253 lines
8.2 KiB
Python
253 lines
8.2 KiB
Python
"""Unit tests for src/exporters/."""
|
|
|
|
import json
|
|
import os
|
|
import tempfile
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
|
|
from src.exporters.markdown import MarkdownExporter, _yaml_escape, _format_timestamp
|
|
from src.exporters.json_export import JSONExporter
|
|
|
|
|
|
SAMPLE_CONV = {
|
|
"id": "abc12345def67890",
|
|
"title": "Test Conversation",
|
|
"provider": "claude",
|
|
"project": "my-project",
|
|
"created_at": "2024-06-10T14:32:00Z",
|
|
"updated_at": "2024-06-10T15:00:00Z",
|
|
"message_count": 2,
|
|
"messages": [
|
|
{
|
|
"role": "user",
|
|
"content": "Hello, how are you?",
|
|
"content_type": "text",
|
|
"timestamp": "2024-06-10T14:32:00Z",
|
|
},
|
|
{
|
|
"role": "assistant",
|
|
"content": "I'm doing well, thank you! How can I help?",
|
|
"content_type": "text",
|
|
"timestamp": "2024-06-10T14:32:10Z",
|
|
},
|
|
],
|
|
}
|
|
|
|
NO_PROJECT_CONV = {
|
|
**SAMPLE_CONV,
|
|
"id": "noproj12345",
|
|
"project": None,
|
|
"title": "No Project Chat",
|
|
}
|
|
|
|
CODE_CONV = {
|
|
**SAMPLE_CONV,
|
|
"id": "code12345",
|
|
"messages": [
|
|
{
|
|
"role": "user",
|
|
"content": "Here is some code:\n```python\nprint('hello')\n```",
|
|
"content_type": "text",
|
|
"timestamp": None,
|
|
}
|
|
],
|
|
}
|
|
|
|
|
|
class TestMarkdownFrontmatter:
|
|
def test_yaml_frontmatter_present(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
content = path.read_text()
|
|
assert content.startswith("---\n")
|
|
assert "title: " in content
|
|
assert "provider: claude" in content
|
|
assert "conversation_id: abc12345def67890" in content
|
|
assert "created_at: 2024-06-10T14:32:00Z" in content
|
|
assert "exported_at: " in content
|
|
assert "message_count: 2" in content
|
|
assert "tags: [claude, my-project]" in content
|
|
|
|
def test_no_project_uses_no_project_label(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(NO_PROJECT_CONV)
|
|
content = path.read_text()
|
|
assert "project: no-project" in content
|
|
assert "tags: [claude]" in content
|
|
|
|
def test_metadata_table_present(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
content = path.read_text()
|
|
assert "| Provider | Claude |" in content
|
|
assert "| Project | my-project |" in content
|
|
assert "| Date | 2024-06-10 |" in content
|
|
assert "| Messages | 2 |" in content
|
|
|
|
def test_messages_rendered(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
content = path.read_text()
|
|
assert "Hello, how are you?" in content
|
|
assert "I'm doing well" in content
|
|
assert "🧑 Human" in content
|
|
assert "🤖 Assistant" in content
|
|
|
|
def test_code_fences_preserved(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(CODE_CONV)
|
|
content = path.read_text()
|
|
assert "```python" in content
|
|
assert "print('hello')" in content
|
|
|
|
|
|
class TestMarkdownFilenameGeneration:
|
|
def test_filename_format(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
assert path.name == "2024-06-10_test-conversation_abc12345.md"
|
|
|
|
def test_no_project_goes_to_no_project_dir(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(NO_PROJECT_CONV)
|
|
assert "no-project" in str(path)
|
|
|
|
def test_project_slug_in_path(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
assert "my-project" in str(path)
|
|
|
|
def test_year_in_path(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
assert "/2024/" in str(path)
|
|
|
|
def test_output_structure_provider_project(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path, output_structure="provider/project")
|
|
path = exp.export(SAMPLE_CONV)
|
|
# Should NOT have year subdirectory
|
|
parts = path.parts
|
|
assert "2024" not in parts
|
|
|
|
|
|
class TestMarkdownEmptyMessages:
|
|
def test_empty_message_skipped(self, tmp_path, caplog):
|
|
