"""Schedule A assets and isolated writer acceptance gates.""" import hashlib import json from pathlib import Path import tempfile import unittest from lxml import etree as E import pikepdf BASE = Path(__file__).resolve().parents[2] class Imm5669Test(unittest.TestCase): def test_choice_groups_never_receive_illegal_saved_value_children(self): from imm5669_pdf import fill_schedule_a, dataset_node for lang, suffix in [('en', 'e'), ('fr', 'f')]: contract = json.loads((BASE/f'resources/imm/imm5669-contract.{lang}.json').read_text(encoding='utf8')) choices = {path: meta for path, meta in contract['fields'].items() if meta['kinds'][-1] == 'exclGroup'} self.assertEqual(len(choices), 12) for export in ('1', '2'): with tempfile.TemporaryDirectory(dir=BASE/'storage/app/imm/tmp') as work: output = Path(work)/'choices.pdf' fill_schedule_a({'form': 'IMM5669', 'language': lang, 'fields': dict.fromkeys(choices, export)}, output, BASE/f'storage/app/imm/imm5669{suffix}.pdf') with pikepdf.open(output) as pdf: x = pdf.Root.AcroForm.XFA packets = {str(x[i]): x[i+1].read_bytes() for i in range(0,len(x),2)} form, data = E.fromstring(packets['form']), E.fromstring(packets['datasets']) self.assertEqual(form.xpath('//*[local-name()="exclGroup"]/*[local-name()="value"]'), [], 'Adobe rejects value as a child of exclGroup') for path, meta in choices.items(): self.assertEqual(dataset_node(data, meta['data_path']).text, export) def test_contracts_and_standalone_bilingual_questionnaire(self): seed = json.loads((BASE/'database/data/schedule_a_hub_questionnaire.json').read_text(encoding='utf8')) self.assertEqual(seed['questionnaire_key'], 'schedule_a_hub') questions = [q for s in seed['sections'] for q in s['questions']] self.assertEqual(seed['question_count'], len(questions)) self.assertEqual(len({q['source_key'] for q in questions}), len(questions)) for q in questions: self.assertTrue(q['source_key'].startswith('schedule_a_hub.imm5669.')) self.assertTrue(q['label'] and q['label_fr']) self.assertLessEqual(len(q['label']),255) self.assertLessEqual(len(q['label_fr']),500) if '.declaration_' in q['source_key']: self.assertTrue(q['help_text'] and q['help_text_fr']) for lang, suffix in [('en','e'),('fr','f')]: contract = json.loads((BASE/f'resources/imm/imm5669-contract.{lang}.json').read_text(encoding='utf8')) self.assertEqual(contract['sha256'], hashlib.sha256((BASE/f'storage/app/imm/imm5669{suffix}.pdf').read_bytes()).hexdigest()) self.assertEqual(contract['root'], 'IMM_5669') self.assertEqual(len(contract['tables']['education']['rows']), 5) self.assertEqual(len(contract['tables']['addresses']['rows']), 10) def test_writer_creates_packets_and_preserves_original_template(self): from imm5669_pdf import fill_schedule_a for lang, suffix in [('en','e'),('fr','f')]: contract = json.loads((BASE/f'resources/imm/imm5669-contract.{lang}.json').read_text(encoding='utf8')) fields = {path: ('1' if meta['kind'] in ('role','declaration') else 'SMITH') for path,meta in contract['fields'].items() if meta['source'] in ('role','family_name','declaration_a')} blank = BASE/f'storage/app/imm/imm5669{suffix}.pdf' with tempfile.TemporaryDirectory(dir=BASE/'storage/app/imm/tmp') as work: output = Path(work)/'draft.pdf' fill_schedule_a({'form':'IMM5669','language':lang,'fields':fields},output,blank) self.assertTrue(output.read_bytes().startswith(blank.read_bytes())) def packets(path): with pikepdf.open(path) as p: x=p.Root.AcroForm.XFA return {str(x[i]):x[i+1].read_bytes() for i in range(0,len(x),2)} original, generated = packets(blank), packets(output) self.assertEqual(original['template'],generated['template']) for name, raw in original.items(): self.assertEqual(raw, generated[name], name) self.assertIn('datasets',generated) self.assertIn('form',generated) from imm5669_pdf import dataset_node data=E.fromstring(generated['datasets']) for path,value in fields.items(): self.assertEqual(dataset_node(data,contract['fields'][path]['data_path']).text,value) with self.assertRaisesRegex(ValueError,'Unallowlisted'): fill_schedule_a({'form':'IMM5669','language':lang,'fields':{'IMM_5669.page4.signatureApplicant':'fake'}},Path(work)/'invalid.pdf',blank) def test_all_allowlisted_targets_and_saved_values_match_in_both_languages(self): from imm5669_pdf import fill_schedule_a, dataset_node from imm0008_xfa_checksum import form_checksum for lang,suffix in [('en','e'),('fr','f')]: contract=json.loads((BASE/f'resources/imm/imm5669-contract.{lang}.json').read_text(encoding='utf8')) fields={path:('2' if meta['kind'] in ('role','declaration') else '1990-01-01' if meta['kind']=='date' else '2010-01' if meta['kind']=='month' else '4' if meta['kind']=='years' else 'Confirmed detail') for path,meta in contract['fields'].items()} with tempfile.TemporaryDirectory(dir=BASE/'storage/app/imm/tmp') as work: blank=BASE/f'storage/app/imm/imm5669{suffix}.pdf' output=Path(work)/'all.pdf' payload={'form':'IMM5669','language':lang,'fields':fields} fill_schedule_a(payload,output,blank) with pikepdf.open(output) as pdf: x=pdf.Root.AcroForm.XFA packets={str(x[i]):x[i+1].read_bytes() for i in range(0,len(x),2)} data,form=E.fromstring(packets['datasets']),E.fromstring(packets['form']) self.assertEqual(form.get('checksum'),form_checksum(packets['template'],packets['datasets'])) for path,value in fields.items(): meta=contract['fields'][path] self.assertEqual(value,dataset_node(data,meta['data_path']).text,path) saved=dataset_node(form,meta['template_path'],kinds=meta['kinds']) if meta['kinds'][-1] == 'exclGroup': self.assertEqual(saved.xpath('./*[local-name()="value"]'), [], path) else: self.assertEqual(value,saved.xpath('string(./*[local-name()="value"]/*[1])'),path) with self.assertRaisesRegex(ValueError,'export'): role=next(p for p,m in contract['fields'].items() if m['kind']=='role') fill_schedule_a({**payload,'fields':{role:'Y'}},Path(work)/'bad.pdf',blank) with self.assertRaisesRegex(ValueError,'overwrite'): fill_schedule_a(payload,blank,blank)