"""

Quick unit tests for ingestion pipeline.

Run with: pytest test_ingest.py -v

"""

from pathlib import Path

import tempfile

import shutil

from ingest import (

list_source_files,

read_text_safely,

chunk_lines,

detect_lang_from_ext,

hash_text,

make_preview,

ingest_repo,

Chunk,

)

def test_list_source_files():

"""Test file discovery with filters."""

# Create a temp directory structure

with tempfile.TemporaryDirectory() as tmpdir:

root = Path(tmpdir)

# Create test files

(root / "main.py").write_text("print('hello')")

(root / "test.ts").write_text("console.log('hi')")

(root / "ignore.txt").write_text("should not appear")

(root / "node_modules").mkdir()

(root / "node_modules" / "lib.js").write_text("module.exports = {}")

files = list_source_files(root, [".py", ".ts"], ["node_modules"])

assert "main.py" in files

assert "test.ts" in files

assert "ignore.txt" not in files

assert "node_modules/lib.js" not in files

def test_read_text_safely():

"""Test file reading with encoding detection."""

with tempfile.TemporaryDirectory() as tmpdir:

path = Path(tmpdir) / "test.py"

path.write_text("def hello():\n return 'world'", encoding="utf-8")

content = read_text_safely(path)

assert content is not None

assert "def hello()" in content

def test_chunk_lines():

"""Test fixed-window chunking."""

lines = [f"line {i}\n" for i in range(100)]

chunks = chunk_lines(lines, window=20, overlap=5, min_lines=5)

# First chunk: lines 1-20

assert chunks[0].start_line == 1

assert chunks[0].end_line == 20

# Second chunk: lines 16-35 (overlap of 5)

assert chunks[1].start_line == 16

# All chunks should have text

for chunk in chunks:

assert len(chunk.text) > 0

def test_detect_lang():

"""Test language detection."""

assert detect_lang_from_ext("main.py") == "python"

assert detect_lang_from_ext("app.ts") == "typescript"

assert detect_lang_from_ext("server.js") == "javascript"

assert detect_lang_from_ext("main.go") == "go"

assert detect_lang_from_ext("unknown.xyz") == "unknown"

def test_hash_text():

"""Test hashing for deduplication."""

text1 = "def hello():\n pass"

text2 = "def hello():\n pass"

text3 = "def world():\n pass"

hash1 = hash_text(text1)

hash2 = hash_text(text2)

hash3 = hash_text(text3)

assert hash1 == hash2 # Same text = same hash

assert hash1 != hash3 # Different text = different hash

assert len(hash1) == 16 # Should be truncated to 16 chars

def test_make_preview():

"""Test preview generation."""

text = "\n".join([f"line {i}" for i in range(50)])

preview = make_preview(text, max_lines=10)

preview_lines = preview.splitlines()

assert len(preview_lines) <= 10

assert "line 0" in preview

assert "line 9" in preview or "line 10" in preview

def test_ingest_repo_integration():

"""Integration test: full pipeline."""

with tempfile.TemporaryDirectory() as tmpdir:

root = Path(tmpdir)

# Create a mini repo

(root / "main.py").write_text(

"\n".join([f"# Line {i}" for i in range(100)])

)

(root / "utils.py").write_text(

"def helper():\n return True"

)

chunks, stats = ingest_repo(

root,

include_exts=[".py"],

window=20,

overlap=5,

min_lines=5,

)

assert stats["files_scanned"] == 2

assert stats["files_read"] == 2

assert stats["chunks_total"] > 0

assert len(chunks) == stats["chunks_total"]

# Check chunk structure

chunk = chunks[0]

assert chunk.repo == root.name

assert chunk.lang == "python"

assert chunk.start_line >= 1

assert chunk.end_line > chunk.start_line

assert len(chunk.text) > 0

assert len(chunk.hash) == 16

assert len(chunk.preview) > 0

if __name__ == "__main__":

# Run tests manually

test_list_source_files()

test_read_text_safely()

test_chunk_lines()

test_detect_lang()

test_hash_text()

test_make_preview()

test_ingest_repo_integration()

print("✅ All tests passed!")