"""IMM5708 contact state regression checks; does not run Adobe Reader.""" import json import hashlib import re from pathlib import Path import tempfile import unittest import xml.etree.ElementTree as ET import pikepdf from prefill_imm_xfa_pdf import fill from imm5708_pdf_state import contact_phone_state, contact_address_state, address_values, ADDRESS_SELECTION BASE = Path(__file__).resolve().parents[2] def raw_packets(path): with pikepdf.open(path) as pdf: xfa = pdf.Root.AcroForm.XFA return {str(xfa[i]): xfa[i + 1].read_bytes() for i in range(0, len(xfa), 2) if str(xfa[i]) in {'template', 'form', 'datasets'}} def packets(path): return {name: ET.fromstring(data) for name, data in raw_packets(path).items()} def named_nodes(root): result = {} def visit(node, path=''): if node.get('name'): path = path + '.' + node.get('name') result[path.lstrip('.')] = node for child in node: visit(child, path) visit(root) return result class Imm5708ContactStateTest(unittest.TestCase): def test_address_normalization_does_not_mutate_answers_or_guess_selection(self): fields = {ADDRESS_SELECTION: 'Y', 'ContactInformation.Resi.AddrLine2.Country': '511'} before = dict(fields) normalized = address_values(fields) self.assertEqual(before, fields) self.assertEqual('', normalized['ContactInformation.Resi.AddrLine2.Country']) self.assertEqual('', normalized['ContactInformation.Resi.AddrLine2.District']) for choice in [None, '', 'unknown']: fields[ADDRESS_SELECTION] = choice self.assertEqual(fields, address_values(fields)) self.assertEqual(b'unchanged', contact_address_state(b'unchanged', fields)) def test_address_selection_and_residential_state_in_generated_pdfs(self): selection = 'ContactInformation.Resi.SameAsAddr.SameAsMailingInd' residential = 'ContactInformation.Resi.' address = {'AddrLine1.AptUnit': '4', 'AddrLine1.StreetNum': '123', 'AddrLine1.Streetname': 'Queen Street', 'AddrLine2.City': 'Toronto', 'AddrLine2.Country': '511', 'AddrLine2.Prov': '06', 'AddrLine2.PostalCode': 'M5V 2T6', 'AddrLine2.District': 'Toronto'} for suffix in ['e', 'f']: blank = BASE / f'storage/app/imm/imm5708{suffix}.pdf' original_hash = hashlib.sha256(blank.read_bytes()).hexdigest() original_raw = raw_packets(blank) original = {name: ET.fromstring(data) for name, data in original_raw.items()} for choice in ['Y', 'N', None]: with self.subTest(language=suffix, choice=choice): fields = {residential + k: v for k, v in address.items()} fields['ComingIntoCda.OrigEntry.Place'] = 'Toronto Pearson Airport' if choice: fields[selection] = choice with tempfile.TemporaryDirectory() as temp: payload, output = Path(temp) / 'payload.json', Path(temp) / 'draft.pdf' payload.write_text(json.dumps({'form': 'IMM5708', 'fields': fields}), encoding='utf-8') fill(str(payload), str(output), str(blank)) written_raw = raw_packets(output) written = {name: ET.fromstring(data) for name, data in written_raw.items()} for name in ['template', 'form']: self.assertEqual(re.findall(br'<\?.*?\?>', original_raw[name], re.S), re.findall(br'<\?.*?\?>', written_raw[name], re.S)) from prefill_imm_xfa_pdf import find_by_path, XFA_DATA_NS node = find_by_path(written['datasets'], selection)[0] if choice: self.assertEqual(choice, node.text) self.assertEqual('dataValue', node.get(f'{{{XFA_DATA_NS}}}dataNode')) else: self.assertNotIn(node.text, ['Y', 'N']) self.assertEqual([], list(node)) self.assertEqual('Toronto Pearson Airport', find_by_path(written['datasets'], 'ComingIntoCda.OrigEntry.Place')[0].text) for packet in ['template', 'form']: nodes = named_nodes(written[packet]) prefix = 'form1.Page2.' if choice: group = nodes[prefix + selection] value = next(c for c in group if c.tag.endswith('value')) self.assertEqual(choice, next(iter(value)).text) for member, on in [('Yes', 'Y'), ('No', 'N')]: member_node = nodes[prefix + selection + '.' + member] member_values = [n.text for c in member_node if c.tag.endswith('value') for n in c] self.assertEqual([on] if choice == on else [None], member_values) for key in address: field = nodes[prefix + residential + key] expected_access = 'readOnly' if choice == 'Y' or key.endswith('Country') else 'open' self.assertEqual(expected_access, field.get('access')) validate = next(c for c in field if c.tag.endswith('validate')) required = choice == 'N' and not key.endswith(('AptUnit', 'StreetNum', 'District')) self.assertEqual('error' if required else 'disabled', validate.get('nullTest')) for key, value in address.items(): node = find_by_path(written['datasets'], residential + key)[0] self.assertEqual('' if choice == 'Y' else value, node.text or '') scripts = lambda root: [n.text for n in root.iter() if n.tag.endswith('script')] self.assertEqual(scripts(original['template']), scripts(written['template'])) self.assertEqual(scripts(original['form']), scripts(written['form'])) self.assertEqual(original_hash, hashlib.sha256(blank.read_bytes()).hexdigest()) def test_packet_structure_namespaces_and_processing_instructions_are_preserved(self): xml = b'' fields = {'ContactInformation.q3-4.Phone.CanOtherInd.CanadaUS': '1', 'ContactInformation.q3-4.Phone.CanOtherInd.Other': '0'} expected = xml.replace(b'name="NANumber" presence="invisible"', b'name="NANumber" presence="visible"').replace(b'name="IntlNumber"/>', b'name="IntlNumber" presence="invisible"/>') self.assertEqual(expected, contact_phone_state(xml, fields)) def test_phone_branches_match_selected_country_and_keep_official_scripts(self): for lang, suffix in [('en', 'e'), ('fr', 'f')]: for primary_nanp, alternate_nanp in [(True, False), (False, True)]: with self.subTest(lang=lang, primary_nanp=primary_nanp): fields = {} for group, nanp in [('Phone', primary_nanp), ('AltPhone', alternate_nanp)]: prefix = 'ContactInformation.q3-4.' + group + '.' fields[prefix + 'CanOtherInd.CanadaUS'] = '1' if nanp else '0' fields[prefix + 'CanOtherInd.Other'] = '0' if nanp else '1' fields[prefix + 'NumberCountry'] = '1' if nanp else '86' blank = BASE / f'storage/app/imm/imm5708{suffix}.pdf' original = packets(blank) temp_root = BASE / 'storage/app/imm/tmp' temp_root.mkdir(parents=True, exist_ok=True) with tempfile.TemporaryDirectory(dir=temp_root, prefix='contact-state-') as temp: payload = Path(temp) / 'payload.json' output = Path(temp) / 'draft.pdf' payload.write_text(json.dumps({'form': 'IMM5708', 'fields': fields}), encoding='utf-8') fill(str(payload), str(output), str(blank)) written = packets(output) for packet in ['template', 'form']: nodes = named_nodes(written[packet]) for group, nanp in [('Phone', primary_nanp), ('AltPhone', alternate_nanp)]: prefix = 'form1.Page2.ContactInformation.q3-4.' + group + '.' self.assertEqual('visible' if nanp else 'invisible', nodes[prefix + 'NANumber'].get('presence', 'visible')) self.assertEqual('invisible' if nanp else 'visible', nodes[prefix + 'IntlNumber'].get('presence', 'visible')) scripts = lambda root: [n.text for n in root.iter() if n.tag.endswith('script')] self.assertEqual(scripts(original['template']), scripts(written['template'])) # No answer guessed for missing residential-address confirmation. self.assertFalse(any(n.text in ['Y', 'N'] for n in written['datasets'].iter() if n.tag.endswith('SameAsMailingInd'))) if __name__ == '__main__': unittest.main()