Merge branch 'master' into bugfix/attribut_data_not_needed_anymore

Author: the-it

binary-search/README.md

@@ -1,3 +1,73 @@
-# How to Do a Binary Search in Python?
+# How to Do a Binary Search in Python
 
-Code snippets supplementing the [How to Do a Binary Search in Python?](https://realpython.com/binary-search-python/) article on [Real Python](https://realpython.com/).
+Code snippets supplementing the [How to Do a Binary Search in Python](https://realpython.com/binary-search-python/) article.
+
+## Usage
+
+### Download Sample Dataset
+
+To download a sample dataset for benchmarking, run the following script:
+
+```shell
+$ python download_imdb.py
+Fetching data from IMDb...
+Created "names.txt" and "sorted_names.txt"
+```
+
+### Benchmark
+
+**Note:** The hash-based search algorithm requires a separate data structure, so it's not included here. Feel free to treat it as an exercise.
+
+Find the index of Arnold Schwarzenegger in unsorted names using random search:
+
+```shell
+$ python benchmark.py -a random -f names.txt 'Arnold Schwarzenegger'
+Loading names...ok
+[1/10] Found at index=215 (21.68 s)
+[2/10] Found at index=215 (148.10 ms)
+[3/10] Found at index=215 (63.49 s)
+[4/10] Found at index=215 (15.93 s)
+[5/10] Found at index=215 (23.38 s)
+[6/10] Found at index=215 (60.98 s)
+[7/10] Found at index=215 (10.14 s)
+[8/10] Found at index=215 (15.81 s)
+[9/10] Found at index=215 (5.10 s)
+[10/10] Found at index=215 (13.36 s)
+best=148.10 ms, worst=63.49 s, avg=23.00 s, median=15.87 s
+```
+
+Find the index of Arnold Schwarzenegger in unsorted names using linear search:
+
+```shell
+$ python benchmark.py -a linear -f names.txt 'Arnold Schwarzenegger'
+Loading names...ok
+[1/10] Found at index=215 (30.95 µs)
+[2/10] Found at index=215 (26.86 µs)
+[3/10] Found at index=215 (26.26 µs)
+[4/10] Found at index=215 (26.55 µs)
+[5/10] Found at index=215 (26.24 µs)
+[6/10] Found at index=215 (25.88 µs)
+[7/10] Found at index=215 (25.77 µs)
+[8/10] Found at index=215 (25.89 µs)
+[9/10] Found at index=215 (25.91 µs)
+[10/10] Found at index=215 (25.99 µs)
+best=25.77 µs, worst=30.95 µs, avg=26.63 µs, median=26.11 µs
+```
+
+Find the index of Arnold Schwarzenegger in *sorted* names using binary search:
+
+```shell
+$ python benchmark.py -a binary -f sorted_names.txt 'Arnold Schwarzenegger'
+Loading names...ok
+[1/10] Found at index=782431 (18.82 µs)
+[2/10] Found at index=782431 (9.19 µs)
+[3/10] Found at index=782431 (11.08 µs)
+[4/10] Found at index=782431 (9.70 µs)
+[5/10] Found at index=782431 (10.14 µs)
+[6/10] Found at index=782431 (9.35 µs)
+[7/10] Found at index=782431 (9.42 µs)
+[8/10] Found at index=782431 (8.79 µs)
+[9/10] Found at index=782431 (8.66 µs)
+[10/10] Found at index=782431 (8.79 µs)
+best=8.66 µs, worst=18.82 µs, avg=10.39 µs, median=9.38 µs
+```

python-bindings/overview_article/cffi_test.py

@@ -0,0 +1,9 @@
+#!/usr/bin/env python3
+import cffi_example
+
+if __name__ == "__main__":
+    # Sample data for our call:
+    x, y = 6, 2.3
+
+    answer = cffi_example.lib.cmult(x, y)
+    print(f"    In Python: int: {x} float {y:.1f} return val {answer:.1f}")

python-bindings/overview_article/cmult.c

@@ -0,0 +1,11 @@
+#include <stdio.h>
+#include "cmult.h"
+
+
+float cmult(int int_param, float float_param) {
+    float return_value = int_param * float_param;
+    printf("    In cmult : int: %d float %.1f returning  %.1f\n", int_param,
+            float_param, return_value);
+    return return_value;
+}
+

python-bindings/overview_article/cmult.h

@@ -0,0 +1 @@
+float cmult(int int_param, float float_param);

python-bindings/overview_article/cppmult.cpp

@@ -0,0 +1,15 @@
+#include <iostream>
+#include <iomanip>
+#include "cppmult.hpp"
+
+
+float cppmult(int int_param, float float_param) {
+    float return_value = int_param * float_param;
+    std::cout << std::setprecision(1) << std::fixed
+	          << "    In cppmul: int: " << int_param
+	          << " float " <<  float_param
+	          << " returning  " << return_value
+	          << std::endl;
+    return return_value;
+}
+

python-bindings/overview_article/cppmult.hpp

@@ -0,0 +1 @@
+float cppmult(int int_param, float float_param);

python-bindings/overview_article/ctypes_test.py

@@ -0,0 +1,26 @@
+#!/usr/bin/env python
+""" Simple examples of calling C functions through ctypes module. """
+import ctypes
+import pathlib
+
+
+if __name__ == "__main__":
+    # Load the shared library into c types.
+    libname = pathlib.Path().absolute() / "libcmult.so"
+    c_lib = ctypes.CDLL(libname)
+
+    # Sample data for our call:
+    x, y = 6, 2.3
+
+    # This will not work:
+    # answer = c_lib.cmult(x, y)
+
+    # This produces a bad answer:
+    answer = c_lib.cmult(x, ctypes.c_float(y))
+    print(f"    In Python: int: {x} float {y:.1f} return val {answer:.1f}")
+    print()
+
+    # You need tell ctypes that the function returns a float
+    c_lib.cmult.restype = ctypes.c_float
+    answer = c_lib.cmult(x, ctypes.c_float(y))
+    print(f"    In Python: int: {x} float {y:.1f} return val {answer:.1f}")

python-bindings/overview_article/cython_example.pyx

@@ -0,0 +1,7 @@
+""" Example cython interface definition """
+
+cdef extern from "cppmult.hpp":
+    float cppmult(int int_param, float float_param)
+
+def pymult( int_param, float_param ):
+    return cppmult( int_param, float_param )

python-bindings/overview_article/cython_test.py

@@ -0,0 +1,8 @@
+#!/usr/bin/env python
+import cython_example
+
+# Sample data for our call
+x, y = 6, 2.3
+
+answer = cython_example.pymult(x, y)
+print(f"    In Python: int: {x} float {y:.1f} return val {answer:.1f}")

python-bindings/overview_article/pybind11_test.py

@@ -0,0 +1,9 @@
+#!/usr/bin/env python
+import pybind11_example
+
+if __name__ == "__main__":
+    # Sample data for our call:
+    x, y = 6, 2.3
+
+    answer = pybind11_example.cpp_function(x, y)
+    print(f"    In Python: int: {x} float {y:.1f} return val {answer:.1f}")