def test_3pg_no_ocr(self, mocks):
init_doc(FakePdf("..."))
trigger_processing()
assert mocks["page_extracted"].call_count == 3
# No pages need to be OCRd when text is extractable
assert mocks["page_ocrd"].call_count == 0
assert mocks["page_text_position_extracted"].call_count == 3
def test_redaction(self, mocks):
init_doc(FakePdf("..."))
trigger_processing()
# Redact the middle page
reset_mocks(mocks)
trigger_redacting([1])
# Only 1 page should be processed, but the cache is rebuilt
assert mocks["cache_miss"].call_count == 3
assert mocks["page_loaded"].call_count == 2 # 1 + 1
assert mocks["page_extracted"].call_count == 1
assert mocks["page_ocrd"].call_count == 1
assert mocks["page_text_position_extracted"].call_count == 1
def test_cache_dirty_batching(self, mocks):
init_doc(FakePdf("....."))
trigger_processing()
# Redact the last two pages
reset_mocks(mocks)
trigger_redacting([3, 4])
# Cache should only be written once
assert mocks["cache_written"].call_count == 1
# Cache should be missed number of pages times
assert mocks["cache_miss"].call_count == 5
# Two pages should be extracted
assert mocks["page_extracted"].call_count == 2
assert mocks["page_ocrd"].call_count == 2
assert mocks["page_text_position_extracted"].call_count == 2
def test_cache_misses(self, mocks):
init_doc(FakePdf("..."))
trigger_processing()
assert mocks["cache_miss"].call_count == 3
def test_5pg_half_ocr(self, mocks):
init_doc(FakePdf(".o.o."))
trigger_processing()
assert mocks["page_extracted"].call_count == 5
assert mocks["page_ocrd"].call_count == 2
assert mocks["page_text_position_extracted"].call_count == 5
def test_3pg_ocr(self, mocks):
init_doc(FakePdf("ooo"))
trigger_processing()
assert mocks["page_extracted"].call_count == 3
assert mocks["page_ocrd"].call_count == 3
assert mocks["page_text_position_extracted"].call_count == 3
def test_cache_batching(self, mocks):
init_doc(FakePdf("....."))
trigger_processing()
# Cache should be written only once
assert mocks["cache_written"].call_count == 1