new feature: can read Micromeritics xls file as input

betsi/gui.py

@@ -89,7 +89,7 @@ def import_file(self):
         ## file_path = QFileDialog.getOpenFileName(
         ##     self, 'Select File', os.getcwd(), '*.csv')[0]
         file_path = QFileDialog.getOpenFileName(
-            self, 'Select File', os.getcwd(), "*.csv *.aif *.txt")[0]
+            self, 'Select File', os.getcwd(), "*.csv *.aif *.txt *.XLS")[0]
         self.betsi_widget.target_filepath = file_path
         self.betsi_widget.populate_table(csv_path=file_path)
         print(f'Imported file: {Path(file_path).name}')
@@ -103,7 +103,7 @@ def dragEnterEvent(self, e):
         # accept files only for now.
 
         ##if all(dt == 'file' for dt in drag_type) and all(ext == '.csv' for ext in extensions):
-        if all(dt == 'file' for dt in drag_type) and all(ext in ['.csv', '.aif', '.txt'] for ext in extensions):
+        if all(dt == 'file' for dt in drag_type) and all(ext in ['.csv', '.aif', '.txt', '.XLS'] for ext in extensions):
             e.accept()
         elif len(drag_type) == 1 and os.path.isdir(paths[0]):
             e.ignore()
@@ -118,7 +118,7 @@ def dropEvent(self, e):
         # Single path to csv file
 
         ##if len(paths) == 1 and Path(paths[0]).suffix == '.csv':
-        if len(paths) == 1 and Path(paths[0]).suffix in ['.csv', '.aif', '.txt']:
+        if len(paths) == 1 and Path(paths[0]).suffix in ['.csv', '.aif', '.txt', '.XLS']:
             self.betsi_widget.target_filepath = paths[0]
             self.betsi_widget.populate_table(csv_path=paths[0])
             print(f'Imported file: {Path(paths[0]).name}')
@@ -188,9 +188,9 @@ def __init__(self, parent=None):
 
         # add a group box containing controls
         self.criteria_box = QGroupBox("BET area selection criteria")
-        self.criteria_box.setMaximumWidth(500)
-        self.criteria_box.setMaximumHeight(800)
-        self.criteria_box.setMinimumWidth(400)
+        self.criteria_box.setMaximumWidth(700)
+        self.criteria_box.setMaximumHeight(1000)
+        self.criteria_box.setMinimumWidth(500)
         #self.criteria_box.setMinimumHeight(650)
         self.min_points_label = QLabel(self.criteria_box)
         self.min_points_label.setText('Minimum number of points in the linear region: [3,10]')
@@ -205,14 +205,14 @@ def __init__(self, parent=None):
         self.rouq2_tick = QCheckBox("Rouquerol criterion 2: Positive C")
         self.rouq3_tick = QCheckBox("Rouquerol criterion 3: Pressure in linear range")
         self.rouq4_tick = QCheckBox("Rouquerol criterion 4: Error in %, [5,75]")
-        self.rouq5_tick = QCheckBox("Rouquerol criterion 5: End at the knee")
+        self.rouq5_tick = QCheckBox("BETSI criterion: End at the knee")
         self.rouq4_edit = QLineEdit()
         self.rouq4_edit.setMaximumWidth(75)
         self.rouq4_slider = QSlider(QtCore.Qt.Horizontal)
 
