.gitignore

@@ -128,8 +128,10 @@ dmypy.json
 # Pyre type checker
 .pyre/
 
+
 # project-specific files
 README_init.md
-A_setup_python/
-B_explore_python/
-Cygwin_setup.md
+
+# solution files and .py files in project directory
+**/*_sol.py
+/*.py

.vscode/settings.json

+{
+    "code-runner.clearPreviousOutput": true,
+    "python.testing.unittestArgs": [
+        "-v",
+        "-s",
+        ".",
+        "-p",
+        "test_*.py"
+    ],
+    "python.testing.pytestEnabled": false,
+    "python.testing.unittestEnabled": true
+}
\ No newline at end of file

A_setup_python/README.md

+# Assignment A: Setup Python   (10 Pts)
+
+This assignment will setup your base Python enviroment. If you already have it, simply run challenges and answer questions (if any). If you cannot run challenges, set up the needed software.
+
+
+### Challenges
+- [Challenge 1:](#1-challenge-terminal) Terminal
+- [Challenge 2:](#2-challenge-python3) Python3
+- [Challenge 3:](#3-challenge-pip) pip
+- [Challenge 4:](#4-challenge-test-python) Test Python
+
+Points: [4, 2, 2, 2]
+
+
+ 
+### 1.) Challenge: Terminal
+
+If you are using *MacOS* or *Linux*, skip steps for *Windows*.
+
+For *Windows*:
+
+- Use a Unix emulator such as
+[cygwin](https://www.cygwin.com),
+the built-in Windows Subsystem for Linux
+[WSL](https://learn.microsoft.com/en-us/windows/wsl/install)
+or a
+[Linux VM](https://ubuntu.com/tutorials/how-to-run-ubuntu-desktop-on-a-virtual-machine-using-virtualbox).
+
+    - Windows *CMD.EXE* is no option and *Powershell* is not recommended since it is not compatible with Unix standards.
+
+- Follow instructions for
+[setting up *cygwin*](../Cygwin_setup.md),
+in particular for:
+  - switching from `/cygdrive/c` to `/c`
+  - selecting your `HOME`-directory
+  - configuring `.bashrc` in your `HOME`-directory and
+  - defining `PATH` in `.bashrc`.
+
+Open a terminal and type commands:
+```sh
+> cd                    # change to HOME-directory
+> pwd                   # print path to working directory
+> ls -la                # show content of HOME-directory
+> whoami                # show user identity
+> cat .profile              # file may not exist
+> cat .bashrc               # for Mac, use .zshrc
+> echo $PATH            # show PATH variable
+```
+Explain commands. If you are not familiar, find out about these basic Unix shell (bash, zsh, ...) commands (e.g. from [introduction](https://cs.lmu.edu/~ray/notes/bash) or 
+[tutorial](https://linuxconfig.org/bash-scripting-tutorial-for-beginners)).
+
+On *Mac*, use file *.zshrc* instead of *.bashrc*. If file does not exist
+in the `HOME`-directory, it can be created as a text file using an editor.
+
+Find out about
+[ differences between](https://www.baeldung.com/linux/bashrc-vs-bash-profile-vs-profile)
+`.profile` and `.bashrc` (Mac: `.zshrc`):
+  - When is `.profile` executed?
+  - When is `.bashrc` (Mac: `.zshrc`) executed?
+
+(4 Pts)
+
+
+ 
+### 2.) Challenge: Python3
+Check if you have Python 3 installed on your system. Name three differences between [Python 2 and 3](https://www.guru99.com/python-2-vs-python-3.html#7). 
+
+Run commands in terminal (version 3.x.y may vary):
+```sh
+> python --version
+Python 3.12.0
+```
+
+If you see error *command not found*, add path to Python installation to
+PATH variable in `.bashrc` (Mac: `.zshrc`).
+
+(2 Pts)
+
+
+ 
+### 3.) Challenge: pip
+Check if you have a Python package manager installed (pip, conda, ... ). [`pip`](https://pip.pypa.io) is Python's default package manager to install additional Python packages and libraries.
+
+Follow [instructions](https://pip.pypa.io/en/stable/installing) for installation:
+- [download](https://bootstrap.pypa.io/get-pip.py) the `get-pip.py` file.
+- run `python get-pip.py`
+- or update pip to latest version: `python -m pip install --upgrade pip`
+
+Run commands in terminal:
+```
+> pip --version
+pip 23.2.1 from C:\Users\svgr2\AppData\Local\Programs\Python\Python312\Lib\site-
+packages\pip (python 3.12)
+```
+(2 Pts)
+
+
+ 
+### 4.) Challenge: Test Python
+Start Python in the terminal and execute commands:
+```py
+> python
+Python 3.12.0 (tags/v3.12.0:0fb18b0, Oct  2 2023, 13:03:39) [MSC v.1935 64 bit (AMD64)] on win32
+Type "help", "copyright", "credits" or "license" for more information.
+
+>>> print('Hello World')
+Hello World
+
+>>> help('modules')
+...lists installed python packages
+
+>>> 2+3*4
+14
+
+>>> x = 2+3*4
+>>> x
+14
+```
+
+Create a file *print_sys.py* with following content.
+```py
+import platform
+
+impl = platform.python_implementation()
+ver = platform.version()
+mach = platform.machine()
+sys = platform.system()
+
+print('Python impl:    ' + impl)
+print('Python version: ' + ver)
+print('Python machine: ' + mach)
+print('Python system:  ' + sys)
+print('Python version: ' + platform.python_version())
+```
+
+Run file *print_sys.py*. Output varies depending on your system.
+```
+> python print_sys.py
+Python impl:    CPython
+Python version: 10.0.19045
+Python machine: AMD64
+Python system:  Windows
+Python version: 3.12.0
+```
+(2 Pts)

A_setup_python/print_sys.py

+import platform
+
+impl = platform.python_implementation()
+ver = platform.version()
+mach = platform.machine()
+sys = platform.system()
+
+print('Python impl:    ' + impl)
+print('Python version: ' + ver)
+print('Python machine: ' + mach)
+print('Python system:  ' + sys)
+print('Python version: ' + platform.python_version())

B_explore_python/README.md

+# Assignment B: Explore Python   (20 Pts)
+
+This assignment demonstrates Python's basic data structures.
+
+### Challenges
+- [Challenge 1:](#1-challenge-indexing-fruits) Indexing Fruits
+- [Challenge 2:](#2-challenge-packaging-fruits) Packaging Fruits
+- [Challenge 3:](#3-challenge-sorting-fruits) Sorting Fruits
+- [Challenge 4:](#4-challenge-income-analysis) Income Analysis
+- [Challenge 5:](#5-challenge-code-income-analysis) Code Income Analysis
+- [Challenge 6:](#6-challenge-explore-python-built-in-functions)
+    Explore Python built-in functions
+
+Points: [1, 7, 2, 3, 6, 1]
+
+
+ 
+### 1.) Challenge: Indexing Fruits
+Review Python's basic
+[data structures](https://www.dataquest.io/blog/data-structures-in-python).
+
+```py
+# Python is known for advanced list processing.
+>>> fruits = ['apple', 'pear', 'orange', 'banana', 'apple']
+>>> print(fruits)
+>>> fruits
+['apple', 'pear', 'orange', 'banana', 'apple']
+
+>>> print(len(fruits))
+5
+
+>>> print(f"the third fruit is: {fruits[2]}")
+the third fruit is: orange
+
+>>> print(f"the second and third fruits are: {fruits[1:3]}")
+the second and third fruits are: ['pear', 'orange']
+
+>>> print(f"the last fruit is: {fruits[-1]}")
+the last fruit is: apple
+
+>>> print(f"the second-last fruit is: {fruits[-2]}")
+the second-last fruit is: banana
+
+>>> print(f"the last three fruits are: {fruits[-3:]}")
+the last three fruits are: ['orange', 'banana', 'apple']
+```
+Perform examples on your laptop.
+
+(1 Pt)
+
+
+ 
+### 2.) Challenge: Packaging Fruits
+
+Review Python's built-in
+[data structures](https://www.dataquest.io/blog/data-structures-in-python).
+Perform examples and answer questions on a piece of paper.
+
+1. What are data types for `fruits`, `fruitbag` and `fruitbox` called? (1 Pt)
+
+1. Name three properties that characterize each data type. (1 Pts)
+
+1. Why does output for `fruitbag` differ from input? (1 Pt)
+
+    ```py
+    >>> fruits = ['apple', 'pear', 'orange', 'banana', 'apple']
+    >>> fruitbag = {'apple', 'pear', 'orange', 'banana', 'apple'}
+    >>> fruitbox = ('apple', 'pear', 'orange', 'banana', 'apple')
+
+    >>> print(fruits)
+    ['apple', 'pear', 'orange', 'banana', 'apple']
+
+    >>> print(fruitbox)
+    ('apple', 'pear', 'orange', 'banana', 'apple')
+
+    >>> print(fruitbag)
+    {'banana', 'apple', 'pear', 'orange'}
+
+    >>> print(fruits[1])
+    pear
+    >>> print(fruitbox[1])
+    pear
+    >>> print(fruitbag[1])
+    TypeError: object is not subscriptable
+    >>>
+    ```
+
+1. How is the structure for `eric1` called? What is the difference to
+    `eric2`? Explain outputs. (1 Pt)
+
+    ```py
+    eric1 = {"name": "Eric", "salary": 5000, "birthday": "Sep 25 2001"}
+
+    eric2 = {"name", "Eric", "salary", 5000, "birthday", "Sep 25 2001"}
+
+    >>> print(eric1)
+    {'name': 'Eric', 'salary': 5000, 'birthday': 'Sep 25 2001'}
+
+    >>> print(eric2)
+    {'Sep 25 2001', 5000, 'name', 'Eric', 'birthday', 'salary'}
+    # 
+    # print(eric2) in same order?
+
+    # print salary for eric1 and eric2?
+    >>> print(eric1["salary"])
+    5000
+    ```
+
+1. Some people say that Arrays in other languages are Lists
+in Python. Other people argue that Tuples are closer to Arrays.
+    - a) Which statement is (more) correct?
+    - b) Name two differences between Arrays and Lists?
+        (1 Pt)
+    - c) What are differences between a Python 2-dimensional List
+        (List with Lists) and a *NumPy* Array *ndarray* (read the first
+        [three paragraphs](https://numpy.org/doc/stable/user/whatisnumpy.html))?
+        (1 Pt)
+
+1. Draw sketches to visualize Python data structures:
+*List*, *Set*, *Tuple*, *Dictionary* and *NumPy Array*. (1 Pt)
+
+
+ 
+### 3.) Challenge: Sorting Fruits
+
+```py
+>>> fruits = ['apple', 'pear', 'orange', 'banana', 'apple']
+
+>>> f1 = sorted(fruits)
+>>> print(f"{f1},\n{fruits}")
+['apple', 'apple', 'banana', 'orange', 'pear'],
+['apple', 'pear', 'orange', 'banana', 'apple']
+
+
+>>> f2 = fruits.sort()
+>>> print(f"{f2},\n{fruits}")
+None,
+['apple', 'apple', 'banana', 'orange', 'pear']
+```
+
+What is the difference between List-function *sort()* and built-in
+function *sorted()*, see
+[link](https://www.python-engineer.com/posts/sort-vs-sorted)?
+
+(2 Pts)
+
+
+ 
+### 4.) Challenge: Income Analysis
+The US tax Income Revenue Service (IRS) annually 
+publishes income statistics by ZIP codes (postal codes)
+([reports](https://www.irs.gov/statistics/soi-tax-stats-individual-income-tax-statistics-2020-zip-code-data-soi)).
+
+For example, California ZIP Code
+[93636](https://simplemaps.com/us-zips/93636)
+is for *Madera* county, an agricultural region north of
+Fresno in the Central Valley.
+The income distribution in this county was for the tax year 2020:
+```
+income bracket:         number of tax returns
+                        filed in bracket
+[$1 to under $25,000]            1,800
+[$25,000 to under $50,000]       1,380
+[$50,000 to under $75,000]         980
+[$75,000 to under $100,000]        830
+[$100,000 to under $200,000]     1,660
+[$200,000 or more, up to $10M]     550
+```
+In the bracket [$50,000 to under $75,000] of taxable income (exact to 74,999),
+a total of 980 tax returns were filed, which means 980 tax payers reported
+taxable income in this band.
+
+We assume $10 million ($10M) as upper limit in the highest bracket.
+
+A common statistical analysis is to compute:
+
+- the *mean (average) income* and the
+
+- the *median income* across all tax payers in *Madera* county.
+
+For calculating the *mean income*, average the income within
+each bracket.
+
+For calculating the *median income*, consider a linear rising
+income from the lower bound to the upper bound in the bracket
+holding the *median income*.
+
+Answer questions:
+
+1. What is the difference between *mean (average)* and
+*median* calculations?
+    - Why are both indicators relevant?
+    - When should one over the other be preferred?
+
+1. Calculate manually the *mean* or *average* income for *Madera* county
+(result: *$453,073* ).
+
+1. Calculate manually the *median* income for *Madera* county
+(result: *$60,714* ).
+
+(3 Pts)
+
+
+ 
+### 5.) Challenge: Code Income Analysis
+
+Write Python code to perform this income analysis for arbitray
+ZIP regions.
+
+Use file
+<b>[income_tax_analysis.py](https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/blob/main/B_explore_python/income_tax_analysis.py)</b>
+as template.
+
+<b>Use pure Python</b>, no libraries such as *Pandas* or *Numpy* or
+built-in library functions.
+
+Think about following steps:
+
+1. Chose a suitable Python data structure to represent tax data for
+    a ZIP code.
+    - Which data is relevant for the analysis?
+    - How can data be structured?
+    - Use only use Python structures: *list*, *set*, *tuple*, *dictionary*.
+    (1 Pt)
+
+1. Implement that data structure for the *Madera* data
+    (fill data in code, no need to read `.xlsx`-files). (1 Pt)
+
+1. Define two functions `mean_income(...)` and `median_income(...)`
+    that take a data structure for a ZIP code as input and return
+    respective numbers as results.
+
+1. Define a function `number_of_returns(...)` that returns what the
+    name suggests.
+
+1. Implement these functions and demonstrate they return correct values.
+
+1. Demonstrate analysis for other ZIP codes using IRS tax
+[reports](https://www.irs.gov/statistics/soi-tax-stats-individual-income-tax-statistics-2020-zip-code-data-soi).
+Select a state (CA: California, IA: Iowa, NY: New York City), download
+the `.xlsx` and navigate to ZIP codes for:
+    - [94040](https://simplemaps.com/us-zips/94040) (Mountain View, CA),
+    - [94304](https://simplemaps.com/us-zips/94304) (Palo Alto, CA),
+    - [94027](https://simplemaps.com/us-zips/94027) (Atherton, CA),
+    - [50860](https://simplemaps.com/us-zips/50860) (Redding, IA) and
+    - [10023](https://simplemaps.com/us-zips/10023) (New York City, NY Upper West side). (1 Pt)
+
+Develop (test, debug) code in
+[income_tax_analysis.py](https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/blob/main/B_explore_python/income_tax_analysis.py)
+in your IDE (*VS Code*, *PyCharm*, *Jupyter*, etc.).
+
+Run the final result in a terminal (*Jupyter*: export Python file):
+
+```sh
+$ python income_tax_analysis.py
+```
+
+Results:
+```
+mean_income in Madera County, CA          is:    453,073 - median_income is:   60,714
+mean_income in Mountain View, CA          is:  1,740,371 - median_income is:  114,820
+mean_income in Palo Alto, CA              is:  2,077,038 - median_income is:  153,658
+mean_income in Atherton, CA               is:  2,623,882 - median_income is:  354,088
+mean_income in Redding, IA                is:     33,333 - median_income is:   31,250
+mean_income in New York City, NY U West   is:  1,544,991 - median_income is:  104,775
+```
+(4 Pts)
+
+
+&nbsp;
+### 6.) Challenge: Explore Python built-in functions
+Learn about Python's
+[built-in functions](https://docs.python.org/3/library/functions.html).
+Test the
+[*globals()*](https://docs.python.org/3/library/functions.html#globals)
+function.
+```py
+  >>> globals()
+  {'__name__': '__main__', '__doc__': None, '__package__': None, '__loader__': <class '_frozen_
+  importlib.BuiltinImporter'>, '__spec__': None, '__annotations__': {}, '__builtins__': <module
+  'builtins' (built-in)>, 'fruits': ['apple', 'pear', 'orange', 'banana']}
+```
+Test the [*input()*](https://docs.python.org/3/library/functions.html#input) function.
+```py
+  >>> s = input('--> ')
+  --> Monty Python's Flying Circus
+  >>> s
+  "Monty Python's Flying Circus"
+  exit()
+```
+(1 Pt)

