Small performance optimization in spectrum.extreme_values() function,…

… added example script for this function

example_scripts/extreme_values.py

@@ -0,0 +1,51 @@
+#!/usr/bin/env python
+
+import pymzml
+from collections import defaultdict as ddict
+import os
+
+
+def main():
+    """
+    Testscript to fetch the extreme m/z values from each spectrum
+    of an example file.
+
+    Usage:
+
+        ./extreme_values.py
+
+    Parses the file '../tests/data/example.mzML' and extracts the smallest and
+    largest m/z from each each spectrum.
+
+    """
+    example_file = os.path.join(
+        os.path.dirname(__file__),
+        os.pardir,
+        'tests',
+        'data',
+        'example.mzML'
+    )
+    run = pymzml.run.Reader(
+        example_file
+    )
+    extreme_mz_values = {}
+
+    number_of_mz_values = 2
+
+    for spectrum in run:
+        # print( spectrum.ID )
+        if spectrum.ms_level == 1:
+            extreme_mz_values[ spectrum.ID ] = spectrum.extreme_values('mz')
+
+    for spectrum_id, extreme_mz_tuple in extreme_mz_values.items():
+        assert len(extreme_mz_tuple) == number_of_mz_values
+        print(
+            'Spectrum {0}; lowest m/z: {1} highest m/z: {2}'.format(
+                spectrum_id,
+                *extreme_mz_tuple
+            )
+        )
+
+
+if __name__ == '__main__':
+    main()

pymzml/spec.py

@@ -1311,7 +1311,10 @@ def estimated_noise_level(self, mode='median'):
                 ) / float(len(self.peaks('centroided')))
             else:
                 print(
-                    'dont understand noise level estimation method call', mode
+                    'Do not understand noise level estimation method call with \
+                    given mode: {0}'.format(
+                        mode
+                    )
                 )
             return_value = self.noise_level_estimate[mode]
         return return_value
@@ -1381,20 +1384,29 @@ def extreme_values(self, key):
             extrema (tuple) : tuple of minimal and maximum m/z or intensity
 
         """
-        if key not in ['mz', 'i']:
-            print("Dont understand extreme request ")
+        available_extreme_values = ['mz', 'i']
+        if key not in available_extreme_values:
+            print(
+                "Do not understand extreme request: '{0}'; available values are: {1}".format(
+                    key,
+                    available_extreme_values
+                )
+            )
+            exit()
         if self._extreme_values is None:
             self._extreme_values = {}
         try:
             if key == 'mz':
+                all_mz_values = [mz for mz, i in self.peaks('raw')]
                 self._extreme_values['mz'] = (
-                    min([mz for mz, i in self.peaks('raw')]),
-                    max([mz for mz, i in self.peaks('raw')])
+                    min(all_mz_values),
+                    max(all_mz_values)
                 )
             else:
-                self._extreme_values['i']  = (
-                    min([i for mz, i in self.peaks('raw')]),
-                    max([i for mz, i in self.peaks('raw')])
+                all_i_values = [i for mz, i in self.peaks('raw')]
+                self._extreme_values['i'] = (
+                    min(all_i_values),
+                    max(all_i_values)
                 )
         except ValueError:
             # emtpy spectrum
@@ -1483,7 +1495,7 @@ def similarity_to(self, spec2, round_precision=0):
             to 1.
         """
         assert isinstance(spec2, Spectrum), \
-            "Spectrum2 is not a pymzML spectrum"
+            "Spectrum 2 is not a pymzML spectrum"
 
         vector1 = ddict(int)
         vector2 = ddict(int)