Gc/pytest utils (#11)

* add test for parse_inputs

* install python package (dev)

* typo

* 🎨 Format Python code with psf/black (#10)

Co-authored-by: gcattan <gcattan@users.noreply.github.com>

* test sample size

* typo

* fix test sample size

* 🎨 Format Python code with psf/black (#12)

Co-authored-by: gcattan <gcattan@users.noreply.github.com>

* add test on sample size

* fix assert for test_sample_greater_size_than_list

* fix test_sample_greater_size_than_list

* 🎨 Format Python code with psf/black (#13)

Co-authored-by: gcattan <gcattan@users.noreply.github.com>

* test skeleton for test_git

* 🎨 Format Python code with psf/black (#14)

Co-authored-by: gcattan <gcattan@users.noreply.github.com>

* finish test common and start test for git

* import datetime missing

* typeerror: Integer required

* return absolute month difference

* typo

* 🎨 Format Python code with psf/black (#17)

Co-authored-by: gcattan <gcattan@users.noreply.github.com>

* test if code runs wihtin a py.test session

* typo

* print traces

* test_branch_is_old: GREEN

* test_branch_are_coupled: GREEN

* change are_coupled -> contains

* test_git pass

* 🎨 Format Python code with psf/black (#18)

Co-authored-by: gcattan <gcattan@users.noreply.github.com>

* test for is_empty_body

* typo

* test is_test_commit: GREEN

* test for not_a_squased commit: GREEN

* test if commit contains bad words: GREEN

* test count coupled : GREEN!

* test count_old_branches : GREEN

* test process logs: GREEN

* tes overall OK

* some helpfull logs

* 🎨 Format Python code with psf/black (#19)

Co-authored-by: gcattan <gcattan@users.noreply.github.com>

* Update run-pytest.yml

* missing "-r" option in "is_old"

* Update run-pytest.yml

Co-authored-by: gcattan <gregoire.cattan@ibm.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: gcattan <gcattan@users.noreply.github.com>

.github/workflows/run-pytest.yml

@@ -9,15 +9,19 @@ jobs:
       max-parallel: 5
     steps:
     - uses: actions/checkout@v2
-    - name: Set up Python 3.10
+    - name: Set up Python 3.9
       uses: actions/setup-python@v2
       with:
-        python-version: "3.10"
+        python-version: "3.9"
     - name: Add conda to system path
       run: |
         # $CONDA is an environment variable pointing to the root of the miniconda directory
         echo $CONDA/bin >> $GITHUB_PATH
+    - name: Install git repo
+      run: |
+        python setup.py develop
     - name: Test with pytest
       run: |
         conda install pytest
+        git fetch
         pytest

example/git-quality-check.py

@@ -16,7 +16,7 @@
 from git_quality_check.indicators.commits import (
     is_empty_body,
     not_a_squashed_commit,
-    count_bad_words,
+    does_contain_bad_words,
     is_test_commit,
 )
 
@@ -31,17 +31,17 @@
     branches = git_all_branches()
 
     bad_commit_index = process_logs(
-        logs, [not_a_squashed_commit, is_empty_body, count_bad_words(bad_words)]
+        logs, [not_a_squashed_commit, is_empty_body, does_contain_bad_words(bad_words)]
     )
     test_index = process_logs(logs, [is_test_commit])
 
     old_branches_index = count_old_branches(branches)
     coupling_index = count_coupled(branches, main_branches)
 
-    print(bad_commit_index)
-    print(test_index)
-    print(old_branches_index)
-    print(coupling_index)
+    print("Percent of bad commits: ", bad_commit_index)
+    print("Percent of test commits: ", test_index)
+    print("Percent of old branches: ", old_branches_index)
+    print("Percent of coupled branches", coupling_index)
 
     overall = compute_score(
         bad_commit_index, test_index, old_branches_index, coupling_index

git_quality_check/indicators/commits/__init__.py

@@ -1,4 +1,4 @@
 from .is_empty_body import is_empty_body
 from .not_a_squashed_commit import not_a_squashed_commit
-from .count_bad_words import count_bad_words
+from .does_contain_bad_words import does_contain_bad_words
 from .is_test_commit import is_test_commit

git_quality_check/indicators/commits/does_contain_bad_words.py

@@ -0,0 +1,9 @@
+def does_contain_bad_words(bad_words: list[str]):
+    def _does_contain_bad_words(log: str):
+        counter = 0
+        for word in bad_words:
+            if word.lower() in log.lower().split():
+                counter += 1
+        return 1 if counter > 0 else 0
+
+    return _does_contain_bad_words

git_quality_check/indicators/counters/count_coupled.py

@@ -1,12 +1,16 @@
-from git_quality_check.utils import sample, are_coupled
+from git_quality_check.utils import sample, contains
 