B_explore_python/fruits.py

+'''
+Code for Assignment B: Explore Python
+ - Challenge 1: Indexing Fruits
+ - Challenge 2: Packaging Fruits
+ - Challenge 3: Sorting Fruits
+'''
+
+fruits = ['apple', 'pear', 'orange', 'banana', 'apple']
+fruitbag = {'apple', 'pear', 'orange', 'banana', 'apple'}
+fruitbox = ('apple', 'pear', 'orange', 'banana', 'apple')
+
+
+def index_fruits():
+    # fruits = ['apple', 'pear', 'orange', 'banana']
+    print(fruits)
+    print(len(fruits))
+    print(f"the third fruit is: {fruits[2]}")
+    print(f"the second and third fruits are: {fruits[1:3]}")
+    print(f"the last fruit is: {fruits[-1]}")
+    print(f"the last two fruits are: {fruits[-2:]}")
+
+
+def package_fruits():
+    print(f'fruits:   {fruits}')
+    print(f'fruitbox: {fruitbox}')
+    print(f'fruitbag: {fruitbag}')
+    print(fruits[1])
+    print(fruitbox[1])
+    # print(fruitbag[1])
+    # 
+    eric = {"name": "Eric", "salary": 5000, "birthday": "Sep 25 2001"}
+    print(eric)
+    print(eric["salary"])
+
+
+def sort_fruits():
+    f1 = sorted(fruits)
+    print(f"{f1},\n{fruits}")
+    f2 = fruits.sort()
+    print(f"{f2},\n{fruits}")
+
+
+if __name__ == '__main__':
+    '''
+    Main driver that runs when this file is executed by the Python interpreter.
+    '''
+    index_fruits()
+    package_fruits()
+    sort_fruits()
+    # 
+    # print(globals())

B_explore_python/income_tax_analysis.py

+'''
+Code for Assignment B: Explore Python
+ - Challenge 4: Income Analysis
+ - Challenge 5: Code Income Analysis
+
+Based on 2020 tax returns, income distribution in Madera County, CA
+with (postal) ZIP code 93636 was:
+#
+income brackets:        number of tax returns
+                        filed in brackets:
+[$1 to under $25,000]             1,800
+[$25,000 to under $50,000]        1,380
+[$50,000 to under $75,000]          980
+[$75,000 to under $100,000]         830
+[$100,000 to under $200,000]      1,660
+[$200,000 or more, up to $10M>]     550
+'''
+
+# design a data structure that stores information about a ZIP area
+# that is relevant for mean/median tax analysis
+zip_93636 = None
+zip_94040 = None
+zip_94304 = None
+zip_94027 = None
+zip_50860 = None
+zip_10023 = None
+
+
+# implement a function that calculates the mean income for a ZIP area
+def mean_income(_zip) -> int:
+    return 25000    # mock result, replace with computed result
+
+
+# implement a function that calculates the median income for a ZIP area
+def median_income(_zip) -> int:
+    return 18500    # mock result, replace with computed result
+
+
+# use this function to print results for a ZIP area
+def print_analysis(_zip):
+    _county = 'Madera, CA'
+    print(
+        f'mean_income in {_county:26} is: {mean_income(_zip):10,} - ' +
+        f'median_income is: {median_income(_zip):8,}'
+    )
+
+
+# attempt to load solution module (if present - ignore)
+try:
+    solution_module = 'income_tax_analysis_sol'
+    mod = __import__(solution_module, globals(), locals(), [], 0)
+    mean_income, median_income, print_analysis = mod.mean_income, mod.median_income, mod.print_analysis
+    zip_93636, zip_94040, zip_94304, zip_94027, zip_50860, zip_10023 = \
+        mod.zip_93636, mod.zip_94040, mod.zip_94304, mod.zip_94027, mod.zip_50860, mod.zip_10023
+#
+except ImportError:
+    pass
+
+
+if __name__ == '__main__':
+    '''
+    driver code that runs when file is directly executed
+    '''
+    print_analysis(zip_93636)
+    print_analysis(zip_94040)
+    print_analysis(zip_94304)
+    print_analysis(zip_94027)
+    print_analysis(zip_50860)
+    print_analysis(zip_10023)

C_expressions/.vscode/settings.json

+{
+    "python.testing.unittestArgs": [
+        "-v",
+        "-s",
+        ".",
+        "-p",
+        "test_*.py"
+    ],
+    "python.testing.pytestEnabled": false,
+    "python.testing.unittestEnabled": true
+}
\ No newline at end of file

C_expressions/README.md

+# Assignment C: Python Expressions & Unit Tests   (16 Pts)
+
+This assignment demonstrates Python's powerful (*"one-liner"*) expressions.
+
+### Challenges
+- [Challenge 1:](#1-challenge-create-new-project) Create New Project
+- [Challenge 2:](#2-challenge-run-code) Run Code
+- [Challenge 3:](#3-challenge-run-unit-tests) Run Unit Tests
+- [Challenge 4:](#4-challenge-write-expressions) Write Expressions
+- [Challenge 5:](#5-challenge-final-test-and-sign-off) Final Test and sign-off
+
+Points: [1, 2, 3, 0, 10]
+
+
+ 
+### 1.) Challenge: Create New Project
+Create a Python project, e.g. named `C_expressions`, and
+[pull files](https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/tree/main/C_expressions)
+from GitLab (above).
+
+Inspect files and figure out their purpose. Write 1-2 sentenses what each file means
+and purpose is:
+
+ - [__init __.py](https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/blob/main/C_expressions/__init__.py)
+    : `_____________________________________`
+ 
+     - What does the init-file contain?
+     - When and how often is this file executed?
+
+ - [expressions.py](https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/blob/main/C_expressions/expressions.py)
+    : `__________________________________`
+
+ - [test_expressions.py](https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/blob/main/C_expressions/test_expressions.py)
+    : `______________________________`
+
+(1 Pt)
+
+
+ 
+### 2.) Challenge: Run Code
+Run file `expressions.py` in your IDE:
+```
+numbers: [4, 12, 3, 8, 17, 12, 1, 8, 7]
+#
+a) number of numbers: 9
+b) first three numbers: []
+c) last three numbers: []
+d) last three numbers reverse: []
+e) odd numbers: []
+f) number of odd numbers: 0
+g) sum of odd numbers: 0
+h) duplicate numbers removed: []
+i) number of duplicate numbers: 0
+j) ascending, de-dup (n^2) numbers: []
+k) length: NEITHER
+
+[Done] exited with code=0 in 0.126 seconds
+```
+(1 Pt)
+
+Run file `expressions.py` in terminal:
+```sh
+cd <project>                # cd into project directory
+pwd                         # print working directory
+/c/.../workspaces/ds_cs4bd_2324/C_expressions
+
+python expressions.py       # run program
+-->
+numbers: [4, 12, 3, 8, 17, 12, 1, 8, 7]
+#
+a) number of numbers: 9
+b) first three numbers: [4, 12, 3]
+c) last three numbers: [1, 8, 7]
+d) last three numbers reverse: [7, 8, 1]
+e) odd numbers: [3, 17, 1, 7]
+f) number of odd numbers: 4
+g) sum of odd numbers: 28
+h) duplicate numbers removed: [1, 3, 4, 7, 8, 12, 17]
+i) number of duplicate numbers: 2
+j) ascending, de-dup (n^2) numbers: [1, 9, 16, 49, 64, 144, 289]
+k) length: ODD_LIST
+```
+(1 Pt)
+
+
+&nbsp;
+### 3.) Challenge: Run Unit Tests
+Unit Tests are used to *"test-a-unit"* of code in isolation. This unit can be
+a function, a file, a class, a module.
+
+In contrast to running code regularly, Unit Tests execute under the
+supervision of a `test runner` that:
+
+ - looks for (discovers) tested units,
+
+ - executes them with test data, collects test results regardless
+    whether a test succeeded or failed and
+
+ - reports test results at the and.
+
+Read *"A Beginner’s Guide to Unit Tests in Python"*,
+[link](https://www.dataquest.io/blog/unit-tests-python/),
+and answer questions:
+
+ - How are tests discovered? Which feature makes the test runner to collect
+    something as a test?
+
+ - What is an
+    [assert](https://docs.python.org/3/library/unittest.html#assert-methods)
+    statement? What happens when a test (assert) passes and fails?
+
+ - Where is the test runner started in given files?
+
+(1 Pt)
+
+Run tests in a terminal. Currently, only one test runs and passes:
+*TestCase_a_number_of_numbers* :
+```sh
+python test_expressions.py      # run tests directly from file calling the
+                                # test runner in __main__
+```
+
+Output:
+
+```
+test_a_number_of_numbers (C_expressions.test_expressions.TestCase_a_number_of_nu
+mbers.test_a_number_of_numbers) ... ok
+
+----------------------------------------------------------------------
+Ran 1 test in 0.001s
+
+OK
+<unittest.runner.TextTestResult run=1 errors=0 failures=0>
+```
+Result: 1 test was performed that passed.
+
+Alternatively, run tests with test discovery. Run the unit test module that
+starts the test runner, which in turn discovers tests that are then executed:
+
+```sh
+python -m unittest              # let test runner discover tests
+```
+Output is the same as above.
+
+(1 Pt)
+
+Configure your IDE so it runs Unit Tests (you can use other IDE than VS Code
+that is used here as example).
+
+VSCode discovers unit tests under the test glass icon (red circled).
+
+The figure shows one unit test that has been discovered passing. Unit tests are
+structured as *"TestCase - Classes"*, which are classes that inherit from class:
+[unittest.TestCase](https://docs.python.org/3/library/unittest.html#unittest.TestCase),
+in the example indirectly through class `Test_case_a`.
+
+VSCode shows discovered test classes in the left panel and their execution result
+with a green check mark when passed or a red cross when failed.
+
+![](../markup/img/C_unit_tests_1.png)
+
+Uncomment tests: *"Test_case_b"* and *"Test_case_c"* in `test_expressions.py`
+above and re-run tests.
+Both tests should fail because expressions they test have not been implemented:
+
+![](../markup/img/C_unit_tests_2.png)
+
+Re-run unit tests with the two tests failing in the terminal:
+
+```sh
+python -m unittest              # let test runner discover tests
+```
+
+Output shows one passing and two failed tests:
+
+```
+======================================================================
+FAIL: test_b_first_three_numbers (test_expressions.TestCase_b_first_three_number
+s.test_b_first_three_numbers)
+----------------------------------------------------------------------
+Traceback (most recent call last):
+  File "C:\Sven1\svgr\workspaces\ds_cs4bd_2324\C_expressions\test_expressions.py
+", line 103, in test_b_first_three_numbers
+    self.assertEqual(self.ut1.b, [4, 12, 3])
+AssertionError: Lists differ: [] != [4, 12, 3]
+
+Second list contains 3 additional elements.
+First extra element 0:
+4
+
+- []
++ [4, 12, 3]
+
+======================================================================
+FAIL: test_c_last_three_numbers (test_expressions.TestCase_c_last_three_numbers.
+test_c_last_three_numbers)
+----------------------------------------------------------------------
+Traceback (most recent call last):
+  File "C:\Sven1\svgr\workspaces\ds_cs4bd_2324\C_expressions\test_expressions.py
+", line 117, in test_c_last_three_numbers
+    td.assertEqual(td.ut1.c, [1, 8, 7])
+AssertionError: Lists differ: [] != [1, 8, 7]
+
+Second list contains 3 additional elements.
+First extra element 0:
+1
+
+- []
++ [1, 8, 7]
+
+----------------------------------------------------------------------
+Ran 3 tests in 0.002s
+
+FAILED (failures=2)
+```
+Output says: `Ran 3 tests`, `FAILED (failures=2)`.
+
+When tests fail, the test report tells which tests have failed and why:
+
+  - *test_b_first_three_numbers* failed in line: 103. The test expected
+    result: `[4, 12, 3]`, but an empty list `[]` was found in the tested
+    expression: `self.b` in file `expressions.py`.
+
+  - *test_c_last_three_numbers* failed in line: 117 where the test expected
+    result: `[1, 8, 7]`, but an empty list `[]` was found in: `self.c`
+
+Tests refer to the `self.numbers` list: `[4, 12, 3, 8, 17, 12, 1, 8, 7]`.
+
+(1 Pt)
+
+
+&nbsp;
+### 4.) Challenge: Write Expressions
+
+In order to let tests pass, write expressions in
+[expressions.py](https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/blob/main/C_expressions/expressions.py)
+for variables `self.b` .. `self.k` according to specification, e.g. write an
+expression for `self.b` that extracts the first three numbers from `self.numbers`.
+
+Use <b>one-line expressions</b> only.
+Python's [built-in functions](https://docs.python.org/3/library/functions.html)
+are allowed, but not own functions.
+
+Tests exercise expressions with various lists. Initialization with constants
+(`self.b = [4, 12, 3]`) will hence not work.
+
+Write expression incrementally, one after the other - not all at once. Some
+expressions require thinking and reading.
+
+Once you have written an expression, uncomment the corresponding test case in 
+[test_expressions.py](https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/blob/main/C_expressions/test_expressions.py):
+and re-run the test. See if it is passing or figure out why it is failing
+from the test report.
+
+![](../markup/img/C_unit_tests_3.png)
+
+Test cases a), b) and c) are now passing.
+
+Continue until all tests pass.
+
+![](../markup/img/C_unit_tests_4.png)
+
+
+&nbsp;
+### 5.) Challenge: Final Test and sign-off
+
+For sign-off, change into `C_expressions` directory and copy commands into a terminal:
+
+```sh
+# Fetch test file from Gitlab and run tests for sign-off.
+# The sed-command removes comments from test cases.
+
+test_url=https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/raw/main/C_expressions/test_expressions.py
+
+curl $test_url | \
+   sed -e 's/^#.*Test_case_/Test_case_/' | \
+   python
+```
+
+Result:
+
+```
+  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
+                                 Dload  Upload   Total   Spent    Left  Speed
+100  7874  100  7874    0     0  55666      0 --:--:-- --:--:-- --:--:-- 56242
+...........
+----------------------------------------------------------------------
+Ran 11 tests in 0.003s
+
+OK
+```
+
+11 tests succeeded.
+
+(10 Pts, 1 Pt for each test passing)

