D it downloaded during the build process. 
 
-    julia> ] 
-    (@v1.4) pkg> develop https://github.com/iancze/DiskJockey.git
+It is a good idea to set the environment variable `JULIA_PROJECT` to the location for this project
 
-Included within the DiskJockey github repo are also several "driver" command line scripts that are used to perform the actual mass fitting. To utilize these files, you should add the scripts directory to your system PATH. To figure out where the package is installed
+    export JULIA_PROJECT='/home/ian/myproject'
+
+You can add this line to your `.bashrc` or shell startup, or local directory initialization script. That way, whenever you start Julia, it will already activate the myproject environment. This makes running the scripts in the next section much easier. Advanced users with multiple active Julia projects may wish to explore alternate package activation patterns.
+
+## DiskJockey Scripts
+
+### Julia Scripts
+
+Included within the DiskJockey github repo are several "driver" command line scripts in the `DiskJockey/scripts` directory that you can use to perform the actual mass fitting and analysis plotting. To utilize these files, you should add the `scripts` directory to your system PATH. To figure out where the package is installed
 
     julia> using DiskJockey
     julia> dirname(pathof(DiskJockey))
@@ -32,13 +49,19 @@ Your PATH will vary. The scripts are located inside of the `DiskJockey/scripts`
 
 inside of your `.bashrc` file.
 
+The scripts depend on other Julia packages for reading configuration files (in YAML) and plotting. You can either install these packages to your `myproject` environment just like you did DiskJockey, or you can use the convenience script 
+
+    $ DJ_initialize_environment.jl 
+
+
+### Python Scripts
+
 Some of the Python scripts also depend on imports from this directory, so you will also need to add
 
     export PYTHONPATH="/home/ian/.julia/dev/DiskJockey/scripts:$PYTHONPATH"
 
 inside of your `.bashrc` file. 
 
-For archival purposes, tagged release versions of this package are available [here](https://github.com/iancze/DiskJockey/releases).
 
 ## Python scripts
 
@@ -63,3 +86,14 @@ If you'd like to run the test suite to make sure everything checks out, start Ju
 this may take about 10 minutes or so. If you catch any errors for your specific machine, please report them via an Issue on the github repository.
 
 The code package is designed to interface with visibilities in the UVHDF5 format, described [here](https://github.com/Astrochem/UVHDF5). UVHDF5 also provides scripts to convert to and from UVFITS and CASA measurement sets.
+
+
+## Development workflow
+
+If you anticipate editing the source code, rather than installing the package with the `add` command, you can install using the `develop` command
+
+    julia> ] 
+    (@v1.4) pkg> develop https://github.com/iancze/DiskJockey.git
+
+
+For archival purposes, tagged release versions of this package are available [here](https://github.com/iancze/DiskJockey/releases).

scripts/DJ_initialize.jl

@@ -4,11 +4,9 @@
 # files into this directory, write out the proper wavelength files, and make sure we can continue
 # with the run properly.
 
-using Pkg; Pkg.activate("DiskJockey")
-
 using ArgParse
 
-s = ArgParseSettings(description="Initialize a new project directory with the appropriate files. Can also be used to update RADMC-3D input files after making changes to config.yaml")
+s = ArgParseSettings(description = "Initialize a new project directory with the appropriate files. Can also be used to update RADMC-3D input files after making changes to config.yaml")
 @add_arg_table s begin
     "--version"
     help = "Print a the version number and exit."
@@ -109,10 +107,10 @@ if fit_every != 0
     nchan = length(dvarr)
     println("Dataset contains a total of $nchan channels.")
 
