WIP:hr_to_mr and mr_to_hr functions (closes #54)

pydatarecognition/utils.py

@@ -213,4 +213,81 @@ def get_formatted_crossref_reference(doi):
 
     return ref, ref_date
 
+
+def hr_to_mr_number_and_esd(number_esd):
+    '''
+    splits human readable numbers with estimated standard deviations (e.g. 343.44(45)) into machine readable numbers and
+    estimated standard deviations (e.g. 343.44 and 0.45).
+
+    Parameters
+    ----------
+    number_esd : array_like or string
+      The array-like object that contains numbers with their estimated standard deviations as strings
+      in the following format: ["343.44(45)", "324908.435(67)", "0.0783(1)"] or
+      The string that contains numbers with their estimated standard deviations separated by new line characters
+      in the following format: "343.44(45)\n324908.435(67)\n0.0783(1)"
+
+    Returns
+    -------
+    number : numpy array
+      The array with the numbers as floats
+
+    esd : numpy array
+      The array with estimated standard deviations as floats
+
+    '''
+    number_esd = np.array(number_esd, dtype='str')
+    number = np.char.split(number_esd, sep="(")
+    esd = np.array([e[1].split(")")[0] for e in number], dtype='float')
+    number = np.array([e[0] for e in number], dtype='str')
+    esd_oom = []
+    for i in range(len(number)):
+        if len(number[i].split(".")) == 1:
+            esd_oom.append(1)
+        else:
+            esd_oom.append(10**-len(number[i].split(".")[1]))
+    esd_oom = np.array(esd_oom, dtype='float')
+    number, esd = np.array(number, dtype='float'), np.array(esd * esd_oom, dtype='float')
+
+    return number, esd
+
+
+def mr_to_hr_number_and_esd(number, esd):
+    '''
+    merges machine readable numbers and estimated standard deviations (e.g. 343.44 and 0.45) into human readable
+    numbers with estimated standard deviations (e.g. 343.44(45)).
+
+    Parameters
+    ----------
+    number : array_like or string
+      The array-like object that contains numbers in the following format: [343.44, 324908.435, 0.0783] or
+      The string that contains numbers in the following format: "343.44\n324908.435\n0.0783"
+
+    esd : array_like or string
+      The array-like object that contains estimated standard deviations in the following format:
+      [0.45, 0.067, 0.0001]
+      The string that contains estimated standard deviations in the following format:
+      "0.45\n0.067\n0.0001"
+
+    Returns
+    -------
+    number_esd : list
+      The list of strings that contains the rounded numbers with estimated standard deviations
+      in the following format: ["343.4(5)", "324908.44(7)", "0.0783(1)" ]
+
+    '''
+    number, esd = np.array(number, dtype='float').astype('str'), np.array(esd, dtype='float').astype('str')
+    number_hr, esd_hr = [], []
+    for i in range(len(number)):
+        if number[i].split(".")[1] == "0":
+            number_hr.append(number[i].split(".")[0])
+            esd_hr.append(esd[i].split(".")[0])
+        else:
+            number_hr.append(number[i])
+            esd_hr.append(int(esd[i].split(".")[1]))
+    number, esd = np.array(number_hr, dtype='str'), np.array(esd_hr, dtype='str')
+    number_esd = np.array([f'{number[i]}({esd[i]})' for i in range(len(esd))])
+
+    return number_esd
+
 # End of file.

tests/test_utils.py

@@ -2,7 +2,8 @@
 import pytest
 from datetime import date
 from habanero import Crossref
-from pydatarecognition.utils import data_sample, pearson_correlate, xy_resample, get_formatted_crossref_reference
+from pydatarecognition.utils import data_sample, pearson_correlate, xy_resample, get_formatted_crossref_reference, \
+     hr_to_mr, mr_to_hr
 
 def test_data_sample():
     test_cif_data = [[10.0413, 10.0913, 10.1413, 10.1913],
@@ -69,4 +70,19 @@ def mockreturn(*args, **kwargs):
     actual = get_formatted_crossref_reference("test")
     assert actual == expected
 
+
+def test_hr_to_mr():
+    number_esd = ["343.44(45)", "324908.435(67)", "0.0783(1)", "11(1)", "51(13)"]
+    actual = hr_to_mr(number_esd)
+    expected = np.array([343.44, 324908.435, 0.0783, 11, 51]), np.array([0.45, 0.067, 0.0001, 1, 13])
+    assert np.allclose(actual[0], expected[0])
+    assert np.allclose(actual[1], expected[1])
+
+
+def test_mr_to_hr():
+    number, esd = [343.44, 324908.435, 0.0783, 11, 51], [0.45, 0.067, 0.0001, 1, 13]
+    actual = mr_to_hr(number, esd)
+    expected = np.array(["343.44(45)", "324908.435(67)", "0.0783(1)", "11(1)", "51(13)"], dtype='str')
+    assert np.array_equal(actual, expected)
+
 # End of file.