         self.adsorbate_label = QLabel('Adsorbate:')
         self.adsorbate_combo_box = QComboBox()
-        self.adsorbate_combo_box.addItems(["N2", "Ar", "Kr", "Xe", "CO2", "Custom"])
+        self.adsorbate_combo_box.addItems(["N2", "Ar", "Kr", "Xe", "Custom"])
         self.adsorbate_cross_section_label = QLabel('Cross sectional area (nm<sup>2</sup>):')
         self.adsorbate_cross_section_edit = QLineEdit()
         self.adsorbate_cross_section_edit.setMaximumWidth(75)
@@ -223,9 +223,17 @@ def __init__(self, parent=None):
             """Hints:
    - For convenience, it is best to first set your desired criteria before importing the input file.
    - Drag and drop the input file into the BETSI window.
-   - Valid input file formats: *.csv, *.txt, *.aif
+   - Valid input file formats: 
+     # Adsorption Information File: *.aif
+     # Two-column data files: *.csv, *.txt
+     # Micromeritics: *.XLS
+   - Valid value range for parameters are given in brackets "[ ]"
    - Make sure to read the warnings that may pop up after BET calculation.  
    - After the first run, by modifiying any of the parameters above, the calculations will rerun automatically.
+   - Regarding the minimum number of points, Rouquerol suggested 10 points, but you can lower the number if the data has insufficient number of points.
+   - Units: "Relative pressure" is dimensionless and "Quantity adsorbed" is in (cm\u00B3 STP/g).
+   - When the calculation is done, by clicking on any points on the "Filtered BET Areas" plot, all the plots will change to the corresponding selected points range.
+   - If the calculation takes longer than expected, bear with it. In case you see "Not responding", it means it is still running, otherwise the software would crash.
    """)
         self.note_label.setStyleSheet("background-color: lightblue")
         self.note_label.setMargin(10)
@@ -411,7 +419,7 @@ def run_calculation(self):
 
         ## assert self.target_filepath, "You must provide a csv file before calling run."
         if not self.target_filepath:
-            warnings = "You must provide an input file (e.g. *.csv, *.txt, *.aif) before calling run. Press \"Clear\" and try again. Please refer to the \"Hints\" box for a quick guide."
+            warnings = "You must provide an input file (e.g. *.csv, *.txt, *.aif, *.XLS) before calling run. Press \"Clear\" and try again. Please refer to the \"Hints\" box for a quick guide."
             information = ""
             self.show_dialog(warnings, information)
             return
@@ -599,7 +607,8 @@ def analyse_directory(self, dir_path):
         csv_paths = Path(dir_path).glob('*.csv')
         aif_paths = Path(dir_path).glob('*.aif')
         txt_paths = Path(dir_path).glob('*.txt')
-        input_file_paths = (*csv_paths, *aif_paths, txt_paths)
+        XLS_paths = Path(dir_path).glob('*.XLS')
+        input_file_paths = (*csv_paths, *aif_paths, *txt_paths, *XLS_paths)
 
         ##for file_path in csv_paths:
         for file_path in input_file_paths:
@@ -629,7 +638,7 @@ def set_defaults(self):
         self.rouq2_tick.setCheckState(True)
         self.rouq3_tick.setCheckState(True)
         self.rouq4_tick.setCheckState(True)
-        self.rouq5_tick.setCheckState(False)
+        self.rouq5_tick.setCheckState(True)
 
         # the ticks can only be on or off - Not sure why I need to do this every time, but doesn't matter
         self.rouq1_tick.setTristate(False)
@@ -772,7 +781,7 @@ def clean_table(self):
         self.results_table.setColumnCount(2)
         self.results_table.setRowCount(0)
         self.results_table.setHorizontalHeaderLabels(
-            ['Relative pressure (p/p\u2080)', 'Quantity adsorbed (cm\u00B3/g)'])
+            ['Relative pressure (p/p\u2080)', 'Quantity adsorbed (cm\u00B3 STP/g)'])
         self.results_table.setColumnWidth(0, 250)
         self.results_table.setColumnWidth(1, 250)
 

betsi/lib.py

@@ -175,9 +175,10 @@ def distance_to_pchip(_s,_mono):
                 self.pc_error[i, j] = (
                     self.error[i, j] / self.corresponding_pressure_pchip[i,j]) * 100.
 
-                # ROUQUEROL CRITERIA 5. Linear region must end at the knee
-                if j == self.knee_index:
-                    self.rouq5[i, j] = 1
+                ### ROUQUEROL CRITERIA 5. Linear region must end at the knee
+                ##if j == self.knee_index:
+                ##    self.rouq5[i, j] = 1
+                self.rouq5[i, j] = 1
 
 class BETFilterAppliedResults:
     """
@@ -209,9 +210,12 @@ def __init__(self, bet_result, **kwargs):
             max_perc_error = kwargs.get('max_perc_error', 20)
             filter_mask = filter_mask * (bet_result.pc_error < max_perc_error)
 
-        if kwargs.get('use_rouq5', False):
+        ##if kwargs.get('use_rouq5', False):
+        ##    filter_mask = filter_mask * bet_result.rouq5
+        if kwargs.get('use_rouq5', True):
             filter_mask = filter_mask * bet_result.rouq5
 