C_expressions/__init__.py

+"""
+Special file __init__.py marks a directory as a Python module.
+The file is executed once when any .py file is imported.
+//
+Python unittest require the presence of (even an empty) file.
+"""
+
+# load setup module when executed in parent directory
+try:
+    __import__('setup')
+#
+except ImportError:
+    pass

C_expressions/expressions.py

+
+class Expressions:
+    """"
+    Fill in one-line expressions (no own functions) to initialize attributes
+    self.b .. self.k with specified values.
+
+    Use Python built-in functions, list expressions and list comprehension,
+    but NOT own functions.
+
+    Complete tasks one after another. Once you are done with one task,
+    uncomment test cases in test_expressions.py. Remove comments for
+      # Test_case_b = Test_case
+      # Test_case_c = Test_case
+      # Test_case_d = Test_case
+      # ...
+    Run tests in IDE and in a terminal:
+      python test_expressions.py
+      python -m unittest
+    """
+
+    default_numbers=[4, 12, 3, 8, 17, 12, 1, 8, 7]
+
+    def __init__(self, _numbers=default_numbers):
+        """
+        Constructor to initialize member variables.
+        """
+        self.numbers = _numbers
+
+        # a) initialize with number of numbers: 9
+        self.a = len(self.numbers)    # <-- given solution, insert one-line expressions below
+
+        # b) initialize with first three numbers: [4, 12, 3]
+        self.b = []      # <-- write expression here
+
+        # c) initialize with last three numbers: [1, 8, 7]
+        self.c = []
+
+        # d) initialize with last three numbers reverse: [7, 8, 1]
+        self.d = []
+
+        # e) initialize with odd numbers: [3, 17, 1, 7]
+        self.e = []
+
+        # f) initialize with number of odd numbers: 4
+        self.f = 0
+
+        # g) initialize with sum_ of odd numbers: 28
+        self.g = 0
+
+        # h) duplicate numbers removed: [4, 12, 3, 8, 17, 1, 7]
+        self.h = []
+
+        # i) number of duplicate numbers: 2
+        self.i = 0
+
+        # j) ascending list of squared numbers with no duplicates: [1, 9, 16, 49, 64, 144, 289]
+        self.j = []
+
+        # k) initialize with "ODD_LIST", "EVEN_LIST" or "EMPTY_LIST" depending on numbers length
+        self.k = "NEITHER"
+
+        # attempt to load solution module (ignore)
+        try:
+            _from, _import = 'expressions_sol', 'Stream'
+            mod = __import__(_from, fromlist=[_import])
+            mod.set_solution(self)  # invoke set_solution() to replace values with solutions
+        #
+        except ImportError:
+            pass
+
+
+    def print_results(self):
+        print(f'\nnumbers: {self.numbers}\n#')
+        fmt = {
+            # key: (value, output string)
+            'a': (self.a, 'number of numbers'),
+            'b': (self.b, 'first three numbers'),
+            'c': (self.c, 'last three numbers'),
+            'd': (self.d, 'last three numbers reverse'),
+            'e': (self.e, 'odd numbers'),
+            'f': (self.f, 'number of odd numbers'),
+            'g': (self.g, 'sum of odd numbers'),
+            'h': (self.h, 'duplicate numbers removed'),
+            'i': (self.i, 'number of duplicate numbers'),
+            'j': (self.j, 'ascending, de-dup (n^2) numbers'),
+            'k': (self.k, 'length'),
+        }
+        # format output, e.g.: "b) first three numbers: [1, 4, 6]"
+        for k in sorted(fmt.keys()):
+            print(f'{k}) {fmt[k][1]}: {fmt[k][0]}')
+
+
+if __name__ == '__main__':
+    '''
+    Driver code that runs when this file is directly executed.
+    '''
+    #
+    n1 = Expressions()  # use default list
+    #
+    # 2nd object with different list
+    n2 = Expressions([1, 4, 6, 67, 6, 8, 23, 8, 34, 49, 67,
+        6, 8, 23, 37, 67, 6, 34, 19, 67, 6, 8])
+    #
+    n1.print_results()
+    # n2.print_results()     # try also other list

C_expressions/test_expressions.py

+"""
+Run unit tests with discovery (-m) or from __main__() with verbosity level 2
+ - python -m unittest
+ - python test_expressions.py
+
+Output with verbosity level < 2:
+================================
+...........
+----------------------------------------------------------------------
+Ran 11 tests in 0.002s
+OK
+<unittest.runner.TextTestResult run=11 errors=0 failures=0>
+"""
+import unittest
+import abc              # import Abstract Base Class (ABC) from module abc
+from expressions import Expressions
+
+
+"""
+tested objects (objects "under test", "ut") as instances of the Expressions class
+"""
+ut1 = Expressions(Expressions.default_numbers)  # [4, 12, 3, 8, 17, 12, 1, 8, 7]
+ut2 = Expressions([1, 4, 6, 67, 6, 8, 23, 8, 34, 49, 67, 6, 8, 23, 37, 67, 6, 34, 19, 67, 6, 8])
+ut3 = Expressions([6, 67, 6, 8, 17, 3, 6, 8])
+ut4 = Expressions([8, 3, 9])
+ut5 = Expressions([1, 1, 1])
+ut6 = Expressions([0, 0])
+ut7 = Expressions([0])
+ut8 = Expressions([])
+
+
+class Test_case(unittest.TestCase):
+    """
+    Top-level class that inherits from class unittest.TestCase
+    and injects test data into derived classes for test cases.
+    Sub-classes are discovered as unit tests.
+    """
+    def setUp(self):
+        self.ut1 = ut1
+        self.ut2 = ut2
+        self.ut3 = ut3
+        self.ut4 = ut4
+        self.ut5 = ut5
+        self.ut6 = ut6
+        self.ut7 = ut7
+        self.ut8 = ut8
+
+
+# disable tests by assigning Python's Abstract Base Class (ABC) to test
+# case classes, which will not be discovered as unit tests
+Test_case_a = Test_case_b = Test_case_c = Test_case_d = \
+Test_case_e = Test_case_f = Test_case_g = Test_case_h = \
+Test_case_i = Test_case_j = Test_case_k = abc.ABC
+
+
+# assign Test_case class (above) as subclass of unittest.TestCase and with
+# attributes of tested objects (self.ut1...ut8)
+# uncomment tests one after another as you progress with expressions
+Test_case_a = Test_case   # test a) passes, solution is given in numbers.py
+# Test_case_b = Test_case
+# Test_case_c = Test_case
+# Test_case_d = Test_case
+# Test_case_e = Test_case
+# Test_case_f = Test_case
+# Test_case_g = Test_case
+# Test_case_h = Test_case
+# Test_case_i = Test_case
+# Test_case_j = Test_case
+# Test_case_k = Test_case
+
+
+class TestCase_a_number_of_numbers(Test_case_a):
+    # 
+    # tests a): number of numbers tests (lengths of numbers lists)
+    def test_a_number_of_numbers(self):
+        self.assertEqual(self.ut1.a, 9)
+        self.assertEqual(self.ut2.a, 22)
+        self.assertEqual(self.ut3.a, 8)
+        self.assertEqual(self.ut4.a, 3)
+        self.assertEqual(self.ut5.a, 3)
+        self.assertEqual(self.ut6.a, 2)
+        self.assertEqual(self.ut7.a, 1)
+        self.assertEqual(self.ut8.a, 0)
+
+
+class TestCase_b_first_three_numbers(Test_case_b):
+    # 
+    # tests b): first three numbers
+    def test_b_first_three_numbers(self):
+        self.assertEqual(self.ut1.b, [4, 12, 3])
+        self.assertEqual(self.ut2.b, [1, 4, 6])
+        self.assertEqual(self.ut3.b, [6, 67, 6])
+        self.assertEqual(self.ut4.b, [8, 3, 9])
+        self.assertEqual(self.ut5.b, [1, 1, 1])
+        self.assertEqual(self.ut6.b, [0, 0])
+        self.assertEqual(self.ut7.b, [0])
+        self.assertEqual(self.ut8.b, [])
+
+
+class TestCase_c_last_three_numbers(Test_case_c):
+    # 
+    # tests c): last three numbers
+    def test_c_last_three_numbers(td):
+        td.assertEqual(td.ut1.c, [1, 8, 7])
+        td.assertEqual(td.ut2.c, [67, 6, 8])
+        td.assertEqual(td.ut3.c, [3, 6, 8])
+        td.assertEqual(td.ut4.c, [8, 3, 9])
+        td.assertEqual(td.ut5.c, [1, 1, 1])
+        td.assertEqual(td.ut6.c, [0, 0])
+        td.assertEqual(td.ut7.c, [0])
+        td.assertEqual(td.ut8.c, [])
+
+
+class TestCase_d_last_threeClass_in_reverse(Test_case_d):
+    # 
+    # tests d): last three numbers in reverse
+    def test_d_last_threeClass_in_reverse(td):
+        td.assertEqual(td.ut1.d, [7, 8, 1])
+        td.assertEqual(td.ut2.d, [8, 6, 67])
+        td.assertEqual(td.ut3.d, [8, 6, 3])
+        td.assertEqual(td.ut4.d, [9, 3, 8])
+        td.assertEqual(td.ut5.d, [1, 1, 1])
+        td.assertEqual(td.ut6.d, [0, 0])
+        td.assertEqual(td.ut7.d, [0])
+        td.assertEqual(td.ut8.d, [])
+
+
+class TestCase_e_odd_numbers(Test_case_e):
+    # 
+    # tests e): odd numbers, order must be preserved
+    def test_e_odd_numbers(td):
+        td.assertEqual(td.ut1.e, [3, 17, 1, 7])
+        td.assertEqual(td.ut2.e, [1, 67, 23, 49, 67, 23, 37, 67, 19, 67])
+        td.assertEqual(td.ut3.e, [67, 17, 3])
+        td.assertEqual(td.ut4.e, [3, 9])
+        td.assertEqual(td.ut5.e, [1, 1, 1])
+        td.assertEqual(td.ut6.e, [])
+        td.assertEqual(td.ut7.e, [])
+        td.assertEqual(td.ut8.e, [])
+
+
+class TestCase_f_number_of_odd_numbers(Test_case_f):
+    # 
+    # tests f): number of odd numbers
+    def test_f_number_of_odd_numbers(td):
+        td.assertEqual(td.ut1.f, 4)
+        td.assertEqual(td.ut2.f, 10)
+        td.assertEqual(td.ut3.f, 3)
+        td.assertEqual(td.ut4.f, 2)
+        td.assertEqual(td.ut5.f, 3)
+        td.assertEqual(td.ut6.f, 0)
+        td.assertEqual(td.ut7.f, 0)
+        td.assertEqual(td.ut8.f, 0)
+
+
+class TestCase_g_sum_of_odd_numbers(Test_case_g):
+    # 
+    # tests g): sum of odd numbers
+    def test_g_sum_of_odd_numbers(td):
+        td.assertEqual(td.ut1.g, 28)
+        td.assertEqual(td.ut2.g, 420)
+        td.assertEqual(td.ut3.g, 87)
+        td.assertEqual(td.ut4.g, 12)
+        td.assertEqual(td.ut5.g, 3)
+        td.assertEqual(td.ut6.g, 0)
+        td.assertEqual(td.ut7.g, 0)
+        td.assertEqual(td.ut8.g, 0)
+
+
+class TestCase_h_duplicateClass_removed(Test_case_h):
+    # 
+    # tests h): duplicate numbers removed - use set() to accept any order
+    def test_h_duplicateClass_removed(td):
+        td.assertEqual(set(td.ut1.h), {4, 12, 3, 8, 17, 1, 7})
+        td.assertEqual(set(td.ut2.h), {1, 4, 6, 67, 8, 23, 34, 49, 37, 19})
+        td.assertEqual(set(td.ut3.h), {6, 67, 8, 17, 3})
+        td.assertEqual(set(td.ut4.h), {8, 3, 9})
+        td.assertEqual(td.ut5.h, [1])
+        td.assertEqual(td.ut6.h, [0])
+        td.assertEqual(td.ut7.h, [0])
+        td.assertEqual(td.ut8.h, [])
+
+
+class TestCase_i_number_of_duplicate_numbers(Test_case_i):
+    # 
+    # tests i): number of duplicate numbers
+    def test_i_number_of_duplicate_numbers(td):
+        td.assertEqual(td.ut1.i, 2)
+        td.assertEqual(td.ut2.i, 12)
+        td.assertEqual(td.ut3.i, 3)
+        td.assertEqual(td.ut4.i, 0)
+        td.assertEqual(td.ut5.i, 2) # [1, 1, 1] has 2 duplicates 1
+        td.assertEqual(td.ut6.i, 1) # [0, 0] has one duplicate number 0
+        td.assertEqual(td.ut7.i, 0)
+        td.assertEqual(td.ut8.i, 0)
+
+
+class TestCase_j_ascending_squaredClass_no_duplicates(Test_case_j):
+    # 
+    # tests j): ascending list of squared numbers with no duplicates
+    def test_j_ascending_squaredClass_no_duplicates(td):
+        td.assertEqual(set(td.ut1.j), {1, 9, 16, 49, 64, 144, 289})
+        td.assertEqual(set(td.ut2.j), {1, 16, 36, 64, 361, 529, 1156, 1369, 2401, 4489})
+        td.assertEqual(set(td.ut3.j), {9, 36, 64, 289, 4489})
+        td.assertEqual(set(td.ut4.j), {9, 64, 81})
+        td.assertEqual(td.ut5.j, [1])
+        td.assertEqual(td.ut6.j, [0])
+        td.assertEqual(td.ut7.j, [0])
+        td.assertEqual(td.ut8.j, [])
+
+
+class TestCase_k_classifyClass_as_odd_even_empty(Test_case_k):
+    # 
+    # tests k): classify as "ODD_LIST", "EVEN_LIST" or "EMPTY_LIST" depending on numbers length
+    def test_k_classifyClass_as_odd_even_empty(td):
+        td.assertEqual(td.ut1.k, "ODD_LIST")
+        td.assertEqual(td.ut2.k, "EVEN_LIST")
+        td.assertEqual(td.ut3.k, "EVEN_LIST")
+        td.assertEqual(td.ut4.k, "ODD_LIST")
+        td.assertEqual(td.ut5.k, "ODD_LIST")
+        td.assertEqual(td.ut6.k, "EVEN_LIST")
+        td.assertEqual(td.ut7.k, "ODD_LIST")
+        td.assertEqual(td.ut8.k, "EMPTY_LIST")
+
+
+if __name__ == '__main__':
+    unittest.main()

