Skip to content

test_samediggi_no_crawler

Test the SamediggiNoPage class.

TestSamediggiNoPage

Bases: unittest.TestCase

Test the SamediggiNoPage class.

Source code in /home/anders/projects/CorpusTools/corpustools/test/test_samediggi_no_crawler.py
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
class TestSamediggiNoPage(unittest.TestCase):
    """Test the SamediggiNoPage class."""

    def test_basics(self):
        """Test initial values."""
        with requests_mock.Mocker() as mocker:
            mocker.get(
                "https://samediggi.no/Odasarkiiva/PRD-Doarjja-julevsami-giellaproseavttaide",
                text=MY_TEXT,
                headers={"content-type": "text/html; charset=UTF-8"},
            )
            result = requests.get(
                "https://samediggi.no/Odasarkiiva/PRD-Doarjja-julevsami-giellaproseavttaide"
            )

            page = samediggi_no_crawler.SamediggiNoPage(result, {})
            self.assertEqual(
                page.corpuspath.orig,
                os.path.join(
                    os.getenv("GTFREE"),
                    "orig/sme/admin/sd/samediggi.no/prd-doarjja-julevsami-giellaproseavttaide.html",
                ),
            )
            self.assertEqual(
                page.url,
                "https://samediggi.no/Odasarkiiva/PRD-Doarjja-julevsami-giellaproseavttaide",
            )
            self.assertListEqual(
                page.parallel_links,
                [
                    "https://www.sametinget.no/Nyhetsarkiv/parallel1",
                    "https://www.saemiedigkie.no/Saernievaaarhkoe/parallel1",
                    "https://www.samedigge.no/AAdaasa/PRD-Doarjju-julevsame-giellaprosjevtajda",
                ],
            )
            self.assertTrue(page.saveable)
            self.assertEqual(page.lang, "sme")
            self.assertSetEqual(
                page.links,
                {
                        "https://samediggi.no/Balvalusat2/Dearvvasvuohta-ja-sosiala",
                        "https://samediggi.no/Vuoigatvuodat",
                        "https://samediggi.no/Samedikki-birra2/Rehket-Samediggai",
                        "https://samediggi.no/Samedikki-birra",
                        "https://samediggi.no/Doarjagat-ja-stipeanddat",
                        "https://samediggi.no/Preassa",
                        "https://samediggi.no/Samedikki-birra2/langlink",
                        "https://samediggi.no/Balvalusat2/Ealahusat",
                        "https://samediggi.no/Politihkka2/Assit-ja-dokumeanttat",
                        "https://samediggi.no/Odasarkiiva",
                        "https://samediggi.no/Balvalusat2/Giella",
                        "https://samediggi.no/Samedikki-birra2/langlink2",
                        "https://samediggi.no/Balvalusat2/Kultuvra",
                        "https://samediggi.no/Valga",
                        "https://samediggi.no/Balvalusat2/Dassearvu",
                        "https://samediggi.no/Politihkka2",
                        "https://samediggi.no/Lagideamit",
                        "https://samediggi.no/Balvalusat2/Oahpahus-ja-oahpponeavvut",
                        "https://samediggi.no/Balvalusat2/Riikkaidgaskasas-bargu",
                        "https://samediggi.no/Girjeradju",
                        "https://samediggi.no/Balvalusat2/Biras-areala-ja-kultursuodjaleapmi",
                        "https://samediggi.no/Balvalusat2/Manaidgardi",
                },
            )

            page.set_initial_metadata()
            self.assertEqual(
                page.corpuspath.metadata.get_variable("title"),
                "PRD: Doarjja julevsámi giellaprošeavttaide",
            )
            self.assertEqual(
                page.corpuspath.metadata.get_variable("filename"),
                "https://samediggi.no/Odasarkiiva/PRD-Doarjja-julevsami-giellaproseavttaide",
            )
            self.assertEqual(page.corpuspath.metadata.get_variable("mainlang"), "sme")
            self.assertEqual(page.corpuspath.metadata.get_variable("genre"), "admin")
            self.assertEqual(
                page.corpuspath.metadata.get_variable("translated_from"), "nob"
            )

test_basics()

Test initial values.