-    to_fit = Int[i for i=1:fit_every:nchan]
+    to_fit = Int[i for i = 1:fit_every:nchan]
 
     # which channels of the dset to exclude
-    exclude = filter(x->(!in(x, to_fit)), Int[i for i=1:nchan])
+    exclude = filter(x->(!in(x, to_fit)), Int[i for i = 1:nchan])
 
     println("To fit only every $(fit_every)-nd/rd/th channel, place the following line in your `config.yaml file.`")
     println("exclude : $exclude")
@@ -185,20 +183,20 @@ if config["gas"]
         # and create an evenly spaced array of velocities
         vels = linspace(vstart, vend, nvel) # [km/s]
 
-        lam0 = lam0s[species*transition]
+        lam0 = lam0s[species * transition]
 
         # convert velocities to wavelengths
-        shift_lams = lam0 * (vels/c_kms + 1)
+        shift_lams = lam0 * (vels / c_kms + 1)
     else
         # read the wavelengths for all data channels
         fid = h5open(config["data_file"], "r")
         freqs = read(fid["freqs"]) # [Hz]
         # Convert from Hz to wavelengths in μm
-        lams = cc ./freqs * 1e4 # [μm]
+        lams = cc ./ freqs * 1e4 # [μm]
         close(fid)
 
         # Doppler shift the dataset wavelength according to the velocity in the parameter file
-        beta = vel/c_kms # relativistic Doppler formula
+        beta = vel / c_kms # relativistic Doppler formula
         shift_lams =  lams .* sqrt((1. - beta) / (1. + beta)) # [microns]
     end
 

scripts/DJ_initialize_environment.jl

@@ -0,0 +1,19 @@
+#!/usr/bin/env julia
+
+using Pkg
+
+# add the packages to the current environment, defined by JULIA_PROJECT 
+Pkg.add(PackageSpec(url = "https://github.com/iancze/DiskJockey.git"))
+Pkg.add("ArgParse")
+Pkg.add("HDF5")
+Pkg.add("PyPlot")
+Pkg.add("LaTeXStrings")
+Pkg.add("YAML")
+Pkg.add("Printf")
+Pkg.add("PyCall")
+Pkg.add("Images")
+Pkg.add("NaNMath")
+Pkg.add("NPZ")
+Pkg.add("Statistics")
+
+Pkg.instantiate()

scripts/DJ_plot_baselines.jl

@@ -1,5 +1,4 @@
 #!/usr/bin/env julia
-using Pkg; Pkg.activate("DiskJockey")
 
 # Read the data file and plot the location of all of the baselines, in k\lambda
 using ArgParse
@@ -27,7 +26,7 @@ using LaTeXStrings
 
 species = config["species"]
 transition = config["transition"]
-lam0 = lam0s[species*transition]
+lam0 = lam0s[species * transition]
 model = config["model"]
 
 pars = convert_dict(config["parameters"], config["model"])
@@ -36,7 +35,7 @@ pars = convert_dict(config["parameters"], config["model"])
 dvarr = DataVis(config["data_file"])
 dv = dvarr[1]
 
-fig,ax = plt[:subplots](figsize=(6,6))
+fig, ax = plt[:subplots](figsize = (6, 6))
 ax[:plot](dv.uu, dv.vv, ".")
 ax[:set_xlabel](L"UU [k$\lambda$]")
 ax[:set_ylabel](L"VV [k$\lambda$]")

scripts/DJ_plot_chmaps.jl

@@ -1,6 +1,4 @@
 #!/usr/bin/env julia
-using Pkg; Pkg.activate("DiskJockey")
-
 
 # Given some model parameters, plot the channel maps and
 # integrated spectrum
@@ -44,7 +42,7 @@ using LaTeXStrings
 
 species = config["species"]
 transition = config["transition"]
-lam0 = lam0s[species*transition]
+lam0 = lam0s[species * transition]
 model = config["model"]
 
 # Read the inclination, position angle, and (grid) radius of the disk. Plot ellipses.
@@ -57,7 +55,7 @@ vel_sys = pars["vel"]
 
 # from matplotlib.patches import Ellipse
 #     ax[:add_artist](PyPlot.matplotlib[:patches][:Ellipse](xy=xy, width=BMIN, height=BMAJ, angle=BPA, facecolor="0.8", linewidth=0.2))
-width = r_c/dpc # [arcsec]
+width = r_c / dpc # [arcsec]
 height = width * cosd(incl) # [arcsec]
 
