simple-nanoshare/tests/test_text_view.py

import asyncio

import pytest

from my_modules.textview import (
  build_segments,
  decode_text,
  parse_share_link,
  tokenize,
  wants_text_view,
)

EXAMPLE = (
  'Ref Barend: https://i.example.dev/-Q6jmHoRYjFi\n'
  'Ref Hanzoku: https://i.example.dev/-eluTduc-npP in female shape like '
  'https://i.example.dev/-kzmIlsf7uUe please\n\n'
  'Hoodie/Skateboard : https://i.example.dev/-to4wPonAadV\n\n'
  'Barend has just returned from a skating trip.'
)

# --- tokenize -----------------------------------------------------------------

def test_tokenize_finds_every_url_in_the_example():
  urls = [value for kind, value in tokenize(EXAMPLE) if kind == 'url']
  assert urls == [
    'https://i.example.dev/-Q6jmHoRYjFi',
    'https://i.example.dev/-eluTduc-npP',
    'https://i.example.dev/-kzmIlsf7uUe',
    'https://i.example.dev/-to4wPonAadV',
  ]

def test_tokenize_keeps_text_lossless():
  assert ''.join(value for _, value in tokenize(EXAMPLE)) == EXAMPLE

@pytest.mark.parametrize('raw,expected', [
  ('see https://example.com/-abc.', 'https://example.com/-abc'),
  ('see https://example.com/-abc, next', 'https://example.com/-abc'),
  ('(https://example.com/-abc)', 'https://example.com/-abc'),
  ('https://example.com/wiki/Foo_(bar)', 'https://example.com/wiki/Foo_(bar)'),
])
def test_tokenize_trims_sentence_punctuation(raw, expected):
  assert [value for kind, value in tokenize(raw) if kind == 'url'] == [expected]

@pytest.mark.parametrize('raw', [
  'javascript:alert(1)',
  'data:text/html,<script>alert(1)</script>',
  'ftp://example.com/file',
])
def test_tokenize_ignores_non_http_schemes(raw):
  assert [kind for kind, _ in tokenize(raw)] == ['text']

# --- parse_share_link ---------------------------------------------------------

@pytest.mark.parametrize('url,expected', [
  ('https://i.example.dev/-abc123', 'abc123'),
  ('http://i.example.dev/-abc123', 'abc123'),
  ('https://I.Example.Dev/-abc123', 'abc123'),
  ('https://i.example.dev/-abc123?download=1', 'abc123'),
  ('https://other.example.dev/-abc123', None),
  ('https://i.example.dev/files/-abc123', None),
  ('https://i.example.dev/-abc/def', None),
  ('https://i.example.dev/-', None),
  ('https://i.example.dev/', None),
])
def test_parse_share_link(url, expected):
  assert parse_share_link(url, host='i.example.dev') == expected

def test_parse_share_link_honours_root_path():
  assert parse_share_link('https://h/sub/-abc', host='h', root_path='/sub') == 'abc'
  assert parse_share_link('https://h/-abc', host='h', root_path='/sub') is None

def test_parse_share_link_rejects_host_lookalike():
  assert parse_share_link('https://i.example.dev.evil.tld/-abc', host='i.example.dev') is None

# --- build_segments -----------------------------------------------------------

FILES = {
  'Q6jmHoRYjFi': {'file_name': 'barend.png', 'content_type': 'image/png'},
  'eluTduc-npP': {'file_name': 'hanzoku.png', 'content_type': 'image/png'},
  'kzmIlsf7uUe': {'file_name': 'gone.png', 'content_type': 'image/png', 'expired': True},
  'to4wPonAadV': {'file_name': 'notes.txt', 'content_type': 'text/plain'},
}

def make_lookup(calls: list):
  async def lookup(file_id):
    calls.append(file_id)
    return FILES.get(file_id)
  return lookup

def run(coro):
  return asyncio.run(coro)

