{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Please download this notebook *(use the links above to find the source file)* and run the following installation script as a notebook.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Running MAGEMin calculations through Julia\n",
    "\n",
    "PetThermoTools provides the ability to run calculations/models that utilize the recent Weller et al. (2024) and Green et al. (2025) thermodynamic models. Once everything is set up you should be able to run these calculations from the same functions used to run models with alphaMELTS, but simply setting the ```Model``` kwarg to ```Green2025``` or ```Weller2024```.\n",
    "\n",
    "I've worked hard to try and ensure a set up routine that works on every operating system, regardless of prior installations etc. There is, however, a good chance you'll come across a bug/issue I haven't seen before. If there is an issue in the following script/functions please email me (gleesonm@berkeley.edu) and I will try to find a fix as quickly as possible!\n",
    "\n",
    "The following notebook and functions will establish a new julia environment on your machine. From PetThermoTools v0.2.51 and later this environment will be activated whenever a MAGEMin calculation is initialized from Python. This should ensure that any prior julia installations/packages on your machine are not influenced by the following installation routine.\n",
    "\n",
    "### Step 1 - install juliacall\n",
    "Uncomment the line below and install juliacall in your Python environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%pip install juliacall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 2 - Install MAGEMin\n",
    "Import PetThermoTools (install it if you haven't done so yet) and then use the inbuilt functions to install MAGEMin_C and MAGEMinCalc using the function below. You will need top uncomment the function to run it. If you've previously run this section successfully import PetThermoTools and then go to **Step 2b** to update the MAGEMinCalc and MAGEMin_C installations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# %pip install petthermotools #### If running on Victor uncomment this line to install petthermotools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0.2.50'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import petthermotools as ptt\n",
    "ptt.__version__  # should ve v0.3.2 or later"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**This code installs MAGEMinCalc and underlying dependencies in julia and links them to the PetThermoTools installation. In a couple of cases the code below has caused the Python kernel to crash AFTER it completed. Therefore if a kernel crash message appears don't worry, just restart the kernel and procede to the test_MAGEMinCalc() function below (step 3).**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Detected IPython. Loading juliacall extension. See https://juliapy.github.io/PythonCall.jl/stable/compat/#IPython\n"
     ]
    }
   ],
   "source": [
    "# ptt.install_MAGEMinCalc()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 2b - Update MAGEMin\n",
    "If you've previously run the cell above but need to update the MAGEMinCalc and MAGEMin_C versions following a PetThermoTools update, uncomment and run the cell below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ptt.update_MAGEMinCalc()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3 - test the installation\n",
    "Test the installation by running the following cell. This is calling the MAGEMinCalc functions directly with Julia code using juliacall. This is different to how the code is usually initiated in PetThermoTools but it provides an initial test of the installation. If successful it should print a Pandas DataFrame containing the predicted melt compositions in this crystallization example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Detected IPython. Loading juliacall extension. See https://juliapy.github.io/PythonCall.jl/stable/compat/#IPython\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  Activating project at `~/.petthermotools_julia_env`\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using libMAGEMin.dylib from MAGEMin_jll\n",
      "ig\n",
      "        SiO2     Al2O3       CaO       MgO       FeO       K2O      Na2O  \\\n",
      "0   0.476276  0.164428  0.116295  0.094047  0.091816  0.003300  0.022550   \n",
      "1   0.476311  0.164655  0.116482  0.093537  0.091797  0.003303  0.022572   \n",