Cygwin_setup.md

+# Cygwin setup on *Windows*
+
+[Cygwin](https://www.cygwin.com) is a Unix-Emulator that provides
+a terminal ([mintty](https://mintty.en.lo4d.com/windows))
+in which Unix commands can be executed on Windows using
+*[bash](https://en.wikipedia.org/wiki/Bash_(Unix_shell))*
+(*Bourne Again Shell*) as command-line interpreter.
+*bash* was developed in 1989 as a successor to the
+*Bourne Shell*
+*[sh](https://en.wikipedia.org/wiki/Bourne_shell)*.
+
+Example of a
+[bash terminal](https://cdn.ttgtmedia.com/rms/onlineimages/REF_bash_command_line_3.jpg).
+
+*Cygwin* is <span style="text-decoration:underline">not</span>
+a Unix system, container or virtual machine (like
+[WSL](https://learn.microsoft.com/en-us/windows/wsl/about)).
+*Cygwin* emulates most (but not all) Unix system calls such that
+most Unix commands can be used on Windows.
+
+[GitBash](https://gitforwindows.org)
+is an alternative to *Cygwin* that uses a different emulator package
+[MinGW](https://www.mingw-w64.org).
+It has has some flaws, for example, it performs path conversions that
+may cause problems, see
+[link](https://stackoverflow.com/questions/54258996/git-bash-string-parameter-with-at-start-is-being-expanded-to-a-file-path).
+
+Read
+[Differences between Cygwin and MinGW](https://stackoverflow.com/questions/771756/what-is-the-difference-between-cygwin-and-mingw)
+for more detail about the differences between *Cygwin* and *MinGW / GitBash*.
+
+Good introductions to *bash* are:
+
+- [Introduction to Bash](https://cs.lmu.edu/~ray/notes/bash).
+
+- [Bash Tutorial for Beginners](https://linuxconfig.org/bash-scripting-tutorial-for-beginners).
+
+
+&nbsp;
+## Steps
+1. [Install *Cygwin*](#1-install-cygwin)
+2. [Configure *Cygwin* and *bash*](#2-configure-cygwin)
+    - switch from `/cygdrive/c` to `/c`
+    - select HOME-directory
+    - configure *.bashrc* in HOME-directory
+    - define PATH in *.bashrc*
+3. [Customize *bash*](#3-customize-bash)
+4. [References](#4-references)
+
+
+&nbsp;
+## 1. Install Cygwin
+
+1. Download the installer `setup-x86_64.exe` from
+[https://www.cygwin.com/install.html](https://www.cygwin.com/install.html).
+
+
+1. Run the installer `setup-x86_64.exe`.
+
+    - select useful Unix commands during installation:
+        - Editors/`vim` - visual editor,
+        - Web/`wget` - web downloader,
+        - Net/`curl` - web downloader.
+
+
+1. Change *cygwin* default path `/cygdrive/c` to: `/c`:
+
+    - navigate to the *cygwin* installation directory called `cygwin64`.
+
+    - inside, edit `/etc/fstab` and replace line
+        ```
+        none /cygdrive cygdrive binary,posix=0,user 0 0
+        with:
+        none / cygdrive binary,posix=0,user 0 0
+        ```
+
+
+1. Add bash-user that works with Windows and create
+   *bash* HOME directory.
+
+    - Find out your Windows user name: <user_name>
+
+    - Create or select a directory to use as HOME-directory for
+      *bash*, e.g. your Windows HOME-directory `C:\Users\<user_name>`
+      (but also any other directory you may create as HOME).
+
+      Read
+      [Change Cygwin home folder after installation](https://stackoverflow.com/questions/1494658/how-can-i-change-my-cygwin-home-folder-after-installation).
+
+    - Edit file `/etc/nsswitch.conf`:
+        - to use your Windows HOME-directory, comment line (put hash # in front)
+            ```
+            #db_home:
+            ```
+        - to use a different HOME directory `<path>`, enter
+            ```sh
+            db_home: /c/<path>      # e.g. db_home: /c/users/svgr
+            ```
+
+
+1. Create a terminal icon on your Desktop.
+
+    - In the installation directory (`cygwin64`), navigate to `./bin`
+        and find file
+        [mintty.exe](https://en.wikipedia.org/wiki/Mintty),
+        which is the terminal emulator executable.
+
+    - Create a shortcut for *mintty.exe*
+        (right-click file -> Create Shortcut -> On Desktop).
+        The terminal icon appears on your desktop.
+
+
+1. Open *mintty* terminal and test:
+
+    ```sh
+    $ whoami
+    <user_name>         # your user name appears
+
+    $ echo $HOME        # print path to HOME-directory
+    <home_directory>    # output, e.g. /c/Users/svgr
+
+    $ cd $HOME          # change into HOME-directory
+    $ ls -la            # show content
+    total 203
+    drwxr-xr-x+ 1 svgr2 Kein      0 Oct  4 22:20 .
+    drwxrwx---+ 1 svgr2 Kein      0 Jan  1  2022 ..
+    -rw-------  1 svgr2 Kein  14476 Oct  4 22:20 .bash_history
+    -rwxr-xr-x+ 1 svgr2 Kein   2717 Oct  4 20:28 .bashrc
+    ...
+
+    $ echo Hello >hello.txt     # create file hello.txt with content Hello
+    $ ls -la                    # show new file exists in $HOME
+    total 203
+    -rw-r--r--  1 svgr2 Kein      6 Oct  7 23:38 hello.txt
+    ...
+
+    $ cat hello.txt             # output content of file
+    Hello
+    ```
+
+
+1. Open Windows file explorer (
+    [ ? ](https://geekflare.com/wp-content/uploads/2021/06/14-alternative-file-managers-to-replace-windows-10-file-explorer.jpg)
+    ) and navigate to file `hello.txt`.
+    Open the file by clicking (ending `.txt` should open the
+    file with *notepad*).
+
+
+1. Return to *bash* terminal:
+    - change to HOME-directory and
+    - show content of file `.bashrc` using a text editor:
+        *[vim](https://www.vim.org)* (already installed with cygwin),
+        *[nano](https://www.nano-editor.org)* or
+        *[sublime](https://www.sublimetext.com)*
+        are good choices.
+        ```sh
+        $ cd            # change to bash HOME-directory
+        $ ls -la        # find file .bashrc
+        total 203
+        -rwxr-xr-x+ 1 svgr2 Kein   2717 Oct  4 20:28 .bashrc
+
+        $ cat .bashrc   # output file .bashrc
+        content of file .bashrc appears...
+        ```
+
+
+&nbsp;
+## 2. Configure *Cygwin*
+
+[PATH](https://en.wikipedia.org/wiki/PATH_(variable)) is an
+environment variable on Unix-like operating systems that
+specifys a set of directories where executable programs are
+located. A *"command not found"* error occurs when PATH is
+not properly configured.
+
+
+1. Inspect the current PATH variable on your system:
+    ```sh
+    echo $PATH
+    .:/usr/bin:/usr/local/bin:/c/WINDOWS:/c/WINDOWS/system32: ...
+    ```
+    PATH has a list of directory paths starting from the root directory '/'
+    separated by colon ( : ).
+
+    To print PATH more readable, replace ( : ) by newlines:
+    ```sh
+    echo $PATH | tr ':' '\n'
+    .
+    /usr/bin
+    /usr/local/bin
+    /c/WINDOWS
+    /c/WINDOWS/system32
+    ...
+    ```
+    Make sure these paths are included in PATH.
+
+
+1. Open `.bashrc` using a text editor
+    (*[sublime](https://www.sublimetext.com)* or
+    *[vim](https://www.vim.org)* - already installed with
+    *cygwin* - are good choices).
+    Append lines for PATH configurations at the end.
+    ```sh
+    $ vim .bashrc       # open file .bashrc in vim editor
+    ```
+    or drag file from file explorer into sublime for editing.
+
+
+1. Append following PATH configurations at the end of the file -
+    only those relevant on your system adjusted for your system:
+
+    ```sh
+    # add Windows system paths
+    export PATH=".:/usr/bin:/usr/local/bin"
+    export PATH="${PATH}:$(cygpath ${SYSTEMROOT}):$(cygpath ${SYSTEMROOT})/system32"
+
+    # if Java, add Java path (as first entry on PATH)
+    # make sure the path to Java exists on your system and is the path
+    # to a JDK (Java Development Kit) and not JRE, which is just the
+    # Java Runtime that has no compiler, no javadoc or jar.
+    export JAVA_HOME="/c/Program Files/Java/jdk-21"
+    export PATH="${JAVA_HOME}/bin:${PATH}"
+
+    # if Python, add Python path
+    export PYTHON_HOME="/c/Users/svgr2/AppData/Local/Programs/Python/Python312"
+    export PATH="${PATH}:${PYTHON_HOME}"
+    export PATH="${PATH}:${PYTHON_HOME}/Scripts"
+
+    # if Docker, add Docker path
+    export DOCKER_HOME="/c/Program Files/Docker/Docker"
+    export PATH="${PATH}:${DOCKER_HOME}/resources/bin"
+    ...
+    ```
+
+
+1. Verify paths have been added to *PATH* variable:
+    ```sh
+    $ source .bashrc    # reload .bashrc to activate PATH definitions
+    $ echo ${PATH}                      # show PATH
+    .:/usr/bin:/usr/local/bin:/c/WINDOWS:/c/WINDOWS/system32:/c/Program ...
+
+    $ echo ${$PATH} | tr ':' '\n'       # pretty print PATH
+    .
+    /c/Program Files/Java/jdk-21/bin    <-- path to JDK executables
+    /usr/bin
+    /usr/local/bin
+    /c/WINDOWS
+    /c/WINDOWS/system32
+    ...
+    ```
+    Paths may vary based on your system.
+
+
+1. Start programs to verify paths have properly been set.
+
+    ```sh
+    $ java --version
+    java 21 2023-09-19 LTS
+    Java(TM) SE Runtime Environment (build 21+35-LTS-2513)
+    Java HotSpot(TM) 64-Bit Server VM (build 21+35-LTS-2513, mixed mode, sharing)
+
+    $ python --version
+    Python 3.12.0
+
+    $ docker --version
+    Docker version 24.0.6, build ed223bc
+    ```
+
+
+&nbsp;
+## 3. Customize *bash*
+
+*bash* is widely customizable, usually by settings in *.bashrc*.
+
+- [Bash Prompt customization](https://wiki.archlinux.org/title/Bash/Prompt_customization)
+
+- [How to Change Colors on LS in Bash](https://linuxhint.com/ls_colors_bash)
+
+- [How to Customize and Colorize the Bash Prompt](https://www.howtogeek.com/307701/how-to-customize-and-colorize-your-bash-prompt/)
+
+Skip when no customization is desired.
+
+
+&nbsp;
+## 4. References
+
+- [What is PATH?](https://en.wikipedia.org/wiki/PATH_(variable))
+
+- Bash tutorials:
+    - [Introduction to Bash](https://cs.lmu.edu/~ray/notes/bash).
+    - [Tutorial for Beginners](https://linuxconfig.org/bash-scripting-tutorial-for-beginners).
+
+- [Differences between Cygwin and MinGW](https://stackoverflow.com/questions/771756/what-is-the-difference-between-cygwin-and-mingw)
+
+- [Change Cygwin HOME directory after installation](https://stackoverflow.com/questions/1494658/how-can-i-change-my-cygwin-home-folder-after-installation).
+