 
 def count_coupled(branches, main_branches):
     branches, count = sample(branches, 10)
     branches.extend(main_branches)
     count += len(main_branches)
     counter = 0
+    max_counter = 0
     for bA in branches:
         for bB in branches:
-            counter += 1 if are_coupled(bA, bB) else 0
-    return counter / count * 100
+            if not bA == bB:
+                print(bA, bB, contains(bA, bB))
+                counter += 1 if contains(bA, bB) else 0
+                max_counter += 1
+    return counter / max_counter * 100

git_quality_check/indicators/counters/process_logs.py

@@ -4,5 +4,6 @@ def process_logs(logs: list[str], functions: list[str:int]):
     for i in range(len(logs)):
         log = logs[i]
         for function in functions:
+            # a same commit can be marked twice if it failes two functions
             counter += function(log)
     return counter / count * 100

git_quality_check/utils/__init__.py

@@ -2,9 +2,16 @@
     is_valid_log,
     remove_header,
     is_old,
-    are_coupled,
+    contains,
     git_logs,
     git_all_branches,
 )
 
-from .common import sample, set_output, format_number, parse_inputs
+from .common import (
+    sample,
+    set_output,
+    format_number,
+    parse_inputs,
+    BADWORDS,
+    MAINBRANCHES,
+)

git_quality_check/utils/common.py

@@ -1,17 +1,26 @@
 import random
 import os
-from datetime import datetime
+from datetime import date, datetime
+
+
+BADWORDS = "INPUT_BADWORDS"
+MAINBRANCHES = "INPUT_MAINBRANCHES"
 
 
 def parse_inputs():
+    """
+    " Gets inputs from environment variables:
+    " - bad_words: A list of words that should be avoided in commit messages.
+    " - main_branches: The list of main branches (e.g. master, main or develop)
+    """
     bad_words = []
     main_branches = []
     try:
-        bad_words = os.environ["INPUT_BADWORDS"].split(", ")
+        bad_words = os.environ[BADWORDS].split(", ")
     except:
         bad_words = ["WIP", "work in progress", "in progress", "TODO"]
     try:
-        main_branches = os.environ["INPUT_MAINBRANCHES"].split(", ")
+        main_branches = os.environ[MAINBRANCHES].split(", ")
     except:
         main_branches = ["origin/develop", "origin/master"]
     return bad_words, main_branches
@@ -25,11 +34,22 @@ def get_date():
     return datetime.today()
 
 
-def sample(li: list[str], min: int):
+def sample(li: list[str], n: int):
+    """
+    " Computes a sample of size 'n' from a list:
+    " - if the size of the list is equal or lower than n, returns the list
+    " - if the size of the list is greater than n, returns a sample of n.
+    "   (some elements might be repetead)
+    "
+    " li: the list
+    " n: expected number of elements in the sample
+    "
+    " Returns the tuple (sample, )
+    """
     count = len(li)
-    if count > min:
-        li = random.sample(li, min)
-        count = min
+    if count > n:
+        li = random.sample(li, n)
+        count = n
     return (li, count)
 
 
@@ -41,5 +61,5 @@ def set_output(output: str):
     print(f"::set-output name=score::{output}")
 
 
-def diff_month(d1, d2):
-    return (d1.year - d2.year) * 12 + d1.month - d2.month
+def diff_month(d1: date, d2: date):
+    return abs((d1.year - d2.year) * 12 + d1.month - d2.month)

git_quality_check/utils/git.py

@@ -1,4 +1,5 @@
 import subprocess
+import sys
 from datetime import datetime
 from .common import get_date, diff_month, strip
 
@@ -30,7 +31,7 @@ def git_all_branches():
     return [strip(r) for r in ret if not strip(r) == ""]
 
 
-def are_coupled(branchA: str, branchB: str):
+def contains(branchA: str, branchB: str):
     if not is_well_formed_branch(branchA) or not is_well_formed_branch(branchB):
         return False
     if branchA == branchB:
@@ -53,7 +54,7 @@ def is_well_formed_branch(branch: str):
 def git_get_branch_date(branch: str):
     if not is_well_formed_branch(branch):
         return None
-    ret = run_git(["log", "-n", "1", '--date=format:"%Y-%m-%d"', branch]).split(
+    ret = run_git(["log", "-n", "1", '--date=format:"%Y-%m-%d"', branch, "-r"]).split(
         "Date: "
     )[1]
     ret = ret.replace('"', "").split("-")
@@ -64,7 +65,7 @@ def git_get_branch_date(branch: str):
 
 
 def remove_first_line(log: str):
-    if is_valid_log:
+    if is_valid_log(log):
         try:
             eol = log.index("\n")
             log = log[eol + 1 :]

tests/indicators/commits/test_does_contain_bad_words.py

@@ -0,0 +1,10 @@
+from git_quality_check.indicators.commits import does_contain_bad_words
+
+
+def test():
+    bad_words = ["bad", "word"]
+    counter = does_contain_bad_words(bad_words)
+    log = "I am a very bad bad word"
+    assert counter(log) == 1
+    log = "I am a good log"
+    assert counter(log) == 0

tests/indicators/commits/test_is_empty_body.py

@@ -1,5 +1,15 @@
 import pytest
+from git_quality_check.indicators.commits import is_empty_body
 