def test_build_segments_resolves_own_links():
  calls = []
  segments = run(build_segments(
    EXAMPLE, host='i.example.dev', root_path='', lookup=make_lookup(calls)
  ))
  by_url = {s['url']: s for s in segments if s['kind'] in {'link', 'file'}}

  assert by_url['https://i.example.dev/-Q6jmHoRYjFi'] == {
    'kind': 'file', 'url': 'https://i.example.dev/-Q6jmHoRYjFi', 'state': 'ok',
    'file_id': 'Q6jmHoRYjFi', 'label': 'barend.png', 'media': 'image',
  }
  assert by_url['https://i.example.dev/-to4wPonAadV']['media'] == 'other'
  assert by_url['https://i.example.dev/-kzmIlsf7uUe']['kind'] == 'link'
  assert by_url['https://i.example.dev/-kzmIlsf7uUe']['state'] == 'expired'
  assert sorted(calls) == sorted(FILES)

def test_build_segments_keeps_foreign_links_unresolved():
  calls = []
  segments = run(build_segments(
    'see https://other.example.dev/-Q6jmHoRYjFi',
    host='i.example.dev', root_path='', lookup=make_lookup(calls),
  ))
  assert segments[-1] == {'kind': 'link', 'url': 'https://other.example.dev/-Q6jmHoRYjFi'}
  assert calls == []

def test_build_segments_marks_missing_files():
  segments = run(build_segments(
    'https://i.example.dev/-nope', host='i.example.dev', root_path='', lookup=make_lookup([])
  ))
  assert segments[0]['kind'] == 'link'
  assert segments[0]['state'] == 'missing'

def test_build_segments_deduplicates_lookups():
  calls = []
  text = 'https://i.example.dev/-Q6jmHoRYjFi and https://i.example.dev/-Q6jmHoRYjFi'
  run(build_segments(text, host='i.example.dev', root_path='', lookup=make_lookup(calls)))
  assert calls == ['Q6jmHoRYjFi']

def test_build_segments_caps_lookups():
  calls = []
  text = ' '.join(f'https://i.example.dev/-id{index}' for index in range(30))
  segments = run(build_segments(
    text, host='i.example.dev', root_path='', lookup=make_lookup(calls), max_lookups=5
  ))
  assert len(calls) == 5
  assert all(s['kind'] == 'link' for s in segments if s['kind'] != 'text')

def test_build_segments_without_lookup_only_linkifies():
  segments = run(build_segments(EXAMPLE, host='i.example.dev', root_path='', lookup=None))
  assert all(segment['kind'] in {'text', 'link'} for segment in segments)

# --- wants_text_view ----------------------------------------------------------

BROWSER_ACCEPT = 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8'

@pytest.mark.parametrize('content_type,accept,raw,download,expected', [
  ('text/plain', BROWSER_ACCEPT, None, False, True),
  ('text/plain; charset=utf-8', BROWSER_ACCEPT, None, False, True),
  ('text/markdown', BROWSER_ACCEPT, None, False, True),
  ('text/plain', '*/*', None, False, False),
  ('text/plain', '', None, False, False),
  ('text/plain', BROWSER_ACCEPT, '1', False, False),
  ('text/plain', BROWSER_ACCEPT, 'true', False, False),
  ('text/plain', BROWSER_ACCEPT, None, True, False),
  ('image/png', BROWSER_ACCEPT, None, False, False),
  ('application/json', BROWSER_ACCEPT, None, False, False),
])
def test_wants_text_view(content_type, accept, raw, download, expected):
  assert wants_text_view(
    content_type=content_type, accept_header=accept, raw_param=raw, force_download=download
  ) is expected

# --- decode_text --------------------------------------------------------------

def test_decode_text_keeps_the_whole_document():
  raw = b'x' * (512 * 1024)
  assert len(decode_text(raw)) == len(raw)

def test_decode_text_survives_invalid_utf8():
  assert decode_text(b'ok \xff\xfe').startswith('ok ')