D_recursion/README.md

+# Assignment D: Recursive Problem Solving   (15 Pts + 4 Extra Pts)
+
+Recursion is not just a *"function calling itself"*, it is a way of thinking
+about a class of problems that can be split into simple "*base cases"* and
+remaining *"sub-problems"* that are *"self-similar"*.
+
+A *"sub-problem"* is self-similar when it exactly looks the same as the
+original problem, just smaller (e.g. reduced by one element). At some point,
+the simple "*base case"* has been reached that yields a primitive solution.
+
+A recursive *solution function* exploiting self-similarity has two phases:
+
+  1. *Reduction:* - slicing the problem (e.g. a list of numbers) into one
+        element (e.g. the first number) and a remaining *sub-problem*
+        (e.g. the list of remaining numbers).
+
+  1. *Recursion:* -  invoke the same function for the *sub-problem*
+        until the *sub-problem* has been reduced to the *base case*.
+        Return the solution for the *base case*.
+
+  1. *Construction:* - results of recursive invocations are considered
+        as solutions of *sub-problems* and are combined with the element
+        that was isolated at the particular level of recursion.
+
+While this approach is elegant from a thinking-about-problems and programming
+point of view, it has cost associated for using the
+[Callstack](https://en.wikipedia.org/wiki/Call_stack)
+using a data structure of an abstract data type
+[Stack](https://en.wikipedia.org/wiki/Stack_(abstract_data_type))
+for recursions.
+
+
+### Challenges
+- [Challenge 1:](#1-challenge-simple-recursion-sum-numbers) Simple recursion: *sum* numbers
+- [Challenge 2:](#2-challenge-fibonacci-numbers) Fibonacci numbers
+- [Challenge 3:](#3-challenge-permutation) Permutation
+- [Challenge 4:](#4-challenge-powerset) Powerset
+- [Challenge 5:](#5-challenge-find-matching-pairs) Find Matching Pairs
+- [Challenge 6:](#6-challenge-combinatorial-problem-of-finding-numbers) Combinatorial Problem of Finding Numbers
+- [Challenge 7:](#7-challenge-hard-problem-of-finding-numbers) Hard Problem of Finding Numbers
+
+Points: [2, 1, 2, 2, 2, 3, 2, +4 extra pts]
+
+File [recursion.py](recursion.py) has function headers defined for each challenge.
+
+Use those functions and complete code.
+
+
+ 
+### 1.) Challenge: Simple recursion: *sum()* numbers
+
+Computing the *sum* of numbers is most often performed *iteratively* as a loop
+over given numbers and adding them in a result variable.
+
+Solving the problem *recursively* illustrates the concept of self-similarity
+and recursive problem solving.
+
+Use the following approach:
+
+  1. *Reduction:* - split the given list of numbers into a first element (first number)
+        and a list of remaining numbers (*sub-problem*). Remember the first element.
+
+  1. *Recursion:* - invoke *sum()* for the list of remaining numbers until the base case
+        has been reached: *sum( [ ] ) = 0* or *sum( [n] )=n*.
+
+  1. *Construction:* - add the remembered element to the value returned from the
+        recursive invocation of *sum()*.
+
+Complete: `sum(_numbers)` using this approach:
+
+```py
+def sum(self, _numbers) -> int:
+    # your code
+    return #...
+```
+
+Remove comment from `run_choices` and run the program:
+
+```py
+    run_choices = [
+        1,      # Challenge 1, Simple recursion: sum numbers
+        ...
+    ]
+```
+
+Output:
+
+```
+n1.numbers: [9, 4, 8, 10, 2, 4, 8, 3, 14, 4, 8]
+sum(n1.numbers): 74
+```
+
+Answer questions:
+
+  1. How many time is the *"first element"* stored?
+    How much memory is used for applying the function to a list of *n* numbers?
+
+  1. What is the run-time estimate for *sum()* given a list of *n* numbers?
+
+  1. How many *stack-frames* are used for a list of *n* numbers?
+
+(2 Pts)
+
+
+ 
+### 2.) Challenge: Fibonacci numbers
+
+[Fibonacci numbers](https://en.wikipedia.org/wiki/Fibonacci_number) were first
+described in Indian mathematics as early as 200 BC in works by *Pingala* on
+enumerating possible patterns of Sanskrit poetry formed from syllables of two lengths.
+
+Italian mathematician *Leonardo of Pisa*, later known as
+*[Fibonacci](https://en.wikipedia.org/wiki/Fibonacci)*,
+introduced the sequence to Western European mathematics in his 1202 book
+*[Liber Abaci](https://en.wikipedia.org/wiki/Liber_Abaci)*.
+
+Numbers of the *Fibonacci sequence* are defined as: *fib(0): 0*, *fib(1): 1*, *...*
+and each following number is the sum of the two preceding numbers.
+
+Fibonacci numbers are widely found in *nature*, *science*, *social behaviors* of
+populations and *arts*, e.g. they form the basis of the
+[Golden Ratio](https://www.adobe.com/creativecloud/design/discover/golden-ratio.html),
+which is widely used in *painting* and *photography*, see also this
+[1:32min](https://www.youtube.com/watch?v=v6PTrc0z4w4) video.
+
+<img src="../markup/img/fibonacci.jpg" alt="drawing" width="640"/>
+<!-- ![image](../markup/img/fibonacci.jpg) -->
+
+&nbsp;
+
+Complete functions `fib(n)` and `fib_gen(n)`.
+
+```py
+def fib(self, _n) -> int:
+    # return value of n-th Fibonacci number
+    return #...
+
+def fib_gen(self, _n):
+    # return a generator object that yields two lists, one with n and the
+    # other with corresponding fib(n)
+    yield #...
+```
+
+Remove comment from `run_choices` and run the program:
+
+```py
+    run_choices = [
+        1,      # Challenge 1, Simple recursion: sum numbers
+        2,      # Challenge 2, Fibonacci numbers
+        ...
+    ]
+```
+
+Output:
+
+```
+n:      [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]
+fib(n): [0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1597, 2584, 4181, 6765]
+```
+
+Answer questions:
+
+  1. Explain the concept of a generator in Python.
+
+  1. Why can't `fib(60)` or `fib(90)` be computed recursively?
+
+  1. What is the more limiting constraint: memory use or needed run time?
+        ```py
+        n = 30
+        print(f'fib({n}): {n1.fib(n)}')
+        n = 60
+        print(f'fib({n}): {n1.fib(n)}')     # ??
+        n = 90
+        print(f'fib({n}): {n1.fib(n)}')     # ??
+        ```
+
+Understand the problem and use a technique called
+[memoization](https://stackoverflow.com/questions/7875380/recursive-fibonacci-memoization)
+to make the solution work for *n=60* and *n=90* - still recursively (!).
+
+Remove comments `#21` and `#22` from `run_choices` and run the program:
+
+Output:
+
+```
+fib(30): 832040
+fib(60): 1548008755920
+fib(90): 2880067194370816120
+```
+
+(2 Pts)
+
+
+&nbsp;
+### 3.) Challenge: Permutation
+
+[Permutation](https://en.wikipedia.org/wiki/Permutation) is a list of all
+arrangements of elements.
+
+For example:
+```py
+perm([1, 2, 3]) -> [[1, 2, 3], [1, 3, 2], [2, 1, 3], [2, 3, 1], [3, 1, 2], [3, 2, 1]]
+
+perm([]) -> [[]]
+perm([1]) -> [[1]]
+perm([1, 2]) -> [[1, 2], [2, 1]]
+perm([1, 2, 3]) -> [[1, 2, 3], [1, 3, 2], [2, 1, 3], [2, 3, 1], [3, 1, 2], [3, 2, 1]]
+perm([1, 2, 3, 4]) -> [[1, 2, 3, 4], [1, 2, 4, 3], ... [4, 3, 1, 2], [4, 3, 2, 1]]
+```
+Find a pattern how numbers are arranged for `perm([1, 2])` and `perm([1, 2, 3])`
+and adapt it for `perm([1, 2, 3, 4])` to understand the algorithm.
+
+Writing non-recursive code for that algorithm can be difficult, but it fits
+well with the recursive sub-problen approach, which is elegant with the
+four steps:
+1. Return solutions for trivial cases: `[]`, `[1]`, `[1, 2]`.
+1. Split the problem by removing the first number `n1` from the list leaving `r` as
+    remaining list (sub-problem).
+1. Invoke `perm(r)` recursively on the remaining list.
+1. Combine the result returned from `perm(r)` by adding `n1` to each element.
+
+```py
+def perm(self, _numbers) -> list:
+    res=[]  # collect result
+    # code...
+    # 1. Return solutions for trivial cases: `[]`, `[1]`, `[1, 2]`.
+    # 2. Split the problem by removing the first number `n1` from the list
+    #    leaving `r` as remaining list (sub-problem).
+    # 3. Invoke `perm(r)` recursively on the remaining list.
+    # 4. Combine the result by adding `n1` to each returned element from `perm(r)`.
+    #
+    return res
+
+lst = [1, 2, 3]
+perm = n1.perm(lst)
+print(f'perm({lst}) -> {perm}')
+
+lst = [1, 2, 3, 4]
+perm = n1.perm(lst)
+print(f'perm({lst}) -> {perm}')
+```
+
+Remove comment `#3` from `run_choices` and run the program:
+
+Output:
+
+```
+perm([1, 2, 3]) -> [[1, 2, 3], [1, 3, 2], [2, 1, 3], [2, 3, 1], [3, 1, 2], [3, 2, 1]]
+perm([1, 2, 3, 4]) -> [[1, 2, 3, 4], [1, 2, 4, 3], ... [4, 3, 1, 2], [4, 3, 2, 1]]
+```
+
+Answer questions:
+
+  - With a rising length of the input list, how does the number of permutations grow?
+
+(2 Pts)
+
+
+&nbsp;
+### 4.) Challenge: Powerset
+
+[Powerset](https://en.wikipedia.org/wiki/Powerset) is a list of all
+subsets of elements including the empty set.
+
+For example:
+```py
+pset([1, 2, 3]) -> [[], [1], [2], [1, 2], [3], [1, 3], [2, 3], [1, 2, 3]]
+```
+Undertstand the pattern and complete function `pset()`.
+
+```py
+def pset(self, _numbers) -> list:
+    res=[]  # collect result
+    # code...
+    # 1. Return solutions for trivial cases: `[]`, `[1]`, `[1, 2]`.
+    # 2. Split the problem by removing the first number `n1` from the list
+    #    leaving `r` as remaining list (sub-problem).
+    # 3. Invoke `pset(r)` recursively on the remaining list.
+    # 4. Combine the result with the first element.
+    #
+    return res
+
+lst = [1, 2, 3]
+pset = n1.pset(lst)
+print(f'pset({lst}) -> {pset}')
+```
+
+Remove comment `#4` from `run_choices` and run the program:
+
+Output:
+
+```py
+pset([1, 2, 3]) -> [[], [1], [2], [1, 2], [3], [1, 3], [2, 3], [1, 2, 3]]
+```
+
+Answer questions:
+
+  - With a rising length of the input list, how does the size of the Powerset grow?
+
+(2 Pts)
+
+
+&nbsp;
+### 5.) Challenge: Find Matching Pairs
+
+Write three functions to `find` elements in a list.
+
+The first function to `find` elements that match a boolean `match_func`.
+
+A second function `find_adjacent` that finds all indexes of adjacent pairs
+of numbers.
+
+The third function `find_pairs` that finds all pairs of numbers (not necessarily
+adjacent) with the sum equal to `n`. For example, `n=12` can be combined from
+the input list with pairs: `[3, 9], [4, 8], [2, 10]`.
+
+```py
+def find(self, _numbers, match_func) -> list:
+    res = []    # code...
+    return res
+
+
+def find_adjacent(self, pair, _numbers) -> list:
+    res = []    # code...
+    return res
+
+
+def find_pairs(self, n, _numbers) -> list:
+    res = []    # code...
+    return res
+
+
+lst = [9, 4, 8, 10, 2, 4, 8, 3, 14, 4, 8]   # input list
+#
+div3 = n1.find(lst, match_func=lambda n : n % 3 == 0)
+print(f'find numbers divisible by 3: {div3}')
+#
+p = [4, 8]  # find all indexes of adjacent numbers [4, 8]
+adj = n1.find_adjacent(p, lst)
+print(f'find_adjacent({p}, list): {adj}')
+#
+n = 12  # find all pairs from the input list that add to n
+pairs = n1.find_pairs(n, lst)
+print(f'find_pairs({n}, list) -> {pairs}')
+```
+
+Remove comments `#5`, `#51` and  `#52` from `run_choices` and run the program:
+
+Output:
+
+```
+find numbers divisible by 3: [9, 3]
+
+find_adjacent([4, 8], list): [1, 5, 9]
+
+find_pairs(12, list) -> [[3, 9], [4, 8], [2, 10]]
+```
+
+Answer questions:
+
+  - With a rising length of the input list, how many steps are needed to
+    complete each function in the best and worst case and on average?
+
+    | function         | answers     |
+    | ---------------- | ----------- |
+    | `find`           | best case: ______, worst case: ______, average: ______ steps.  |
+    | `find_adjacent`  | best case: ______, worst case: ______, average: ______ steps.  |
+    | `find_pairs`       | best case: ______, worst case: ______, average: ______ steps.  |
+
+(3 Pts)
+
+
+&nbsp;
+### 6.) Challenge: Combinatorial Problem of Finding Numbers
+
+`find_all_sums` is a function that returns any combination of numbers from the
+input list that add to `n`. For example, `n=14` can be combined from an
+input list: `[8, 10, 2, 14, 4]` by combinations: `[4, 8, 2], [4, 10], [14]`.
+
+A first approach to the problem is to understand the nature of possible
+combinations from the input list. If all those combinations could be
+generated, each could be tested whether their elements add to `n` and if,
+collect them for the final result.
+
+The order of numbers in solutions is not relevant (summation is commutative).
+Duplicate solutions with same numbers, but in different order need be to removed.
+
+```py
+def find_all_sums(self, n, _numbers) -> list:
+    res = []    # code...
+    return res
+
+
+lst = [8, 10, 2, 14, 4]     # input list
+n = 14
+all = n1.find_all_sums(n, lst)
+print(f'find_all_sums({n}, lst) -> {all}')
+```
+
+Output:
+```py
+find_all_sums(14, lst) -> [[4, 8, 2], [4, 10], [14]]
+```
+
+Test your solution with a larger input set:
+```py
+lst = [     # input list
+    260, 720, 225, 179, 101, 767, 167, 200, 157, 289,
+    318, 303, 153, 290, 201, 594, 457, 607, 592, 246,
+]
+n = 469
+all = n1.find_all_sums(n, lst)
+print(f'find_all_sums({n}, lst) -> {all}')
+```
+
+Remove comments `#6`, and  `#61` from `run_choices` and run the program:
+
+Output:
+
+```
+find_all_sums(469, lst) -> [[179, 290], [101, 167, 201]]
+```
+
+Answer questions:
+
+  - With a rising length of the input list, how does the number of possible
+    solutuions rise that must be tested?
+
+(2 Pts)
+
+
+&nbsp;
+### 7.) Challenge: Hard Problem of Finding Numbers
+
+Larger data sets can no longer be solved *"brute force"* by exploring all possible
+2^n combinations.
+
+Find a solution using a recursive approach exploring a decision tree or
+with tabulation.
+
+```py
+lst = [     # input list
+    260, 720, 225, 179, 101, 767, 167, 200, 157, 289,
+    318, 303, 153, 290, 201, 594, 457, 607, 592, 246,
+    132, 135, 584, 432, 591, 204, 417, 405, 362, 658,
+    136, 751, 583, 536, 293, 493, 431, 780, 563, 703,
+    400, 618, 397, 320, 513, 708, 319, 317, 685, 347,
+    758, 439, 145, 378, 158, 384, 551, 110, 408, 648,
+    847, 498,  50,  19,     # 64 numbers
+]
+n = 469
+all = n1.find_all_sums(n, lst)
+for i, s in enumerate(all):
+    print(f' - {i+1:2}: sum({sum(s)}) -> {s}')
+```
+
+Sort output by lenght of solution (use length as primary and numeric value
+of first element as secondary criteria).
+
+Remove comment `#7` if you tackled the challenge and run the program:
+
+Output:
+
+```
+  1: sum(469) -> [290, 179]
+  2: sum(469) -> [19, 157, 293]
+  3: sum(469) -> [19, 246, 204]
+  4: sum(469) -> [19, 318, 132]
+  5: sum(469) -> [19, 400, 50]
+  6: sum(469) -> [50, 101, 318]
+  7: sum(469) -> [110, 201, 158]
+  8: sum(469) -> [136, 201, 132]
+  9: sum(469) -> [145, 167, 157]
+ 10: sum(469) -> [158, 179, 132]
+ 11: sum(469) -> [201, 101, 167]
+ 12: sum(469) -> [19, 101, 204, 145]
+ 13: sum(469) -> [19, 157, 135, 158]
+ 14: sum(469) -> [19, 179, 135, 136]
+ 15: sum(469) -> [19, 204, 136, 110]
+ 16: sum(469) -> [19, 290, 110, 50]
+ 17: sum(469) -> [19, 101, 167, 132, 50]
+ 18: sum(469) -> [19, 132, 158, 110, 50]
+```
+
+( +4 Extra Pts)

