Merge pull request #52 from JaimieDodd/feature/inputoutput

Manage calls to GeodePy functions to transform large amounts of data in csv format

geodepy/inputoutput.py

@@ -0,0 +1,192 @@
+
+__all__ = ["grid2geoio", "geo2gridio", "gdatrans7"]
+
+import pandas as pd
+import tkinter as ttk
+import geodepy.transform as tf
+import geodepy.convert as cv
+import geodepy.constants as cs
+
+'''
+The inputoutput module acts as the backend for the GUI and manages the calls to the GeodePy functions. The functions
+are rewritten  from the transform module but use pandas to improve efficiency. Functions placed here to not overcrowd
+the transform module.
+'''
+
+
+def grid2geoio(fn, fn_out, easting, northing, utmzone, geotypeout):
+    """
+        Input:
+         The CSV data must have headers.
+
+         Inputs passed the the function:
+         :param fn: input file path
+         :param fn_out: file output path
+         :param easting: is the column name in the csv where the eastings are stored
+         :param northing: is the column name in the csv where the northings are stored
+         :param utmzone: is the column name in the csv where the UTM Zone is stored
+         :param geotypeout: format of latitude and longitude output e.g. DD or DMS
+
+        Output:
+        Data in the file is output to the fn_out location as a CSV.
+        """
+
+    # Check whether geotypeout value is from GUI or if function called within other code, then check that a valid
+    # geotypeout value was provided
+    if isinstance(geotypeout, ttk.StringVar):
+        geotypeout = geotypeout.get()
+    geotypeout_options = ["DD", "DMS"]
+    if geotypeout not in geotypeout_options:
+        raise ValueError("Invalid geotypeout. Expected one of: %s" % geotypeout_options)
+
+    # Opens the CSV as a pandas dataframe
+    csvdf = pd.read_csv(fn, low_memory=False)
+    # Sends data to tf.grid2geo
+    returned = csvdf.apply(lambda x: tf.grid2geo(x[utmzone], x[easting], x[northing]), axis=1)
+    # Convert tuple returned to DafaFrame
+    resultdf = pd.DataFrame(list(returned), columns=['Latitude', 'Longitude', 'Point Scale Factor', 'Grid Convergence'])
+    if geotypeout == 'DMS':
+        resultdf['Latitude'] = resultdf['Latitude'].apply(cv.dd2dms)
+        resultdf['Longitude'] = resultdf['Longitude'].apply(cv.dd2dms)
+    # Adds the results to the original dataframe from the csv
+    csvdf = pd.concat([csvdf, resultdf], axis=1)
+    # Writes the dataframe to a csv
+    csvdf.to_csv(fn_out, index=False)
+
+
+def geo2gridio(fn, fn_out, latitude, longitude, geotypein):
+    """
+        Input:
+        The CSV data must have headers.
+
+         Inputs passed the the function:
+         :param fn: input file path
+         :param fn_out: file output path
+         :param latitude: is the column name in the csv where the latitudes are stored
+         :param longitude: is the column name in the csv where the longitudes are stored
+         :param geotypein: format of latitude and longitude e.g. DD or DMS
+
+        Output:
+        Data in the file is output to the fn_out location as a CSV.
+        """
+
+    # Check whether geotypein value is from GUI or if function called within other code, then check that a valid
+    # geotypein value was provided
+    if isinstance(geotypein, ttk.StringVar):
+        geotypein = geotypein.get()
+    geotypein_options = ["DD", "DMS"]
+    if geotypein not in geotypein_options:
+        raise ValueError("Invalid geotypein. Expected one of: %s" % geotypein_options)
+
+    # Opens the CSV as a pandas dataframe
+    csvdf = pd.read_csv(fn, low_memory=False)
+
+    # Converts DMS lat and long to Decimal Degrees if required
+    if geotypein == 'DMS':
+        csvdf['LatitudeDD'] = csvdf[latitude].apply(cv.dms2dd)
+        csvdf['LongitudeDD'] = csvdf[longitude].apply(cv.dms2dd)
+        latitude = 'LatitudeDD'
+        longitude = 'LongitudeDD'
+
+    # Sends data to tf.geo2grid
+    returned = csvdf.apply(lambda x: tf.geo2grid(x[latitude], x[longitude]), axis=1)
+    # Convert tuple returned from tf.gepo2grid into DafaFrame
+    resultdf = pd.DataFrame(list(returned), columns=['Hemisphere', 'UTMZone', 'Easting', 'Northing',
+                                                     'Point Scale Factor', 'Grid Convergence'])
+    # Adds the results to the original dataframe from the csv
+    csvdf = pd.concat([csvdf, resultdf], axis=1)
+    # Writes the dataframe to a csv
+    csvdf.to_csv(fn_out, index=False)
+
+
+def gdatrans7(fn, fn_out, latitude, longitude, ellht, gdageotypein, direction):
