Robert Helewka
99bdb4ac92
- Implemented custom form widgets for date, time, and datetime fields with DaisyUI styling. - Created utility functions for formatting dates, times, and numbers according to user preferences. - Developed views for profile settings, API key management, and notifications, including health check endpoints. - Added URL configurations for Themis tests and main application routes. - Established test cases for custom widgets to ensure proper functionality and integration. - Defined project metadata and dependencies in pyproject.toml for package management.
122 lines
3.8 KiB
Python
122 lines
3.8 KiB
Python
"""
|
|
Tests for text sanitization utilities.
|
|
"""
|
|
|
|
from django.test import TestCase
|
|
|
|
from library.services.text_utils import (
|
|
clean_pdf_artifacts,
|
|
remove_excessive_whitespace,
|
|
sanitize_text,
|
|
truncate_text,
|
|
)
|
|
|
|
|
|
class SanitizeTextTests(TestCase):
|
|
"""Tests for the sanitize_text function."""
|
|
|
|
def test_empty_string(self):
|
|
self.assertEqual(sanitize_text("", log_changes=False), "")
|
|
|
|
def test_none_input(self):
|
|
self.assertIsNone(sanitize_text(None, log_changes=False))
|
|
|
|
def test_clean_text_unchanged(self):
|
|
text = "Hello, this is clean text."
|
|
self.assertEqual(sanitize_text(text, log_changes=False), text)
|
|
|
|
def test_removes_null_bytes(self):
|
|
text = "Hello\x00World"
|
|
result = sanitize_text(text, log_changes=False)
|
|
self.assertNotIn("\x00", result)
|
|
self.assertEqual(result, "HelloWorld")
|
|
|
|
def test_removes_control_characters(self):
|
|
text = "Hello\x07World\x0eTest"
|
|
result = sanitize_text(text, log_changes=False)
|
|
self.assertNotIn("\x07", result)
|
|
self.assertNotIn("\x0e", result)
|
|
|
|
def test_preserves_newlines_and_tabs(self):
|
|
text = "Hello\nWorld\tTest\r\n"
|
|
result = sanitize_text(text, log_changes=False)
|
|
self.assertIn("\n", result)
|
|
self.assertIn("\t", result)
|
|
|
|
def test_removes_zero_width_characters(self):
|
|
text = "Hello\u200bWorld"
|
|
result = sanitize_text(text, log_changes=False)
|
|
self.assertNotIn("\u200b", result)
|
|
|
|
def test_normalizes_unicode(self):
|
|
# é as combining characters vs. precomposed
|
|
combining = "e\u0301" # e + combining acute
|
|
result = sanitize_text(combining, log_changes=False)
|
|
self.assertEqual(result, "\u00e9") # precomposed é
|
|
|
|
def test_cleans_pdf_ligatures(self):
|
|
text = "finding the flow of effort"
|
|
result = sanitize_text(text, log_changes=False)
|
|
self.assertIn("fi", result)
|
|
self.assertIn("fl", result)
|
|
self.assertIn("ff", result)
|
|
|
|
|
|
class CleanPdfArtifactsTests(TestCase):
|
|
"""Tests for clean_pdf_artifacts."""
|
|
|
|
def test_replaces_smart_quotes(self):
|
|
text = "\u201cHello\u201d \u2018World\u2019"
|
|
result = clean_pdf_artifacts(text)
|
|
self.assertEqual(result, '"Hello" \'World\'')
|
|
|
|
def test_replaces_dashes(self):
|
|
text = "word\u2013word\u2014end"
|
|
result = clean_pdf_artifacts(text)
|
|
self.assertEqual(result, "word-word-end")
|
|
|
|
def test_replaces_ellipsis(self):
|
|
text = "wait\u2026"
|
|
result = clean_pdf_artifacts(text)
|
|
self.assertEqual(result, "wait...")
|
|
|
|
def test_replaces_nbsp(self):
|
|
text = "non\u00a0breaking"
|
|
result = clean_pdf_artifacts(text)
|
|
self.assertEqual(result, "non breaking")
|
|
|
|
|
|
class RemoveExcessiveWhitespaceTests(TestCase):
|
|
"""Tests for remove_excessive_whitespace."""
|
|
|
|
def test_collapses_spaces(self):
|
|
self.assertEqual(remove_excessive_whitespace("a b"), "a b")
|
|
|
|
def test_collapses_newlines(self):
|
|
self.assertEqual(
|
|
remove_excessive_whitespace("a\n\n\n\nb"), "a\n\nb"
|
|
)
|
|
|
|
def test_strips_line_whitespace(self):
|
|
self.assertEqual(
|
|
remove_excessive_whitespace(" hello \n world "),
|
|
"hello\nworld",
|
|
)
|
|
|
|
|
|
class TruncateTextTests(TestCase):
|
|
"""Tests for truncate_text."""
|
|
|
|
def test_short_text_unchanged(self):
|
|
self.assertEqual(truncate_text("hello", 100), "hello")
|
|
|
|
def test_truncates_at_word_boundary(self):
|
|
text = "hello beautiful world"
|
|
result = truncate_text(text, 15)
|
|
self.assertTrue(result.endswith("..."))
|
|
self.assertLessEqual(len(result), 15)
|
|
|
|
def test_custom_suffix(self):
|
|
result = truncate_text("hello beautiful world", 15, suffix="…")
|
|
self.assertTrue(result.endswith("…"))
|