 
-def test():
-    assert True == False
+@pytest.mark.parametrize("log", ["", "commit 1234455"])
+def test_is_empty_body(log):
+    assert is_empty_body(log) == 1
+
+
+def test_is_not_empty_body():
+    log = """commit 7a1e2a6a76e6967bde14e95900996ca17811de47 (origin/gc/pytest)
+    Author: gcattan <gregoire.cattan@ibm.com>
+    Date:   2022-04-17
+    remove dependencies"""
+    assert is_empty_body(log) == 0

tests/indicators/commits/test_is_test_commit.py

@@ -1,5 +1,12 @@
 import pytest
+from git_quality_check.indicators.commits import is_test_commit
 
 
-def test():
-    assert True == False
+@pytest.mark.parametrize("log", ["test", "testing", "test it", "I am testing it"])
+def test_is_test_commit(log):
+    assert is_test_commit(log) == 1
+
+
+def test_is_not_test_commit():
+    log = """Tset"""
+    assert is_test_commit(log) == 0

tests/indicators/commits/test_not_a_squashed_commit.py

@@ -1,5 +1,6 @@
-import pytest
+from git_quality_check.indicators.commits import not_a_squashed_commit
 
 
 def test():
-    assert True == False
+    assert not_a_squashed_commit("PR(#7)") == False
+    assert not_a_squashed_commit("I am not squased") == True

tests/indicators/counters/test_count_coupled.py

@@ -1,5 +1,6 @@
-import pytest
+from git_quality_check.indicators.counters import count_coupled
 
 
 def test():
-    assert True == False
+    assert count_coupled(["origin/update-readme"], ["origin/master"]) == 50
+    assert count_coupled(["origin/test-action"], ["origin/formatter"]) == 0

tests/indicators/counters/test_count_old_branches.py

@@ -1,5 +1,5 @@
-import pytest
+from git_quality_check.indicators.counters import count_old_branches
 
 
 def test():
-    assert True == False
+    assert count_old_branches(["origin/gc/pytest_utils", "origin/update-readme"]) == 50

tests/indicators/counters/test_process_logs.py

@@ -1,5 +1,18 @@
-import pytest
+from git_quality_check.indicators.counters import process_logs
+from git_quality_check.indicators.commits import not_a_squashed_commit, is_empty_body
 
 
 def test():
-    assert True == False
+    logs = [
+        """Commit 1234
+    Author: 
+    Date:
+    Test""",
+        """Commit 1234
+    Author: 
+    Date:
+    Merge (#5)""",
+        "",
+    ]
+    marked_commits = process_logs(logs, [not_a_squashed_commit, is_empty_body])
+    assert marked_commits == 50

tests/scoring/test_overall.py

@@ -1,5 +1,6 @@
-import pytest
+from git_quality_check.scoring import compute_score
 
 
 def test():
-    assert True == False
+    assert compute_score(100, 0, 100, 100) == 0
+    assert compute_score(0, 100, 0, 0) == 100

tests/utils/test_common.py

@@ -1,5 +1,53 @@
-import pytest
+import os
+from git_quality_check.utils.common import (
+    BADWORDS,
+    MAINBRANCHES,
+    parse_inputs,
+    sample,
+    diff_month,
+)
+from datetime import datetime
 
 
-def test():
-    assert True == False
+def test_parse_inputs():
+    os.environ[BADWORDS] = "WIP, todo"
+    os.environ[MAINBRANCHES] = "origin/develop, origin/main"
+    bad_words, main_branches = parse_inputs()
+    assert bad_words[0] == "WIP"
+    assert bad_words[1] == "todo"
+    assert main_branches[0] == "origin/develop"
+    assert main_branches[1] == "origin/main"
+
+
+def test_sample_same_size_than_list():
+    li = [3, 10, 6, 15, 20]
+    size = len(li)
+    sample_list, count = sample(li, size)
+    assert count == size
+    for i in range(size):
+        assert li[i] == sample_list[i]
+
+
+def test_sample_greater_size_than_list():
+    li = [3, 10, 6, 15, 20]
+    size = len(li)
+    sample_list, count = sample(li, size + 1)
+    assert count == size
+    for i in range(size):
+        assert li[i] == sample_list[i]
+
+
+def test_sample_lower_size_than_list():
+    li = [3, 10, 6, 15, 20]
+    size = len(li) - 1
+    sample_list, count = sample(li, size)
+    assert count == size
+    for i in range(size):
+        assert sample_list[i] in li
+
+
+def test_diff_months():
+    date1 = datetime(2022, 8, 1)
+    date2 = datetime(2023, 8, 1)
+    assert diff_month(date1, date2) == 12
+    assert diff_month(date2, date1) == 12