 # cmap = plt[:get_cmap]("viridis")
@@ -66,7 +64,7 @@ height = width * cosd(incl) # [arcsec]
 cmap = plt.get_cmap("Blues")
 
 # First contour is at 3 sigma, and then contours go up (or down) in multiples of spacing
-function get_levels(rms::Float64, vmax::Float64, spacing=3)
+function get_levels(rms::Float64, vmax::Float64, spacing = 3)
     levels = Float64[]
 
     val = 3 * rms
@@ -81,7 +79,7 @@ end
 # Plot the channel maps using sky convention
 """Plot the channel maps using the sky convention. If log is true, plot intensity using
 a log scale."""
-function plot_chmaps(img::image.SkyImage; log=false, contours=true, fname="channel_maps_sky.png")
+function plot_chmaps(img::image.SkyImage; log = false, contours = true, fname = "channel_maps_sky.png")
 
     if log
         ldata = log10.(img.data .+ 1e-99)
@@ -104,17 +102,17 @@ function plot_chmaps(img::image.SkyImage; log=false, contours=true, fname="chann
 
     # Figure out how many plots we'll have.
     ncols = 8
-    nrows = ceil(Int, nlam/ncols)
+    nrows = ceil(Int, nlam / ncols)
 
     xx = 1.5 * 9
     dx = 1.5
     yy = (nrows + 1) * 1.5
     dy = 1.5
 
-    fig, ax = plt.subplots(nrows=nrows, ncols=ncols, figsize=(xx, yy))
+    fig, ax = plt.subplots(nrows = nrows, ncols = ncols, figsize = (xx, yy))
 
-    for row=1:nrows
-        for col=1:ncols
+    for row = 1:nrows
+        for col = 1:ncols
             iframe = col + (row - 1) * ncols
 
             if col != 1 || row != nrows
@@ -127,65 +125,65 @@ function plot_chmaps(img::image.SkyImage; log=false, contours=true, fname="chann
 
             if iframe > nlam
                 # Plot a blank square if we run out of channels
-                ax[row, col].imshow(zeros((im_ny, im_nx)), cmap=cmap, vmin=0, vmax=20, extent=ext, origin="lower")
+                ax[row, col].imshow(zeros((im_ny, im_nx)), cmap = cmap, vmin = 0, vmax = 20, extent = ext, origin = "lower")
 
             else
-                #Flip the frame for Sky convention
+                # Flip the frame for Sky convention
                 # frame = flipdim(img.data[:,:,iframe], 2)
-                frame = reverse(img.data[:,:,iframe], dims=2)
+                frame = reverse(img.data[:,:,iframe], dims = 2)
 
                 if log
-                    frame += 1e-15 #Add a tiny bit so that we don't have log10(0)
+                    frame += 1e-15 # Add a tiny bit so that we don't have log10(0)
                     lframe = log10.(frame)
-                    im = ax[row, col].imshow(lframe, extent=ext, interpolation="none", origin="lower", cmap=cmap, norm=norm)
+                    im = ax[row, col].imshow(lframe, extent = ext, interpolation = "none", origin = "lower", cmap = cmap, norm = norm)
 
                 else
-                    im = ax[row, col].imshow(frame, extent=ext, interpolation="none", origin="lower", cmap=cmap, norm=norm)
+                    im = ax[row, col].imshow(frame, extent = ext, interpolation = "none", origin = "lower", cmap = cmap, norm = norm)
 
                     if contours
-                        ax[row, col].contour(frame, origin="lower", colors="k", levels=levels, extent=ext, linestyles="solid", linewidths=0.2)
+                        ax[row, col].contour(frame, origin = "lower", colors = "k", levels = levels, extent = ext, linestyles = "solid", linewidths = 0.2)
                     end
 
                 end
 
-                ax[row, col].add_artist(PyPlot.matplotlib.patches.Ellipse((0,0), width, height, PA, linewidth=0.15, facecolor="none", edgecolor="w"))
+                ax[row, col].add_artist(PyPlot.matplotlib.patches.Ellipse((0, 0), width, height, PA, linewidth = 0.15, facecolor = "none", edgecolor = "w"))
 
