compounding-intelligence/tests/test_diff_analyzer.py

"""Tests for diff_analyzer module."""

import sys
import os
sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..', 'scripts'))

from diff_analyzer import DiffAnalyzer, ChangeCategory


def test_parse_simple_addition():
    diff = """diff --git a/foo.py b/foo.py
new file mode 100644
--- /dev/null
+++ b/foo.py
@@ -0,0 +1,3 @@
+def hello():
+    return "world"
+# end
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)

    assert summary.files_changed == 1
    assert summary.files_added == 1
    assert summary.files_modified == 0
    assert summary.total_added == 3
    assert summary.total_deleted == 0
    assert summary.hunks_added == 1
    assert len(summary.file_changes) == 1
    assert summary.file_changes[0].is_new is True
    assert summary.file_changes[0].path == "foo.py"


def test_parse_simple_deletion():
    diff = """diff --git a/old.py b/old.py
deleted file mode 100644
--- a/old.py
+++ /dev/null
@@ -1,2 +0,0 @@
-x = 1
-y = 2
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)

    assert summary.files_changed == 1
    assert summary.files_deleted == 1
    assert summary.total_deleted == 2
    assert summary.total_added == 0
    assert summary.hunks_deleted == 1
    assert summary.file_changes[0].is_deleted is True


def test_parse_modification():
    diff = """diff --git a/bar.py b/bar.py
--- a/bar.py
+++ b/bar.py
@@ -10,3 +10,4 @@ def foo():
     existing()
-    old_call()
+    new_call()
+    extra_step()
     return
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)

    assert summary.files_changed == 1
    assert summary.files_modified == 1
    assert summary.total_added == 2   # +new_call(), +extra_step()
    assert summary.total_deleted == 1  # -old_call()
    assert summary.total_context == 2  # 2 context lines
    assert summary.hunks_modified == 1


def test_parse_multiple_files():
    diff = """diff --git a/a.py b/a.py
--- a/a.py
+++ b/a.py
@@ -1,1 +1,2 @@
 existing
+added
diff --git a/b.py b/b.py
new file mode 100644
--- /dev/null
+++ b/b.py
@@ -0,0 +1,1 @@
+new file
diff --git a/c.py b/c.py
deleted file mode 100644
--- a/c.py
+++ /dev/null
@@ -1,1 +0,0 @@
-gone
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)

    assert summary.files_changed == 3
    assert summary.files_added == 1
    assert summary.files_deleted == 1
    assert summary.files_modified == 1
    assert summary.total_added == 2
    assert summary.total_deleted == 1


def test_parse_rename():
    diff = """diff --git a/old_name.py b/new_name.py
rename from old_name.py
rename to new_name.py
--- a/old_name.py
+++ b/new_name.py
@@ -1,1 +1,1 @@
-old_func()
+new_func()
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)

    assert summary.files_changed == 1
    assert summary.files_renamed == 1
    assert summary.file_changes[0].is_renamed is True
    assert summary.file_changes[0].old_path == "old_name.py"
    assert summary.file_changes[0].path == "new_name.py"


def test_parse_mixed_hunks():
    """A file with one add hunk and one delete hunk."""
    diff = """diff --git a/mixed.py b/mixed.py
--- a/mixed.py
+++ b/mixed.py
@@ -5,0 +6,2 @@
+new_line_1
+new_line_2
@@ -20,2 +22,0 @@
-removed_1
-removed_2
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)

    assert summary.files_changed == 1
    assert summary.hunks_added == 1
    assert summary.hunks_deleted == 1
    assert summary.total_added == 2
    assert summary.total_deleted == 2


def test_empty_diff():
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze("")

    assert summary.files_changed == 0
    assert summary.total_added == 0
    assert summary.total_deleted == 0


def test_to_dict():
    diff = """diff --git a/test.py b/test.py
new file mode 100644
--- /dev/null
+++ b/test.py
@@ -0,0 +1,2 @@
+line1
+line2
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)
    d = summary.to_dict()

    assert d["files_changed"] == 1
    assert d["files_added"] == 1
    assert d["total_added"] == 2
    assert d["total_deleted"] == 0
    assert len(d["files"]) == 1
    assert d["files"][0]["path"] == "test.py"
    assert d["files"][0]["is_new"] is True


def test_context_only_hunk():
    """A hunk with only context lines (rare but possible)."""
    diff = """diff --git a/noop.py b/noop.py
--- a/noop.py
+++ b/noop.py
@@ -5,3 +5,3 @@
 context1
 context2
 context3
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)

    assert summary.total_context == 3
    assert summary.total_added == 0
    assert summary.total_deleted == 0


def test_binary_files_skipped():
    """Binary file diffs have no content lines — just headers."""
    diff = """diff --git a/image.png b/image.png
--- a/image.png
+++ b/image.png
Binary files a/image.png and b/image.png differ
"""
    analyzer = DiffAnalyzer()
    summary = analyzer.analyze(diff)

    assert summary.files_changed == 1
    assert summary.total_added == 0
    assert summary.total_deleted == 0


if __name__ == "__main__":
    test_parse_simple_addition()
    test_parse_simple_deletion()
    test_parse_modification()
    test_parse_multiple_files()
    test_parse_rename()
    test_parse_mixed_hunks()
    test_empty_diff()
    test_to_dict()
    test_context_only_hunk()
    test_binary_files_skipped()
    print("All 10 tests passed.")