D_recursion/recursion.py

+from functools import cmp_to_key
+"""
+Assignment_D: recursion
+"""
+
+class Recursion:
+
+    numbers = [9, 4, 8, 10, 2, 4, 8, 3, 14, 4, 8]    #[4, 12, 3, 8, 17, 12, 1, 3, 8, 7]
+
+
+    def sum(self, _numbers) -> int:
+        """
+        Return sum of numbers using recursion.
+        Follow steps:
+            1. Return 0 for an empty list of numbers.
+            2. Split the problem by removing the first number `n1` from the list leaving `r` as remaining list (sub-problem).
+            3. Invoke `sum(r)` recursively on the remaining list.
+            4. Combine the result for the sub-problem with the first number `n1`: `return n1 + sum(r)`.
+        """
+        # your code
+        return 0
+
+
+    def fib(self, _n, memo=None) -> int:
+        """
+        Return value of n-th Fibonacci number.
+        - input: n=8
+        - output: 21
+        """
+        # your code
+        return 0
+
+
+    def fib_gen(self, _n):
+        """
+        Return a generator object that yields two lists, one with n and the
+        other with corresponding fib(n).
+        - input: n=16
+        - output: generator object that produces:
+            ([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
+             [0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987])
+        """
+        # your code
+        yield ([], [])
+
+
+    def perm(self, _numbers) -> list:
+        """
+        Return permutation (all possible arrangements) for a given input list.
+        - input: [1, 2, 3]
+        - output: [[1, 2, 3], [1, 3, 2], [2, 1, 3], [2, 3, 1], [3, 1, 2], [3, 2, 1]]
+        """
+        # your code
+        return []
+
+
+    def pset(self, _numbers) -> list:
+        """
+        Return powerset (set of all subsets) for a given input list.
+        - input: [1, 2, 3]
+        - output: powerset, [[], [1], [2], [1, 2], [3], [1, 3], [2, 3], [1, 2, 3]]
+        """
+        # your code
+        return []
+
+
+    def find(self, _numbers, match_func) -> list:
+        """
+        Return list of elements n for which match_func(n) evaluates True.
+        """
+        # your code
+        return []
+
+
+    def find_adjacent(self, pair, _numbers, _i=0) -> list:
+        """
+        Return list of indexes of adjacent numbers in _numbers.
+        """
+        # your code
+        return []
+
+
+    def find_pairs(self, n, _numbers) -> list:
+        """
+        Return list of pairs from _numbers that add to n,
+        any pair, any order, no duplicates.
+        """
+        # your code
+        return []
+
+
+    def find_all_sums(self, n, _numbers) -> list:
+        """
+        Return all combinations of numbers in _numbers that add to n,
+        (any pair, any order, no duplicates).
+        """
+        # your code
+        return []
+
+
+    def __init__(self, _numbers=numbers):
+        """
+        Constructor to initialize member variables.
+        """
+        self.numbers = _numbers
+
+    run_choices = [
+        1,      # Challenge 1, Simple recursion: sum numbers
+        2,      # Challenge 2, Fibonacci numbers
+        # 21,     # Challenge 2.1, fig_gen()
+        # 22,     # Challenge 2.2, memoization, fib(60), fib(90)
+        # 3,      # Challenge 3, Permutation
+        # 4,      # Challenge 4, Powerset
+        # 5,      # Challenge 5, Finding matches, find()
+        # 51,     # Challenge 5.1, find_adjacent() pairs
+        # 52,     # Challenge 5.2, find_pairs() that add to n
+        # 6,      # Challenge 6, Find all combinations that add to n
+        # 61,     # Challenge 6.1, Find all in medium set
+        # 7       # Challenge 7, Hard problem finding numbers (extra points)
+    ]
+
+
+# Ignore this code that loads solution from file, if exists.
+# The solution is not distributed.
+try:
+    _from, _import = 'recursion_sol', 'Recursion'
+    Recursion = getattr(__import__(_from, fromlist=[_import]), _import)
+#
+except ImportError:
+    pass
+
+
+if __name__ == '__main__':
+    """
+    Main driver that runs when this file is executed by Python interpreter.
+    """
+    run_choices = Recursion.run_choices
+
+    numbers = [9, 4, 8, 10, 2, 4, 8, 3, 14, 4, 8]
+    n1 = Recursion(numbers)
+    print(f'n1.numbers: {n1.numbers}')
+
+    # Challenge 1, Simple recursion: sum numbers
+    if 1 in run_choices:
+        s = n1.sum(n1.numbers)
+        print(f'sum(n1.numbers): {s}')
+
+
+    # Challenge 2, Fibonacci numbers
+    if 2 in run_choices:
+        n = 30
+        print(f'\nfib({n}): {n1.fib(n)}')
+
+    # Challenge 2.1, fig_gen()
+    if 21 in run_choices:
+        gen = n1.fib_gen(20)    # yield generator object
+        n, fib = next(gen)      # trigger generator
+        print(f'n:      {n}')
+        print(f'fib(n): {fib}')
+
+    # Challenge 2.2, memoization, fib(60), fib(90)
+    if 22 in run_choices:
+        n = 60
+        print(f'fib({n}): {n1.fib(n)}')     # ??
+        n = 90
+        print(f'fib({n}): {n1.fib(n)}')     # ??
+
+
+    # Challenge 3, Permutation
+    if 3 in run_choices:
+        lst = [1, 2, 3]
+        perm = n1.perm(lst)
+        print(f'\nperm({lst}) -> {perm}')
+
+
+    # Challenge 4, Powerset
+    if 4 in run_choices:
+        lst = [1, 2, 3]
+        pset = n1.pset(lst)
+        print(f'\npset({lst}) -> {pset}')
+
+
+    lst = n1.numbers
+    #
+    # Challenge 5, Finding matches, find()
+    if 5 in run_choices:
+        div3 = n1.find(lst, match_func=lambda n : n % 3 == 0)
+        print(f'\nfind numbers divisible by 3: {div3}')
+
+    # Challenge 5.1, find_adjacent() pairs
+    if 51 in run_choices:
+        pair = [4, 8]
+        adj = n1.find_adjacent(pair, lst)
+        print(f'find_adjacent({pair}, list): {adj}')
+
+    # Challenge 5.2, find_pairs() that add to n
+    if 52 in run_choices:
+        n = 12
+        pairs = n1.find_pairs(n, lst)
+        print(f'find_pairs({n}, list) -> {pairs}')
+
+
+    lst = [8, 10, 2, 14, 4]     # input list
+    #
+    # Challenge 6, Find all combinations that add to n
+    if 6 in run_choices:
+        print(f'\nlist: {lst}\n\\\\')
+        n = 14
+        all = n1.find_all_sums(n, lst)
+        print(f'find_all_sums({n}, lst) -> {all}')
+        #
+        n = 20
+        all = n1.find_all_sums(n, lst)
+        print(f' - find_all_sums({n}, lst) -> {all}')
+        #
+        n = 32
+        all = n1.find_all_sums(n, lst)
+        print(f' - find_all_sums({n}, lst) -> {all}')
+
+
+    lst = [     # input list
+        260, 720, 225, 179, 101, 767, 167, 200, 157, 289,
+        318, 303, 153, 290, 201, 594, 457, 607, 592, 246,
+    ]
+    #
+    # Challenge 6.1, Find all in medium set
+    if 61 in run_choices:
+        print(f'\nlist({len(lst)}): {lst}\n\\\\')
+        n = 101 + 201 + 167     # 469 -> [[179, 290], [101, 167, 201]]
+        all = n1.find_all_sums(n, lst)
+        for i, s in enumerate(all):
+            print(f' {i+1:2}: find_all_sums({sum(s)}) -> {s}')
+
+
+    lst = [     # input list
+        260, 720, 225, 179, 101, 767, 167, 200, 157, 289,
+        318, 303, 153, 290, 201, 594, 457, 607, 592, 246,
+        132, 135, 584, 432, 591, 204, 417, 405, 362, 658,
+        136, 751, 583, 536, 293, 493, 431, 780, 563, 703,
+        400, 618, 397, 320, 513, 708, 319, 317, 685, 347,
+        758, 439, 145, 378, 158, 384, 551, 110, 408, 648,
+        847, 498,  50,  19,     # 64 numbers
+    ]
+    # Challenge 7, Hard problem finding numbers (extra points)
+    if 7 in run_choices:
+        print(f'\nlist({len(lst)}) with {len(lst)} numbers.\n\\\\')
+        n = 101 + 201 + 167     # 469
+        all = n1.find_all_sums(n, lst)
+        #
+        sort_cpm = lambda x, y: -1 if len(x) < len(y) else 1 if len(x) > len(y) else \
+            -1 if x <= y else 1 if x > y else 0
+        all.sort(key=cmp_to_key(sort_cpm))  # sort by len(solution)
+        #
+        for i, s in enumerate(all):
+            print(f' {i+1:2}: find_all_sums({sum(s)}) -> {s}')
+        print()
+
+
+# #
+# n = 899 # 720 + 179, [[720, 179], [260, 179, 157, 303], [167, 289, 153, 290], [289, 153, 457]]
+# n = 6240
\ No newline at end of file

E_streams/.vscode/settings.json

+{
+    "python.testing.unittestArgs": [
+        "-v",
+        "-s",
+        ".",
+        "-p",
+        "test_*.py"
+    ],
+    "python.testing.pytestEnabled": false,
+    "python.testing.unittestEnabled": true
+}
\ No newline at end of file