-                if iframe==1
+                if iframe == 1
                     # Plot the colorbar
-                    cax = fig.add_axes([(xx - 0.35 * dx)/xx, (yy - 1.5 * dy)/yy, (0.1 * dx)/xx, dy/yy])
-                    cbar = fig.colorbar(mappable=im, cax=cax)
+                    cax = fig.add_axes([(xx - 0.35 * dx) / xx, (yy - 1.5 * dy) / yy, (0.1 * dx) / xx, dy / yy])
+                    cbar = fig.colorbar(mappable = im, cax = cax)
 
-                    cbar.ax.tick_params(labelsize=6)
-                    fig.text(0.99, (yy - 1.7 * dy)/yy, "Jy/beam", size=8, ha="right")
+                    cbar.ax.tick_params(labelsize = 6)
+                    fig.text(0.99, (yy - 1.7 * dy) / yy, "Jy/beam", size = 8, ha = "right")
                 end
 
-                ax[row, col].annotate(@sprintf("%.1f", vels[iframe]), (0.1, 0.8), xycoords="axes fraction", size=8)
-                ax[row, col].annotate(@sprintf("%.1f", vels_obs[iframe]), (0.1, 0.9), xycoords="axes fraction", size=8)
+                ax[row, col].annotate(@sprintf("%.1f", vels[iframe]), (0.1, 0.8), xycoords = "axes fraction", size = 8)
+                ax[row, col].annotate(@sprintf("%.1f", vels_obs[iframe]), (0.1, 0.9), xycoords = "axes fraction", size = 8)
             end
 
         end
     end
 
-    fig.subplots_adjust(hspace=0.00, wspace=0.00, top=(yy - 0.5 * dy)/yy, bottom=(0.5 * dy)/yy, left=(0.5 * dx)/xx, right=(xx - 0.5 * dy)/xx)
+    fig.subplots_adjust(hspace = 0.00, wspace = 0.00, top = (yy - 0.5 * dy) / yy, bottom = (0.5 * dy) / yy, left = (0.5 * dx) / xx, right = (xx - 0.5 * dy) / xx)
 
-    plt.savefig(fname, dpi=600)
+    plt.savefig(fname, dpi = 600)
 
 end
 
 # Plot the spatially-integrated spectrum
-function plot_spectrum(img::image.SkyImage; fname="spectrum.png")
+function plot_spectrum(img::image.SkyImage; fname = "spectrum.png")
 
     fig = plt.figure()
     ax = fig.add_subplot(111)
 
     spec = imToSpec(img)
 
-    ax.plot(vels, spec[:,2], ls="steps-mid")
+    ax.plot(vels, spec[:,2], ls = "steps-mid")
 
     ax.set_ylabel(L"$f_\nu$ [Jy]")
     ax.set_xlabel(L"$v$ [km/s]")
 
-    fig.subplots_adjust(left=0.15, bottom=0.15, right=0.85)
+    fig.subplots_adjust(left = 0.15, bottom = 0.15, right = 0.85)
 
     plt.savefig(fname)
 
@@ -203,7 +201,7 @@ skim = imToSky(im, pars.dpc)
 # Do the velocity conversion here for plot labels in the disk-frame
 global nlam = length(skim.lams)
 # convert wavelengths to velocities
-global vels = c_kms * (skim.lams .- lam0)/lam0
+global vels = c_kms * (skim.lams .- lam0) / lam0
 
 # calculate velocities in the observed frame (whatever the dataset is)
 global vels_obs = vels .+ vel_sys
@@ -217,11 +215,11 @@ if vels[2] < vels[1]
 end
 
 if parsed_args["linear"]
-    plot_chmaps(skim, fname="chmaps_linear.png", contours=false)
+    plot_chmaps(skim, fname = "chmaps_linear.png", contours = false)
 end
 
 if parsed_args["log"]
-    plot_chmaps(skim, fname="chmaps_log.png", log=true, contours=false)
+    plot_chmaps(skim, fname = "chmaps_log.png", log = true, contours = false)
 end
 
 if parsed_args["spectrum"]