Source code in /home/anders/projects/CorpusTools/corpustools/test/test_samediggi_no_crawler.py
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
def test_basics(self):
    """Test initial values."""
    with requests_mock.Mocker() as mocker:
        mocker.get(
            "https://samediggi.no/Odasarkiiva/PRD-Doarjja-julevsami-giellaproseavttaide",
            text=MY_TEXT,
            headers={"content-type": "text/html; charset=UTF-8"},
        )
        result = requests.get(
            "https://samediggi.no/Odasarkiiva/PRD-Doarjja-julevsami-giellaproseavttaide"
        )

        page = samediggi_no_crawler.SamediggiNoPage(result, {})
        self.assertEqual(
            page.corpuspath.orig,
            os.path.join(
                os.getenv("GTFREE"),
                "orig/sme/admin/sd/samediggi.no/prd-doarjja-julevsami-giellaproseavttaide.html",
            ),
        )
        self.assertEqual(
            page.url,
            "https://samediggi.no/Odasarkiiva/PRD-Doarjja-julevsami-giellaproseavttaide",
        )
        self.assertListEqual(
            page.parallel_links,
            [
                "https://www.sametinget.no/Nyhetsarkiv/parallel1",
                "https://www.saemiedigkie.no/Saernievaaarhkoe/parallel1",
                "https://www.samedigge.no/AAdaasa/PRD-Doarjju-julevsame-giellaprosjevtajda",
            ],
        )
        self.assertTrue(page.saveable)
        self.assertEqual(page.lang, "sme")
        self.assertSetEqual(
            page.links,
            {
                    "https://samediggi.no/Balvalusat2/Dearvvasvuohta-ja-sosiala",
                    "https://samediggi.no/Vuoigatvuodat",
                    "https://samediggi.no/Samedikki-birra2/Rehket-Samediggai",
                    "https://samediggi.no/Samedikki-birra",
                    "https://samediggi.no/Doarjagat-ja-stipeanddat",
                    "https://samediggi.no/Preassa",
                    "https://samediggi.no/Samedikki-birra2/langlink",
                    "https://samediggi.no/Balvalusat2/Ealahusat",
                    "https://samediggi.no/Politihkka2/Assit-ja-dokumeanttat",
                    "https://samediggi.no/Odasarkiiva",
                    "https://samediggi.no/Balvalusat2/Giella",
                    "https://samediggi.no/Samedikki-birra2/langlink2",
                    "https://samediggi.no/Balvalusat2/Kultuvra",
                    "https://samediggi.no/Valga",
                    "https://samediggi.no/Balvalusat2/Dassearvu",
                    "https://samediggi.no/Politihkka2",
                    "https://samediggi.no/Lagideamit",
                    "https://samediggi.no/Balvalusat2/Oahpahus-ja-oahpponeavvut",
                    "https://samediggi.no/Balvalusat2/Riikkaidgaskasas-bargu",
                    "https://samediggi.no/Girjeradju",
                    "https://samediggi.no/Balvalusat2/Biras-areala-ja-kultursuodjaleapmi",
                    "https://samediggi.no/Balvalusat2/Manaidgardi",
            },
        )

        page.set_initial_metadata()
        self.assertEqual(
            page.corpuspath.metadata.get_variable("title"),
            "PRD: Doarjja julevsámi giellaprošeavttaide",
        )
        self.assertEqual(
            page.corpuspath.metadata.get_variable("filename"),
            "https://samediggi.no/Odasarkiiva/PRD-Doarjja-julevsami-giellaproseavttaide",
        )
        self.assertEqual(page.corpuspath.metadata.get_variable("mainlang"), "sme")
        self.assertEqual(page.corpuspath.metadata.get_variable("genre"), "admin")
        self.assertEqual(
            page.corpuspath.metadata.get_variable("translated_from"), "nob"
        )