import logging
|
|
conv = {
|
|
**SAMPLE_CONV,
|
|
"messages": [
|
|
{"role": "user", "content": " ", "content_type": "text", "timestamp": None},
|
|
{"role": "assistant", "content": "Real response", "content_type": "text", "timestamp": None},
|
|
],
|
|
}
|
|
exp = MarkdownExporter(tmp_path)
|
|
with caplog.at_level(logging.WARNING, logger="src.exporters.markdown"):
|
|
path = exp.export(conv)
|
|
content = path.read_text()
|
|
assert "Real response" in content
|
|
assert any("empty" in r.message.lower() for r in caplog.records)
|
|
|
|
|
|
class TestMarkdownAtomicWrite:
|
|
def test_permissions_600(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
mode = oct(os.stat(path).st_mode)[-3:]
|
|
assert mode == "600"
|
|
|
|
def test_no_tmp_files_left(self, tmp_path):
|
|
exp = MarkdownExporter(tmp_path)
|
|
exp.export(SAMPLE_CONV)
|
|
tmp_files = list(tmp_path.rglob("*.tmp"))
|
|
assert tmp_files == []
|
|
|
|
|
|
class TestJSONExporter:
|
|
def test_produces_valid_json(self, tmp_path):
|
|
exp = JSONExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
data = json.loads(path.read_text())
|
|
assert data["id"] == "abc12345def67890"
|
|
assert data["title"] == "Test Conversation"
|
|
assert len(data["messages"]) == 2
|
|
|
|
def test_includes_exported_at(self, tmp_path):
|
|
exp = JSONExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
data = json.loads(path.read_text())
|
|
assert "exported_at" in data
|
|
|
|
def test_permissions_600(self, tmp_path):
|
|
exp = JSONExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
mode = oct(os.stat(path).st_mode)[-3:]
|
|
assert mode == "600"
|
|
|
|
def test_json_extension(self, tmp_path):
|
|
exp = JSONExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
assert path.suffix == ".json"
|
|
|
|
def test_pretty_printed(self, tmp_path):
|
|
exp = JSONExporter(tmp_path)
|
|
path = exp.export(SAMPLE_CONV)
|
|
raw = path.read_text()
|
|
# Pretty-printed JSON has newlines and indentation
|
|
assert "\n" in raw
|
|
assert " " in raw
|
|
|
|
|
|
class TestBothFormats:
|
|
"""T-38: Markdown and JSON exporters produce matching filenames for the same conversation."""
|
|
|
|
def test_both_formats_produce_files(self, tmp_path):
|
|
md_exp = MarkdownExporter(tmp_path)
|
|
json_exp = JSONExporter(tmp_path)
|
|
md_path = md_exp.export(SAMPLE_CONV)
|
|
json_path = json_exp.export(SAMPLE_CONV)
|
|
assert md_path.exists()
|
|
assert json_path.exists()
|
|
|
|
def test_both_formats_have_matching_stems(self, tmp_path):
|
|
md_exp = MarkdownExporter(tmp_path)
|
|
json_exp = JSONExporter(tmp_path)
|
|
md_path = md_exp.export(SAMPLE_CONV)
|
|
json_path = json_exp.export(SAMPLE_CONV)
|
|
assert md_path.suffix == ".md"
|
|
assert json_path.suffix == ".json"
|
|
assert md_path.stem == json_path.stem
|
|
|
|
def test_both_formats_same_directory(self, tmp_path):
|
|
md_exp = MarkdownExporter(tmp_path)
|
|
json_exp = JSONExporter(tmp_path)
|
|
md_path = md_exp.export(SAMPLE_CONV)
|
|
json_path = json_exp.export(SAMPLE_CONV)
|
|
assert md_path.parent == json_path.parent
|
|
|
|
|
|
class TestYamlEscape:
|
|
def test_escapes_double_quotes(self):
|
|
assert _yaml_escape('Say "hello"') == 'Say \\"hello\\"'
|
|
|
|
def test_escapes_backslash(self):
|
|
assert _yaml_escape("path\\to\\file") == "path\\\\to\\\\file"
|
|
|
|
def test_no_change_for_plain_string(self):
|
|
assert _yaml_escape("Hello World") == "Hello World"
|
|
|
|
|
|
class TestFormatTimestamp:
|
|
def test_strips_fractional_seconds(self):
|
|
result = _format_timestamp("2024-06-10T14:32:00.123456Z")
|
|
assert "." not in result
|
|
|
|
def test_replaces_T_with_space(self):
|
|
result = _format_timestamp("2024-06-10T14:32:00Z")
|
|
assert "T" not in result
|
|
assert "2024-06-10 14:32:00" == result
|
|
|
|
def test_empty_string(self):
|
|
assert _format_timestamp("") == ""
|