+        self.use_rouq5 = kwargs.get('use_rouq5', True)
         adsorbate = kwargs.get('adsorbate', "N2")
 
         if adsorbate == "Custom":
@@ -272,8 +276,11 @@ def __init__(self, bet_result, **kwargs):
             filtered_pcerrors = bet_result.pc_error + 1000.0 * (1 - filter_mask)
             knee_filtered_pcerrors = bet_result.pc_error + \
                 1000.0 * (1 - knee_filter)
-            min_i, min_j = np.unravel_index(
-                np.argmin(knee_filtered_pcerrors), filtered_pcerrors.shape)
+
+            if self.use_rouq5:
+                min_i, min_j = np.unravel_index(np.argmin(knee_filtered_pcerrors), filtered_pcerrors.shape)
+            else:
+                min_i, min_j = np.unravel_index(np.argmin(filtered_pcerrors), filtered_pcerrors.shape)
             self.min_i = min_i
             self.min_j = min_j
 

betsi/utils.py

@@ -3,12 +3,13 @@
 """
 from pathlib import Path
 
+import pandas as pd
 import numpy as np
 from scipy.interpolate import splrep, PchipInterpolator, pchip_interpolate
 
 
 def get_data(input_file):
-    """Read pressure and Nitrogen uptake data from file. Asserts pressures in units of bar.
+    """Read pressure and adsorbate uptake data from file. Asserts pressures in units of bar.
 
     Args:
         input_file: Path the path to the input file
@@ -19,7 +20,9 @@ def get_data(input_file):
     input_file = Path(input_file)
 
     if str(input_file).find('.txt') != -1 or str(input_file).find('.aif') != -1 or str(input_file).find('.csv') != -1:
-        pressure, q_adsorbed = get_data_for_txt_aif_csv(str(input_file))
+        pressure, q_adsorbed = get_data_for_txt_aif_csv_two_columns(str(input_file))
+    elif str(input_file).find('.XLS') != -1:
+        pressure, q_adsorbed = get_data_from_micromeritics_xls(str(input_file))
     else:
         pressure = np.array([])
         q_adsorbed = np.array([])
@@ -84,7 +87,7 @@ def get_fitted_spline(pressure, q_adsorbed):
 
     Args:
         pressure: Array of relative pressure values.
-        q_adsorbed: Array of Nitrogen uptake values.
+        q_adsorbed: Array of adsorbate uptake values.
 
     Returns:
         tck tuple of spline parameters.
@@ -96,7 +99,7 @@ def get_pchip_interpolation(pressure,q_adsorbed):
     
     Args:
         pressure: Array of relative pressure values
-        q_adsorbed: Array of Nitrogen uptake values
+        q_adsorbed: Array of adsorbate uptake values
 
     Returns:
         Pchip parameters
@@ -108,9 +111,9 @@ def isotherm_pchip_reconstruction(pressure, q_adsorbed, num_of_interpolated_poin
     calculate BET area for difficult isotherms
     
     Args: pressure: Array of relative pressure values
-    q_adsorbed: Array of Nitrogen uptake values
+    q_adsorbed: Array of adsorbate uptake values
     
-    Returns: Array of interpolated nitrogen uptake values
+    Returns: Array of interpolated adsorbate uptake values
     
     """
     ## x_range = np.linspace(pressure[0], pressure[len(pressure) -1 ], 500)
@@ -144,8 +147,8 @@ def isotherm_pchip_reconstruction(pressure, q_adsorbed, num_of_interpolated_poin
 
     return pressure_new, q_adsorbed_new
 
-def get_data_for_txt_aif_csv(input_file):
-    """ Read pressure and Nitrogen uptake data from file if the file extension is *.txt or *.aif.    
+def get_data_for_txt_aif_csv_two_columns(input_file):
+    """ Read pressure and adsorbate uptake data from file if the file extension is *.txt or *.aif.    
     this function will be called in the get_data() function, and should not be called alone anywhere in the code
     as it might cause some errors.
     