      "2   0.476372  0.164888  0.116645  0.093026  0.091774  0.003308  0.022591   \n",
      "3   0.476457  0.165156  0.116821  0.092516  0.091697  0.003310  0.022631   \n",
      "4   0.476543  0.165356  0.116959  0.092003  0.091674  0.003320  0.022656   \n",
      "5   0.476600  0.165608  0.117127  0.091486  0.091618  0.003329  0.022694   \n",
      "6   0.476679  0.165815  0.117307  0.090978  0.091546  0.003333  0.022732   \n",
      "7   0.476789  0.166014  0.117475  0.090462  0.091522  0.003335  0.022770   \n",
      "8   0.476880  0.166223  0.117647  0.089953  0.091484  0.003334  0.022805   \n",
      "9   0.476950  0.166455  0.117817  0.089452  0.091448  0.003335  0.022832   \n",
      "10  0.477000  0.166718  0.117983  0.088940  0.091416  0.003341  0.022873   \n",
      "11  0.477076  0.166931  0.118146  0.088434  0.091400  0.003346  0.022902   \n",
      "12  0.477175  0.167173  0.118310  0.087938  0.091324  0.003351  0.022929   \n",
      "13  0.477512  0.166383  0.118264  0.087365  0.091864  0.003391  0.023051   \n",
      "14  0.477879  0.165479  0.118174  0.086766  0.092514  0.003433  0.023204   \n",
      "15  0.478265  0.164558  0.118119  0.086183  0.093117  0.003471  0.023343   \n",
      "16  0.478611  0.163675  0.118050  0.085598  0.093742  0.003508  0.023492   \n",
      "17  0.478997  0.162752  0.118011  0.085013  0.094356  0.003543  0.023624   \n",
      "18  0.479347  0.161876  0.117955  0.084430  0.094978  0.003584  0.023763   \n",
      "19  0.479771  0.161012  0.117863  0.083845  0.095535  0.003623  0.023925   \n",
      "20  0.480103  0.160162  0.117809  0.083265  0.096122  0.003669  0.024066   \n",
      "21  0.480534  0.159303  0.117741  0.082700  0.096644  0.003703  0.024214   \n",
      "22  0.480895  0.158471  0.117714  0.082138  0.097164  0.003750  0.024355   \n",
      "23  0.481259  0.157630  0.117662  0.081582  0.097704  0.003789  0.024484   \n",
      "24  0.481612  0.156804  0.117614  0.081026  0.098244  0.003825  0.024617   \n",
      "25  0.481952  0.155991  0.117584  0.080473  0.098763  0.003862  0.024749   \n",
      "26  0.482334  0.155181  0.117545  0.079926  0.099240  0.003904  0.024882   \n",
      "27  0.482719  0.154380  0.117509  0.079371  0.099720  0.003947  0.025018   \n",
      "28  0.483126  0.153597  0.117465  0.078832  0.100174  0.003983  0.025159   \n",
      "29  0.483514  0.152812  0.117411  0.078288  0.100661  0.004017  0.025291   \n",
      "30  0.483862  0.152069  0.117362  0.077749  0.101109  0.004057  0.025432   \n",
      "31  0.484211  0.151313  0.117314  0.077214  0.101582  0.004092  0.025564   \n",
      "32  0.484575  0.150554  0.117281  0.076686  0.102022  0.004128  0.025697   \n",
      "33  0.484889  0.149806  0.117269  0.076159  0.102501  0.004163  0.025813   \n",
      "34  0.485042  0.149335  0.116734  0.075553  0.103250  0.004211  0.026014   \n",
      "35  0.485015  0.149040  0.115916  0.074903  0.104195  0.004276  0.026244   \n",
      "36  0.484982  0.148746  0.115101  0.074247  0.105152  0.004341  0.026478   \n",
      "37  0.485004  0.148464  0.114279  0.073590  0.106026  0.004408  0.026728   \n",
      "38  0.485038  0.148184  0.113454  0.072940  0.106889  0.004477  0.026976   \n",
      "39  0.485008  0.147910  0.112657  0.072295  0.107785  0.004542  0.027209   \n",
      "40  0.484987  0.147646  0.111858  0.071646  0.108649  0.004603  0.027460   \n",
      "41  0.484978  0.147392  0.111066  0.070993  0.109497  0.004671  0.027712   \n",
      "42  0.484962  0.147126  0.110273  0.070343  0.110348  0.004733  0.027960   \n",
      "43  0.484908  0.146875  0.109500  0.069696  0.111193  0.004805  0.028195   \n",
      "44  0.484893  0.146627  0.108723  0.069059  0.111993  0.004873  0.028432   \n",
      "45  0.484844  0.146382  0.107957  0.068416  0.112815  0.004937  0.028674   \n",
      "46  0.484855  0.146123  0.107177  0.067779  0.113597  0.005016  0.028898   \n",
      "47  0.484798  0.145888  0.106427  0.067160  0.114373  0.005075  0.029125   \n",
      "\n",
      "        TiO2         O  Cr2O3       H2O  \n",
      "0   0.022949  0.001523    0.0  0.006816  \n",
      "1   0.022999  0.001524    0.0  0.006820  \n",
      "2   0.023036  0.001527    0.0  0.006833  \n",
      "3   0.023041  0.001530    0.0  0.006842  \n",
      "4   0.023102  0.001532    0.0  0.006855  \n",
      "5   0.023143  0.001535    0.0  0.006859  \n",
      "6   0.023204  0.001537    0.0  0.006869  \n",
      "7   0.023219  0.001539    0.0  0.006876  \n",
      "8   0.023247  0.001540    0.0  0.006887  \n",