E_streams/README.md

+# Assignment E: Data Streams in Python   (12 Pts)
+
+
+### Challenges
+- [Challenge 1:](#1-challenge-data-streams-in-python) Data Streams in Python
+- [Challenge 2:](#2-challenge-map-function) *map()* function
+- [Challenge 3:](#3-challenge-reduce-function) *reduce()* function
+- [Challenge 4:](#4-challenge-sort-function) *sort()* function
+- [Challenge 5:](#5-challenge-pipeline-for-product-codes) Pipeline for Product Codes
+- [Challenge 6:](#6-challenge-run-unit-tests) Run Unit Tests
+- [Challenge 7:](#7-challenge-sign-off) Sign-off
+
+Points: [1, 2, 3, 3, 2, 0, 1]
+
+
+ 
+### 1.) Challenge: Data Streams in Python
+
+Data streams are powerful abstractions for data-driven applications that also work in distributed environments. Big Data platforms often build on streams such as
+[Spark Streams](https://spark.apache.org/docs/latest/streaming-programming-guide.html) or
+[Kafka](https://kafka.apache.org/documentation/streams).
+
+A data stream starts with a *source* (here just a list of names) followed by a pipeline of *chainable operations* performed on each data element passing through the stream. Results can be collected at the *terminus* of the stream.
+
+Pull Python file [stream.py](stream.py).
+
+```py
+class Stream:
+    """
+    Class of a data stream comprised of a sequence of stream operations:
+    """
+
+    class __Stream_op:
+        """
+        Inner class of one stream operation with chainable functions.
+        Instances comprise the stream pipeline.
+        """
+
+        def slice(self, i1, i2=None, i3=1):
+            # function that returns new __Stream_op instance that slices stream
+            if i2 == None:
+                i2, i1 = i1, 0
+            #
+            return self.__new(self.__data[i1:i2:i3])
+
+        def filter(self, filter_func=lambda d : True) ...
+            # return new __Stream_op instance that passes only elements for
+            # which filter_func yields True
+
+        def map(self, map_func=lambda d : d) ...
+            # return new __Stream_op instance that passes elements resulting
+            # from map_func of corresponding elements in the inbound stream
+
+        def reduce(self, reduce_func, start=0) -> any: ... 
+            # terminal function that returns single value compounded by reduce_func
+
+        def sort(self, comperator_func=lambda d1, d2 : True) ...
+            # return new __Stream_op instance that passes stream sorted by
+            # comperator_func
+
+        def cond(self, cond: bool, conditional): ...
+            # return same __Stream_op instance or apply conditional function
+            # on __Stream_op instance if condition yields True
+
+        def print(self) ...
+            # return same, unchanged __Stream_op instance and print as side effect
+
+        def count(self) -> int: ...
+            # terminal function that returns number of elements in terminal stream
+
+        def get(self) -> any: ...
+            # terminal function that returns final stream __data
+```
+
+Application of the stream can demonstrated by the example of a stream of names. The stream is instantiated from the `names` list. The `source()` - method returns the first `__Stream_op` - instance onto which chainable stream methods can be attached.
+
+The stream in the example filters names of lenght = 4, prints those names and counts their number. The *lambda*-expression controls the filter process. Only names of length 4 pass to subsequent pipeline operations.
+
+```py
+names = ['Gonzalez', 'Gill', 'Hardin', 'Richardson', 'Buckner', 'Marquez',
+    'Howe', 'Ray', 'Navarro', 'Talley', 'Bernard', 'Gomez', 'Hamilton',
+    'Case', 'Petty', 'Lott', 'Casey', 'Hall', 'Pena', 'Witt', 'Joyner',
+    'Raymond', 'Crane', 'Hendricks', 'Vance', 'Cleveland', 'Duncan', 'Soto',
+    'Brock', 'Graham', 'Nielsen', 'Rutledge', 'Strong', 'Cox']
+
+result = Stream(names).source() \
+    .filter(lambda n : len(n) == 4) \
+    .print() \
+    .count()
+
+print(f'found {result} names with 4 letters.')
+```
+
+Output:
+```c++
+['Gill', 'Howe', 'Case', 'Lott', 'Hall', 'Pena', 'Witt', 'Soto']
+found 8 names with 4 letters.
+```
+**Questions:**
+
+ - How does method chaining work?
+    - What is required for chainable methods? 
+ - How does a data pipeline gets formed in the example?
+    - Draw a sketch of data objects and how they are linked from the example above.
+
+(1 Pts)
+
+
+ 
+### 2.) Challenge: *map()* function
+
+Complete the `map()` function in [stream.py](stream.py) so that the example produces
+the desired result: Names are mapped to name lengths for the first 8 names.
+Name lengths are then compounded to a single result.
+
+```py
+result = Stream(names).source() \
+    .slice(8) \
+    .print() \
+    .map(lambda n : len(n)) \
+    .print()
+```
+
+Output:
+```c++
+['Gonzalez', 'Gill', 'Hardin', 'Richardson', 'Buckner', 'Marquez', 'Howe', 'Ray']
+[8, 4, 6, 10, 7, 7, 4, 3]
+```
+(2 Pts)
+
+
+ 
+### 3.) Challenge: *reduce()* function
+
+Complete the `reduce()` function in [stream.py](stream.py) so that name lengths are
+compounded (added one after another) to a single result.
+
+```py
+result = Stream(names).source() \
+    .slice(8) \
+    .print() \
+    .map(lambda n : len(n)) \
+    .print() \
+    .reduce(lambda x, y : x + y)
+#
+print(f'compound number of letters in names is: {result}.')
+```
+
+Output:
+```c++
+['Gonzalez', 'Gill', 'Hardin', 'Richardson', 'Buckner', 'Marquez', 'Howe', 'Ray']
+[8, 4, 6, 10, 7, 7, 4, 3]
+compound number of letters in names is: 49.
+```
+(2 Pts)
+
+
+3.1) Test your implementation to also work for the next example that produces
+a single string of all n-letter names:
+
+```py
+n = 5
+result = Stream(names).source() \
+    .filter(lambda name : len(name) == n) \
+    .print() \
+    .map(lambda n : n.upper()) \
+    .reduce(lambda x, y : str(x) + str(y), '')
+#
+print(f'compounded {n}-letter names: {result}.')
+```
+
+Output for n=3 and n=5:
+```c++
+['Ray', 'Cox']
+compounded 3-letter names: RAYCOX.
+
+['Gomez', 'Petty', 'Casey', 'Crane', 'Vance', 'Brock']
+compounded 5-letter names: GOMEZPETTYCASEYCRANEVANCEBROCK.
+```
+(1 Pts)
+
+
+ 
+### 4.) Challenge: *sort()* function
+
+Complete the `sort()` function in [stream.py](stream.py) so that the example produces
+the desired result (use Python's built-in `sort()` or `sorted()` functions).
+
+```py
+Stream(names).source() \
+    .slice(8) \
+    .print('unsorted: ') \
+    .sort() \
+    .print('  sorted: ')
+```
+
+Output:
+```c++
+unsorted: ['Gonzalez', 'Gill', 'Hardin', 'Richardson', 'Buckner', 'Marquez', 'Howe', 'Ray']
+  sorted: ['Buckner', 'Gill', 'Gonzalez', 'Hardin', 'Howe', 'Marquez', 'Ray', 'Richardson']
+```
+(1 Pts)
+
+4.1) Understand the sorted sequence below and define a `comperator` (expression that compares two elements (n1, n2) and yields `-1` if n1 should come before n2, `+1` if n1 must be after n2 or `0` if n1 is equal to n2):
+
+```py
+len_alpha_comperator = lambda ...
+
+Stream(names).source() \
+    .sort(len_alpha_comperator) \
+    .print('sorted: ')
+```
+
+Output:
+```c++
+sorted: ['Cox', 'Ray', 'Case', 'Gill', 'Hall', 'Howe', 'Lott', 'Pena', 'Soto', 'Witt', 'Brock', 'Casey', 'Crane', 'Gomez', 'Petty', 'Vance', 'Duncan', 'Graham', 'Hardin', 'Joyner', 'Strong', 'Talley', 'Bernard', 'Buckner', 'Marquez', 'Navarro', 'Nielsen', 'Raymond', 'Gonzalez', 'Hamilton', 'Rutledge', 'Cleveland', 'Hendricks', 'Richardson']
+```
+(1 Pts)
+
+4.2) Extend the pipeline so that it produces the following output:
+```c++
+sorted: [('Cox', 'Xoc', 3), ('Ray', 'Yar', 3), ('Brock', 'Kcorb', 5), ('Casey', 'Yesac', 5), ('Crane', 'Enarc', 5), ('Gomez', 'Zemog', 5), ('Petty', 'Yttep', 5), ('Vance', 'Ecnav', 5), ('Bernard', 'Dranreb', 7), ('Buckner', 'Renkcub', 7), ('Marquez', 'Zeuqram', 7), ('Navarro', 'Orravan', 7), ('Nielsen', 'Neslein', 7), ('Raymond', 'Dnomyar', 7), ('Cleveland', 'Dnalevelc', 9), ('Hendricks', 'Skcirdneh', 9)]
+\\
+16 odd-length names found.
+```
+(1 Pts)
+
+
+ 
+### 5.) Challenge: Pipeline for Product Codes
+
+Build a pipeline that produces batches of five 6-digit numbers with prefix 'X'.
+Numbers are in ascending order within each batch and end with a 1-digit checksum
+after a dash. The checksum is the sum of all six digits of the random number modulo 10.
+
+```py
+for i in range(1, 5):
+    # Stream of 5 random numbers from integer range, feel free to change
+    codes = Stream([random.randint(100000,999999) for j in range(5)]).source() \
+        ... \
+        .get()
+    #
+    print(f'batch {i}: {codes}')
+```
+
+Output:
+```c++
+batch 1: ['X102042-9', 'X102180-2', 'X103228-6', 'X104680-9', 'X106782-4']
+batch 2: ['X200064-2', 'X200732-4', 'X202090-3', 'X209056-2', 'X211464-8']
+batch 3: ['X300186-8', 'X301416-5', 'X305962-5', 'X307938-0', 'X312524-7']
+batch 4: ['X400216-3', 'X401436-8', 'X401682-1', 'X405256-2', 'X406376-6']
+```
+(1 Pts)
+
+5.1) Alter the pipeline such that it produces only even digit codes:
+```c++
+batch 1: ['X226840-2', 'X284240-0', 'X448288-4', 'X804080-0', 'X888620-2']
+batch 2: ['X220640-4', 'X248066-6', 'X648466-4', 'X680404-2', 'X882868-0']
+batch 3: ['X262626-4', 'X608662-8', 'X626404-2', 'X662424-4', 'X846228-0']
+batch 4: ['X224200-0', 'X282204-8', 'X448426-8', 'X600282-8', 'X802882-8']
+```
+(1 Pts)
+
+
+ 
+### 6.) Challenge: Run Unit Tests
+
+Pull file
+[test_stream.py](test_stream.py)
+into same directory. Run unit tests to confirm the correctness of your solution.
+```sh
+cd E_streams                        # change to directory where stream.py and test_strean.py are
+test_url=https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/raw/main/E_streams/test_stream.py
+curl -O $(echo $test_url)           # download file with unit tests from URL
+python test_stream.py               # run tests from test file
+python -m unittest --verbose        # run unit tests with discovery
+```
+
+Output:
+
+```sh
+Ran 12 tests in 0.001s
+
+OK
+Unit testing using test objects:
+ - test_filter_1()
+ - test_filter_11()
+ - test_filter_12()
+ - test_filter_13()
+ - test_map_2()
+ - test_map_21()
+ - test_reduce_3()
+ - test_reduce_31()
+ - test_sort_4()
+ - test_sort_41()
+ - test_sort_42()
+ - test_stream_generation()
+---> 12/12 TESTS SUCCEEDED
+```
+
+
+ 
+### 7.) Challenge: Sign-off
+
+For sign-off, change into `E_streams` directory and copy commands into a terminal:
+
+```sh
+test_url=https://gitlab.bht-berlin.de/sgraupner/ds_cs4bd_2324/-/raw/main/E_streams/test_stream.py
+
+curl $test_url | python     # run tests from URL (use for sign-off)
+```
+
+Result:
+
+```
+  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
+                                 Dload  Upload   Total   Spent    Left  Speed
+100  6264  100  6264    0     0  38354      0 --:--:-- --:--:-- --:--:-- 38666
+Unit testing using test objects:
+<stdin>:153: DeprecationWarning: unittest.makeSuite() is deprecated and will be
+removed in Python 3.13. Please use unittest.TestLoader.loadTestsFromTestCase() i
+nstead.
+----------------------------------------------------------------------
+Ran 12 tests in 0.001s
+
+OK
+```
+
+12 tests succeeded.
+
+(1 Pts)

E_streams/__init__.py

+"""
+Special file __init__.py marks a directory as a Python module.
+The file is executed once when any .py file is imported.
+//
+Python unittest require the presence of (even an empty) file.
+"""
+
+# load setup module when executed in parent directory
+try:
+    __import__('setup')
+#
+except ImportError:
+    pass