@@ -220,4 +223,92 @@ def get_data_for_txt_aif_csv(input_file):
     pressure = np.array(pressure)
     q_adsorbed = np.array(q_adsorbed)
 
+    return pressure, q_adsorbed
+
+def get_data_from_micromeritics_xls(input_file):
+    """ Read pressure and adsorbate uptake data from Micromeritics output files with *.XLS extension    
+    this function will be called in the get_data() function, and should not be called alone anywhere in the code
+    as it might cause some errors.
+    
+    Args:
+        input_file: String of the path to the input file
+
+    Returns:
+        Pressure and Quantity adsorbed.
+    """
+
+    # pressure = []
+    # q_adsorbed = []
+
+    excel_file = pd.ExcelFile(input_file)
+
+    if len(excel_file.sheet_names) == 1:
+        excel_df = pd.read_excel(input_file, sheet_name=0)
+        column_num = 0
+        for column in excel_df.columns:
+            if len(excel_df.loc[excel_df[column]=='Isotherm Linear Plot']) > 0:
+                keyword_column_name = column
+                keyword_index = excel_df.loc[excel_df[column]=='Isotherm Linear Plot'].index[0]
+
+                adsorption_data_start_index = 0
+                for row_num in range(keyword_index+1, len(excel_df.index)):
+                    if excel_df[keyword_column_name][row_num] == "Relative Pressure (p/p°)":
+                        adsorption_data_start_index = row_num + 1
+
+                nan_indexes_in_column = np.array(excel_df.loc[excel_df[column].isnull()].index)
+                adsorption_data_stop_index = nan_indexes_in_column[np.argmax(nan_indexes_in_column > adsorption_data_start_index)] - 1
+
+                if adsorption_data_stop_index < adsorption_data_start_index:
+                    adsorption_data_stop_index = len(excel_df.index) - 1
+
+                if adsorption_data_start_index > 0:
+                    pressure = np.array(excel_df[keyword_column_name][adsorption_data_start_index:adsorption_data_stop_index+1])
+                    q_adsorbed = np.array(excel_df[excel_df.columns[column_num+1]][adsorption_data_start_index:adsorption_data_stop_index+1])
+                else:
+                    pressure = np.array([])
+                    q_adsorbed = np.array([])
+
+                break
+            column_num += 1
+
+    elif len(excel_file.sheet_names) == 0:
+        pressure = np.array([])
+        q_adsorbed = np.array([])
+    else:
+        try:
+            excel_df = pd.read_excel(input_file, sheet_name="Isotherm Linear Plot")
+            column_num = 0
+            for column in excel_df.columns:
+                if len(excel_df.loc[excel_df[column]=='Isotherm Linear Plot']) > 0:
+                    keyword_column_name = column
+                    keyword_index = excel_df.loc[excel_df[column]=='Isotherm Linear Plot'].index[0]
+
+                    adsorption_data_start_index = 0
+                    for row_num in range(keyword_index+1, len(excel_df.index)):
+                        if excel_df[keyword_column_name][row_num] == "Relative Pressure (p/p°)":
+                            adsorption_data_start_index = row_num + 1
+
+                    nan_indexes_in_column = np.array(excel_df.loc[excel_df[column].isnull()].index)
+                    adsorption_data_stop_index = nan_indexes_in_column[np.argmax(nan_indexes_in_column > adsorption_data_start_index)] - 1
+
+                    if adsorption_data_stop_index < adsorption_data_start_index:
+                        adsorption_data_stop_index = len(excel_df.index) - 1
+
+                    if adsorption_data_start_index > 0:
+                        pressure = np.array(excel_df[keyword_column_name][adsorption_data_start_index:adsorption_data_stop_index+1])
+                        q_adsorbed = np.array(excel_df[excel_df.columns[column_num+1]][adsorption_data_start_index:adsorption_data_stop_index+1])
+                    else:
+                        pressure = np.array([])
+                        q_adsorbed = np.array([])
+
+                    break
+                column_num += 1
+
+        except ValueError:
+            pressure = np.array([])
+            q_adsorbed = np.array([])
+
+    pressure = np.float64(pressure)
+    q_adsorbed = np.float64(q_adsorbed)
+
     return pressure, q_adsorbed

requirements.txt

@@ -4,4 +4,5 @@ matplotlib==3.2.2
 PyQt5==5.9.2
 pandas==1.1.5
 seaborn==0.11.0
-statsmodels==0.12.1
+statsmodels==0.12.1
+xlrd==2.0.1