+
+    """
+    Input:
+     The CSV data must have headers.
+
+     Inputs passed the the function:
+     :param fn: input file path
+     :param fn_out: file output path
+     :param latitude: is the column name in the csv where the latitudes are stored
+     :param longitude: is the column name in the csv where the longitudes are stored
+     :param ellht: is the column name in the csv where the ellipsoidal heights are stored
+     :param gdageotypein: format of latitude and longitude e.g. DD or DMS
+     :param direction: either "94to2020" or "2020to94". Specifies the datum to transform from and to.
+
+     trans: Transformation parameter from the geodepy.constants module. The trans parameters are hard coded below.
+     Parameters were sourced from GDA2020 Technical Manual. Need to work on a way of entering and passing different
+     transformation parameters to the function.
+
+    Output:
+    Data in the file is output to the fn_out location as a CSV.
+    """
+
+    # Check whether direction value is from GUI or if function called within other code, then check that a valid
+    # direction value was provided
+    if isinstance(direction, ttk.StringVar):
+        direction = direction.get()
+    direction_options = ["94to2020", "2020to94"]
+    if direction not in direction_options:
+        raise ValueError("Invalid direction. Expected one of: %s" % direction_options)
+
+    # Check whether gdageotypein value is from GUI or if function called within other code, then check that a valid
+    # gdageotypein value was provided
+    if isinstance(gdageotypein, ttk.StringVar):
+        gdageotypein = gdageotypein.get()
+    gdageotypein_options = ["DD", "DMS"]
+    if gdageotypein not in gdageotypein_options:
+        raise ValueError("Invalid gdageotypein. Expected one of: %s" % gdageotypein_options)
+
+    # Sets up the direction of the transformation and sets up the output column naming
+    if direction == "94to2020":
+        trans = cs.Transformation('MGA94', 'MGA2020', '1994', 0.06155, -0.01087, -0.04019, -0.009994, -0.0394924,
+                                  -0.0327221, -0.0328979)
+        translat = "GDA2020Latitude"
+        translong = "GDA2020Longitude"
+        transellht = "GDA2020EllHt"
+    else:
+        trans = cs.Transformation('MGA94', 'MGA2020', '1994', 0.06155, -0.01087, -0.04019, -0.009994, -0.0394924,
+                                  -0.0327221, -0.0328979).__neg__()
+        translat = "GDA1994Latitude"
+        translong = "GDA1994Longitude"
+        transellht = "GDA1994EllHt"
+
+    # Opens the CSV as a pandas dataframe
+    csvdf = pd.read_csv(fn, low_memory=False)
+
+    # Converts DMS lat and long to Decimal Degrees if required
+    if gdageotypein == 'DMS':
+        csvdf['LatitudeDMS'] = csvdf[latitude]
+        csvdf['LongitudeDMS'] = csvdf[longitude]
+        csvdf['Latitude'] = csvdf[latitude].apply(cv.dms2dd)
+        csvdf['Longitude'] = csvdf[longitude].apply(cv.dms2dd)
+
+    # Converts Lat, Long & ElipHt XYZ
+    returned = csvdf.apply(lambda x: tf.llh2xyz(x[latitude], x[longitude], x[ellht]), axis=1)
+    # Converts the results from llh2xyz into a dataframe
+    xyzresultdf = pd.DataFrame(list(returned), columns=['CartesianX', 'CartesianY', 'CartesianZ'])
+
+    # Combines the results with the original dataframe
+    csvdf = pd.concat([csvdf, xyzresultdf], axis=1)
+
+    # Transforms Lat, Long & ElipHt XYZ using a 7 parameter transformation
+    transformed = csvdf.apply(lambda x: tf.conform7(x.CartesianX, x.CartesianY, x.CartesianZ, trans), axis=1)
+    # Converts the results from conform7 into a dataframe
+    transresultdf = pd.DataFrame(list(transformed), columns=['TransCartesianX', 'TransCartesianY', 'TransCartesianZ'])
+
+    # Combines the results with the original dataframe
+    csvdf = pd.concat([csvdf, transresultdf], axis=1)
+
+    # Converts the the transformed XYZ coordinates back to Lats, Longs and Ellipsoidal Heights
+    returned = csvdf.apply(lambda x: tf.xyz2llh(x.TransCartesianX, x.TransCartesianY, x.TransCartesianZ), axis=1)
+    # Converts the results from llh2xyz into a dataframe
+    llhresultdf = pd.DataFrame(list(returned), columns=[translat, translong, transellht])
+
+    # Combines the results with the original dataframe
+    csvdf = pd.concat([csvdf, llhresultdf], axis=1)
+    csvdf = csvdf.drop(['CartesianX', 'CartesianY', 'CartesianZ', 'TransCartesianX', 'TransCartesianY',
+                        'TransCartesianZ'], axis=1)
+
+    # Writes the csv dataframe to a csv file
+    csvdf.to_csv(fn_out, index=False)