      "9   0.023271  0.001543    0.0  0.006897  \n",
      "10  0.023279  0.001545    0.0  0.006906  \n",
      "11  0.023301  0.001547    0.0  0.006916  \n",
      "12  0.023329  0.001548    0.0  0.006923  \n",
      "13  0.023601  0.001566    0.0  0.007003  \n",
      "14  0.023880  0.001584    0.0  0.007088  \n",
      "15  0.024170  0.001602    0.0  0.007173  \n",
      "16  0.024446  0.001622    0.0  0.007259  \n",
      "17  0.024727  0.001639    0.0  0.007337  \n",
      "18  0.024990  0.001658    0.0  0.007420  \n",
      "19  0.025249  0.001675    0.0  0.007502  \n",
      "20  0.025529  0.001692    0.0  0.007581  \n",
      "21  0.025786  0.001709    0.0  0.007665  \n",
      "22  0.026039  0.001728    0.0  0.007747  \n",
      "23  0.026318  0.001747    0.0  0.007826  \n",
      "24  0.026587  0.001766    0.0  0.007905  \n",
      "25  0.026857  0.001783    0.0  0.007985  \n",
      "26  0.027125  0.001800    0.0  0.008064  \n",
      "27  0.027380  0.001815    0.0  0.008141  \n",
      "28  0.027608  0.001832    0.0  0.008225  \n",
      "29  0.027855  0.001848    0.0  0.008303  \n",
      "30  0.028111  0.001864    0.0  0.008385  \n",
      "31  0.028366  0.001882    0.0  0.008463  \n",
      "32  0.028620  0.001901    0.0  0.008535  \n",
      "33  0.028872  0.001919    0.0  0.008610  \n",
      "34  0.029197  0.001941    0.0  0.008722  \n",
      "35  0.029589  0.001966    0.0  0.008856  \n",
      "36  0.029977  0.001991    0.0  0.008987  \n",
      "37  0.030368  0.002011    0.0  0.009121  \n",
      "38  0.030751  0.002035    0.0  0.009256  \n",
      "39  0.031142  0.002059    0.0  0.009393  \n",
      "40  0.031538  0.002082    0.0  0.009529  \n",
      "41  0.031921  0.002104    0.0  0.009667  \n",
      "42  0.032327  0.002127    0.0  0.009801  \n",
      "43  0.032743  0.002149    0.0  0.009937  \n",
      "44  0.033147  0.002172    0.0  0.010079  \n",
      "45  0.033565  0.002195    0.0  0.010217  \n",
      "46  0.033984  0.002217    0.0  0.010354  \n",
      "47  0.034410  0.002242    0.0  0.010502  \n"
     ]
    }
   ],
   "source": [
    "import petthermotools as ptt\n",
    "ptt.test_MAGEMinCalc()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 4 - Restart the kernel and test running calculations through PetThermoTools\n",
    "Assuming the crystallization model above ran without any issues we can now progress to testing the integration of MAGEMin with PetThermoTools. First make sure you have the up to date version of PetThermoTools installed (v0.2.51 or later). Restart the kernel and then run the 3 cells below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0.2.50'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import petthermotools as ptt\n",
    "ptt.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running Green2025 calculation..."
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Complete (time taken = 37.55 seconds)"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  Activating project at `~/.petthermotools_julia_env`\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Detected IPython. Loading juliacall extension. See https://juliapy.github.io/PythonCall.jl/stable/compat/#IPython\n",
      "Using libMAGEMin.dylib from MAGEMin_jll\n",
      "ig\n"
     ]
    }
   ],
   "source": [
    "comp = {\"SiO2_Liq\": 47.5, \n",
    "        \"Al2O3_Liq\": 16.4, \n",
    "        \"CaO_Liq\": 11.6, \n",
    "        \"MgO_Liq\": 9.38,\n",
    "        \"FeOt_Liq\": 9.16, \n",
    "        \"K2O_Liq\": 0.329, \n",
    "        \"Na2O_Liq\": 2.25, \n",
    "        \"TiO2_Liq\": 2.29,\n",
    "        \"Fe3Fet_Liq\": 0.15, \n",
    "        \"Cr2O3_Liq\": 0.0, \n",
    "        \"H2O_Liq\": 0.68}\n",
    "\n",
    "Results = ptt.isobaric_crystallisation(bulk=comp,\n",
    "                             Model = \"Green2025\",\n",
    "                             dt_C = 2.0,\n",
    "                             T_end_C=1100.0,\n",
    "                             P_bar=1000.0,\n",
    "                             find_liquidus=True,\n",
    "                             Frac_solid=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'FeO$_t$')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.plot(Results[\"All\"][\"MgO_Liq\"], Results[\"All\"][\"FeOt_Liq\"], '-r')\n",
    "plt.xlabel('MgO')\n",
    "plt.ylabel('FeO$_t$')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 5 - use PetThermoTools alongside MAGEMin in your research!"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.0"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}