E_streams/stream.py

+import random
+
+"""
+Class of a data stream comprised of a sequence of stream operations:
+    - slice(i1, i2, i3)    # slice stream in analogy to python slicing
+    - filter(filter_func)  # pass only elements for which filter_func yields True
+    - map(map_func)        # pass stream where each element is mapped by map_func
+    - sort(comperator_func) # pass stream sorted by comperator_func
+    - cond(cond, cond_func) # pass stream or apply conditional function
+    - print()              # pass unchanged stream and print as side effect
+
+and with terminal functions:
+    - reduce(reduce_func, start)   # compound stream to single value with reduce_func
+    - count()              # return number of elements in terminal stream
+    - get()                # return final stream data
+"""
+class Stream:
+
+    def __init__(self, _data=[]):
+        # constructor to initialize instance member variables
+        #
+        self.__streamSource = self.__new_op(_data)
+
+
+    class __Stream_op:
+        """
+        Inner class of one stream operation with chainable functions.
+        Instances comprise the stream pipeline.
+        """
+        def __init__(self, _new_op_func, _data):
+            self.__data = _data
+            self.__new = _new_op_func    # __new_op() function injected from outer context
+
+
+        def slice(self, i1, i2=None, i3=1):
+            # function that returns new __Stream_op instance that slices stream
+            if i2 == None:
+                # flip i1, i2 for single arg, e.g. slice(0, 8), slice(8)
+                i2, i1 = i1, 0
+            #
+            # return new __Stream_op instance with sliced __data
+            return self.__new(self.__data[i1:i2:i3])
+
+
+        def filter(self, filter_func=lambda d : True):
+            # return new __Stream_op instance that passes only elements for
+            # which filter_func yields True
+            #
+            return self.__new([d for d in self.__data if filter_func(d)])
+
+
+        def map(self, map_func=lambda d : d):
+            # return new __Stream_op instance that passes elements resulting
+            # from map_func of corresponding elements in the inbound stream
+            #
+            # input data is list of current instance: self.__data
+            # mapping means a new list needs to be created with same number of
+            # elements, each obtained by applying map_func
+
+            # create new data for next __Stream_op instance from current instance
+            # data: self.__data
+            new_data = self.__data      # <-- compute new data here
+
+            # create new __Stream_op instance with new stream data
+            new_stream_op_instance = self.__new(new_data)
+            return new_stream_op_instance
+
+
+        def reduce(self, reduce_func=lambda compound, d : compound + d, start=0) -> any:
+            # terminal function that returns single value compounded by reduce_func
+            #
+            compound = 0                # <-- compute compound result here
+
+            return compound
+
+
+        def sort(self, comperator_func=lambda n1, n2 : -1 if n1 < n2 else 1):
+            # return new __Stream_op instance that passes stream sorted by
+            # comperator_func
+            #
+            # create new data for next __Stream_op instance from current instance
+            # data: self.__data
+            new_data = self.__data      # <-- compute new data here
+
+            # create new __Stream_op instance with new stream data
+            new_stream_op_instance = self.__new(new_data)
+            return new_stream_op_instance
+
+
+        def cond(self, cond: bool, conditional):
+            # return same __Stream_op instance or apply conditional function
+            # on __Stream_op instance if condition yields True
+            #
+            return conditional(self) if cond else self
+
+
+        def print(self, prefix=''):
+            # return same, unchanged __Stream_op instance and print as side effect
+            #
+            print(f'{prefix}{self.__data}')
+            return self
+
+
+        def count(self) -> int:
+            # terminal function that returns number of elements in terminal stream
+            #
+            return len(self.__data)
+
+
+        def get(self) -> any:
+            # terminal function that returns final stream __data
+            #
+            return self.__data
+
+
+    def source(self):
+        # return first __Stream_op instance of stream as source
+        #
+        return self.__streamSource
+
+
+    def __new_op(self, *argv):
+        # private method to create new __Stream_op instance
+        return Stream.__Stream_op(self.__new_op, *argv)
+
+
+# attempt to load solution module (ignore)
+try:
+    _from, _import = 'stream_sol', 'Stream'
+    # fetch Stream class from solution, if present
+    Stream = getattr(__import__(_from, fromlist=[_import]), _import)
+#
+except ImportError:
+    pass
+
+
+if __name__ == '__main__':
+
+    run_choice = 3
+    #
+    run_choices = {
+        1:  "Challenge 1, Data streams in Python, run the first example",
+        2:  "Challenge 2, complete map() function",
+        3:  "Challenge 3, complete reduce() function",
+        31: "Challenge 3.1, example RAYCOX",
+        4:  "Challenge 4, complete sort() function",
+        41: "Challenge 4.1, len-alpha comperator",
+        42: "Challenge 4.2, tuple output: ('Cox', 'Xoc', 3)",
+        5:  "Challenge 5, Pipeline for product codes",
+        51: "Challenge 5.1, even digit codes"
+    }
+
+    names = ['Gonzalez', 'Gill', 'Hardin', 'Richardson', 'Buckner', 'Marquez',
+        'Howe', 'Ray', 'Navarro', 'Talley', 'Bernard', 'Gomez', 'Hamilton',
+        'Case', 'Petty', 'Lott', 'Casey', 'Hall', 'Pena', 'Witt', 'Joyner',
+        'Raymond', 'Crane', 'Hendricks', 'Vance', 'Cleveland', 'Duncan', 'Soto',
+        'Brock', 'Graham', 'Nielsen', 'Rutledge', 'Strong', 'Cox']
+
+
+    if run_choice == 1:
+        # Challenge 1, Data streams in Python, run the first example
+        result = Stream(names).source() \
+            .filter(lambda n : len(n) == 4) \
+            .print() \
+            .count()
+        #
+        print(f'found {result} names with 4 letters.')
+
+    if run_choice == 2:
+        # Challenge 2, complete map() function
+        # to map names to name lengths for the first 8 names
+        Stream(names).source() \
+            .slice(8) \
+            .print() \
+            .map(lambda n : len(n)) \
+            .print()
+
+    if run_choice == 3:
+        # Challenge 3, complete reduce() function
+        # to compound all name lengths to a single result
+        result = Stream(names).source() \
+            .slice(8) \
+            .print() \
+            .map(lambda n : len(n)) \
+            .print() \
+            .reduce(lambda x, y : x + y)
+        #
+        print(f'compound number of letters in names is: {result}.')
+
+    if run_choice == 31:
+        # Challenge 3.1, example RAYCOX
+        # compound single string of all n-letter names
+        n = 3
+        result = Stream(names).source() \
+            .filter(lambda name : len(name) == n) \
+            .print() \
+            .map(lambda n : n.upper()) \
+            .reduce(lambda x, y : str(x) + str(y), '')
+        #
+        print(f'compounded {n}-letter names: {result}.')
+
+    if run_choice == 4:
+        # Challenge 4, complete sort() function
+        Stream(names).source() \
+            .slice(8) \
+            .print('unsorted: ') \
+            .sort() \
+            .print('  sorted: ')
+
+    alpha_comperator = lambda n1, n2 : -1 if n1 < n2 else 1
+    len_alpha_comperator = lambda n1, n2 : -1 if len(n1) < len(n2) else 1 if len(n1) > len(n2) else alpha_comperator(n1, n2)
+    #
+    if run_choice == 41:
+        # Challenge 4.1, len-alpha comperator
+        Stream(names).source() \
+            .sort(len_alpha_comperator) \
+            .print('sorted: ')
+
+    if run_choice == 42:
+        # Challenge 4.2, tuple output: ('Cox', 'Xoc', 3)
+        result = Stream(names).source() \
+            .sort(len_alpha_comperator) \
+            .map(lambda n : (n, n[::-1].capitalize(), len(n))) \
+            .filter(lambda n1 : n1[2] % 2 == 1) \
+            .print('sorted: ') \
+            .count()
+        #
+        print(f'\\\\\n{result} odd-length names found.')
+
+    # rand_numbers = [random.randint(100000,999999) for i in range(30)]
+    # print(f'random numbers: {rand_numbers}')
+    #
+    if run_choice == 5 or run_choice == 51:
+        # Challenge 5, Pipeline for product codes
+        # Challenge 5.1, even digit codes
+        #
+        for i in range(1, 5):
+            # Stream of 5 random numbers from integer range, feel free to change
+            codes = Stream([random.randint(100000,999999) for j in range(1000)]).source() \
+                .filter(lambda n : n % 2 == 0) \
+                .cond( run_choice == 51, \
+                    # use only numbers with even digits, test by split up number in sequence of digits
+                    lambda op : op.filter(lambda n : len(set(map(int, str(n))).intersection([1, 3, 5, 7, 9])) == 0) \
+                ) \
+                .slice(5) \
+                .sort() \
+                .map(lambda n : f'X{n}-{sum(list(map(int, str(n)))) % 10}') \
+                .get()
+            #
+            print(f'batch {i}: {codes}')
+

E_streams/test_stream.py

+import unittest
+from stream import Stream
+
+
+class Stream_test(unittest.TestCase):
+    """
+    Test class.
+    """
+    list_1 = [4, 12, 3, 8, 17, 12, 1, 8, 7]
+    list_1_str = [str(d) for d in list_1]
+    names = ['Gonzalez', 'Gill', 'Hardin', 'Richardson', 'Buckner', 'Marquez',
+        'Howe', 'Ray', 'Navarro', 'Talley', 'Bernard', 'Gomez', 'Hamilton',
+        'Case', 'Petty', 'Lott', 'Casey', 'Hall', 'Pena', 'Witt', 'Joyner',
+        'Raymond', 'Crane', 'Hendricks', 'Vance', 'Cleveland', 'Duncan', 'Soto',
+        'Brock', 'Graham', 'Nielsen', 'Rutledge', 'Strong', 'Cox']
+
+    # tests for stream generation function
+    def test_stream_generation(self):
+        #
+        result = Stream(self.list_1).source() \
+            .get()
+        self.assertEqual(self.list_1, result)
+
+
+    # tests for filter() function
+    def test_filter_1(self):
+        #
+        # test Challenge 1
+        result = Stream(self.list_1).source() \
+            .filter(lambda n : n % 2 == 1) \
+            .get()
+        self.assertEqual([3, 17, 1, 7], result)
+
+    def test_filter_11(self):
+        result = Stream(self.list_1).source() \
+            .filter(lambda d : False) \
+            .get()
+        self.assertEqual([], result)
+
+    def test_filter_12(self):
+        result = Stream(self.list_1).source() \
+            .filter(lambda d : True) \
+            .get()
+        self.assertEqual(self.list_1, result)
+
+    def test_filter_13(self):
+        result = Stream(self.names).source() \
+            .filter(lambda n : len(n) == 4) \
+            .get()
+        self.assertEqual(['Gill', 'Howe', 'Case', 'Lott', 'Hall', 'Pena', 'Witt', 'Soto'], result)
+
+
+    # tests for map() function
+    def test_map_2(self):
+        #
+        # test Challenge 2
+        result = Stream(self.names).source() \
+            .slice(8) \
+            .map(lambda n : len(n)) \
+            .get()
+        self.assertEqual([8, 4, 6, 10, 7, 7, 4, 3], result)
+
+    def test_map_21(self):
+        result = Stream(self.names).source() \
+            .filter(lambda n : len(n) == 3) \
+            .map(lambda n : (n, len(n))) \
+            .get()
+        self.assertEqual([('Ray', 3), ('Cox', 3)], result)
+
+
+    # tests for reduce() function
+    def test_reduce_3(self):
+        #
+        # test Challenge 3
+        result = Stream(self.names).source() \
+            .slice(8) \
+            .map(lambda n : len(n)) \
+            .reduce(lambda x, y : x + y)
+        self.assertEqual(49, result)
+
+    def test_reduce_31(self):
+        # test Challenge 3.1
+        n = 3
+        result = Stream(self.names).source() \
+            .filter(lambda name : len(name) == n) \
+            .map(lambda n : n.upper()) \
+            .reduce(lambda x, y : str(x) + str(y), '')
+        self.assertEqual('RAYCOX', result)
+        #
+        n = 5
+        result = Stream(self.names).source() \
+            .filter(lambda name : len(name) == n) \
+            .map(lambda n : n.upper()) \
+            .reduce(lambda x, y : str(x) + str(y), '')
+        self.assertEqual('GOMEZPETTYCASEYCRANEVANCEBROCK', result)
+
+
+    # tests for sort() function
+    def test_sort_4(self):
+        # test Challenge 4
+        result = Stream(self.names).source() \
+            .slice(8) \
+            .sort() \
+            .get()
+        expected = ['Buckner', 'Gill', 'Gonzalez', 'Hardin', 'Howe', 'Marquez', 'Ray', 'Richardson']
+        self.assertEqual(expected, result)
+
+    def alpha_comperator(self, n1, n2):
+        return -1 if n1 < n2 else 1
+
+    def len_alpha_comperator(self, n1, n2):
+        return -1 if len(n1) < len(n2) else 1 if len(n1) > len(n2) else self.alpha_comperator(n1, n2)
+
+    def test_sort_41(self):
+        # test Challenge 4.1
+        result = Stream(self.names).source() \
+            .sort(self.len_alpha_comperator) \
+            .get()
+        #
+        expected = ['Cox', 'Ray', 'Case', 'Gill', 'Hall', 'Howe', 'Lott', 'Pena', 'Soto', 'Witt',
+            'Brock', 'Casey', 'Crane', 'Gomez', 'Petty', 'Vance', 'Duncan', 'Graham', 'Hardin',
+            'Joyner', 'Strong', 'Talley', 'Bernard', 'Buckner', 'Marquez', 'Navarro', 'Nielsen',
+            'Raymond', 'Gonzalez', 'Hamilton', 'Rutledge', 'Cleveland', 'Hendricks', 'Richardson'
+        ]
+        self.assertEqual(expected, result)
+
+    def test_sort_42(self):
+        # test Challenge 4.2
+        result = Stream(self.names).source() \
+            .sort(self.len_alpha_comperator) \
+            .map(lambda n : (n, n[::-1].capitalize(), len(n))) \
+            .filter(lambda n1 : n1[2] % 2 == 1) \
+            .get()
+        #
+        expected = [('Cox', 'Xoc', 3), ('Ray', 'Yar', 3), ('Brock', 'Kcorb', 5), ('Casey', 'Yesac', 5),
+            ('Crane', 'Enarc', 5), ('Gomez', 'Zemog', 5), ('Petty', 'Yttep', 5), ('Vance', 'Ecnav', 5),
+            ('Bernard', 'Dranreb', 7), ('Buckner', 'Renkcub', 7), ('Marquez', 'Zeuqram', 7),
+            ('Navarro', 'Orravan', 7), ('Nielsen', 'Neslein', 7), ('Raymond', 'Dnomyar', 7),
+            ('Cleveland', 'Dnalevelc', 9), ('Hendricks', 'Skcirdneh', 9)
+        ]
+        self.assertEqual(expected, result)
+        #
+        result = Stream(self.names).source() \
+            .sort(self.len_alpha_comperator) \
+            .map(lambda n : (n, n[::-1].capitalize(), len(n))) \
+            .filter(lambda n1 : n1[2] % 2 == 1) \
+            .count()
+        self.assertEqual(16, result)
+
+
+    # report results,
+    # see https://stackoverflow.com/questions/28500267/python-unittest-count-tests
+    # currentResult = None
+
+    # @classmethod
+    # def setResult(cls, amount, errors, failures, skipped):
+    #     cls.amount, cls.errors, cls.failures, cls.skipped = \
+    #         amount, errors, failures, skipped
+
+    # def tearDown(self):
+    #     amount = self.currentResult.testsRun
+    #     errors = self.currentResult.errors
+    #     failures = self.currentResult.failures
+    #     skipped = self.currentResult.skipped
+    #     self.setResult(amount, errors, failures, skipped)
+
+    # @classmethod
+    # def tearDownClass(cls):
+    #     print("\ntests run: " + str(cls.amount))
+    #     print("errors: " + str(len(cls.errors)))
+    #     print("failures: " + str(len(cls.failures)))
+    #     print("success: " + str(cls.amount - len(cls.errors) - len(cls.failures)))
+    #     print("skipped: " + str(len(cls.skipped)))
+
+    # def run(self, result=None):
+    #     self.currentResult = result # remember result for use in tearDown
+    #     unittest.TestCase.run(self, result) # call superclass run method
+
+
+if __name__ == '__main__':
+    result = unittest.main()
+