"""IMM1344-only incremental XFA writer. Original signed template stays unchanged.""" from pathlib import Path import hashlib,json,os from lxml import etree as E def replace_answer_data(raw, datasets): """Change only xfa:data, preserving official lookup bytes and XML framing.""" from xml.parsers import expat namespace='http://www.xfa.org/schema/xfa-data/1.0/' parser=expat.ParserCreate(namespace_separator='}') span=[] depth=0 active=None def start(tag,attrs): nonlocal depth,active depth+=1 if tag==namespace+'}data': if span:raise ValueError('Ambiguous IMM1344 data packet') span.append(parser.CurrentByteIndex);active=depth def end(tag): nonlocal depth if active==depth and tag==namespace+'}data': span.append(raw.index(b'>',parser.CurrentByteIndex)+1) depth-=1 parser.StartElementHandler=start;parser.EndElementHandler=end parser.Parse(raw,True) nodes=datasets.xpath('./*[local-name()="data"]') if len(span)!=2 or len(nodes)!=1:raise ValueError('Missing IMM1344 data packet') return raw[:span[0]]+E.tostring(nodes[0],encoding='utf-8',with_tail=False)+raw[span[1]:] def resolve(root,path,kinds=None,ns=None): node=root if kinds is None: matches=root.xpath('./*[local-name()="data"]') if len(matches)!=1:raise ValueError('Missing IMM1344 data packet') node=matches[0] for i,name in enumerate(path.split('.')): found=[n for n in node if isinstance(n.tag,str) and (n.get('name') if kinds else E.QName(n).localname)==name] if len(found)>1:raise ValueError('Ambiguous IMM1344 path: '+path) if not found: if not kinds:raise ValueError('Unresolved IMM1344 dataset: '+path) found=[E.SubElement(node,'{'+ns+'}'+kinds[i],name=name)] node=found[0] return node def restore_on_open(form,template,contract,fields,ns): """Keep saved UI state and restore postal answers cleared by initial country merge. Always run the official saved-delta restore before optional hydration. Adobe defers saved access/presence changes until this restore; skipping it leaves North American country code editable and incorrectly validates code 1. Template and all validation scripts remain untouched. """ if not fields:return root=resolve(form,'form1',['subform'],ns) original=E.fromstring(template).xpath('//*[local-name()="subform" and @name="form1"]')[0] postals=[(path,value) for path,value in fields.items() if path.endswith('.PostalCode')] if postals: initial=original.xpath('./*[local-name()="event" and @activity="initialize"]')[0] event=E.SubElement(root,'{'+ns+'}event',**dict(initial.attrib)) capture=[''.join(initial.xpath('./*[local-name()="script"]/text()'))] for i,(path,value) in enumerate(postals): country=path.rsplit('.',1)[0]+'.Country' capture.append('var state=xfa.datasets.data.form1.imm1344PrefillState; var d=xfa.datasets.data.resolveNode('+json.dumps(path)+'); var c=xfa.datasets.data.resolveNode('+json.dumps(country)+'); var postal=d && d.value ? String(d.value) : ""; var country=c && c.value ? String(c.value) : ""; state.resolveNode("postal'+str(i)+'").value=postal; state.resolveNode("country'+str(i)+'").value=country; state.resolveNode("armed'+str(i)+'").value=postal ? "1" : "0"; var countries=xfa.resolveNodes("xfa.datasets.LOVFile.LOV.CountryOfBirthList.CountryOfBirth[*]"); var label=""; for(var j=0;jm['max_chars']:raise ValueError('Invalid IMM1344 field value: '+path) if m['choices'] and value not in m['choices']:raise ValueError('Invalid IMM1344 export: '+path) if os.name=='nt' and not os.environ.get('SystemRoot'): import ctypes buffer=ctypes.create_unicode_buffer(32768) if not ctypes.windll.kernel32.GetWindowsDirectoryW(buffer,len(buffer)):raise OSError('Cannot resolve Windows directory') os.environ['SystemRoot']=buffer.value from pyhanko.pdf_utils.incremental_writer import IncrementalPdfFileWriter from pyhanko.pdf_utils import generic from imm0008_xfa_checksum import form_checksum out.parent.mkdir(parents=True,exist_ok=True) with blank.open('rb') as source: writer=IncrementalPdfFileWriter(source,strict=False) if writer.prev.encrypted:writer.encrypt('') x=writer.root['/AcroForm']['/XFA'];pk={str(x[i]):x[i+1].data for i in range(0,len(x),2)} parser=E.XMLParser(remove_comments=False,remove_pis=False,resolve_entities=False) data=E.fromstring(pk['datasets'],parser);form=E.fromstring(pk['form'],parser);ns=c['form_namespace'] for path,value in fields.items(): m=c['fields'][path];resolve(data,path).text=value widget=resolve(form,path,m['kinds'],ns) # Ordinary answers bind from datasets; phone UI values are saved below # together with access/presence for the template's manual delta restore. for old in widget.xpath('./*[local-name()="value"]'):widget.remove(old) if not path.endswith('.ActualNumber'):widget.set('access','open') if fields: state=E.SubElement(resolve(data,'form1'),'imm1344PrefillState');E.SubElement(state,'active').text='1' for i,path in enumerate(path for path in fields if path.endswith('.PostalCode')): for name in ['postal','country','label','armed']:E.SubElement(state,name+str(i)).text='0' def sub(path):return resolve(form,path,['subform']*len(path.split('.')),ns) co=fields.get('form1.Part1.SponsorDetails.genDetails.q3.CoSignerInd') if co is not None:sub('form1.Part1.CoSigner').set('presence','visible' if co=='Y' else 'hidden') for key,m in c['answers'].items(): if m['type']!='phone':continue p=m['phone'];f=p['fields'];north=fields.get(f['CanadaUS']);other=fields.get(f['Other']) if north is None and other is None:continue for name,path in f.items(): if path not in fields or name=='ActualNumber':continue meta=c['fields'][path];widget=resolve(form,path,meta['kinds'],ns) val=E.SubElement(widget,'{'+ns+'}value') E.SubElement(val,'{'+ns+'}'+meta['value_type']).text=fields[path] resolve(form,f['NumberCountry'],c['fields'][f['NumberCountry']]['kinds'],ns).set('access','readOnly' if north=='1' else 'open') sub(p['root']+'.NANumber').set('presence','visible' if north=='1' else 'invisible') sub(p['root']+'.IntlNumber').set('presence','invisible' if north=='1' else 'visible') # Saved UI state is explicit; never invoke click events which erase digits. for name in ['NANumber.AreaCode','NANumber.FirstThree','NANumber.LastFive','IntlNumber.IntlNumber']: resolve(form,f[name],c['fields'][f[name]]['kinds'],ns).set('access','open') for address in ['sponsor_mailing','sponsor_residential']: country=c['answers'][address+'.country']['paths'][0];code=fields.get(country) if code is None:continue province=c['answers'][address+'.province']['paths'][0];widget=resolve(form,province,c['fields'][province]['kinds'],ns) widget.set('access','open' if code in ['511','461'] else 'readOnly') lookup='StateAbbrevList.StateAbbrev[*]' if code=='461' else 'ProvinceAbbrevList.ProvinceAbbrev[*]' for old in widget.xpath('./*[local-name()="bindItems"]'):widget.remove(old) E.SubElement(widget,'{'+ns+'}bindItems',ref='xfa.datasets.LOVFile.LOV.'+lookup,labelRef='$',valueRef='lic') for n in list(form.xpath('//*[local-name()="subform"]')): parent=n.getparent();name=n.get('name') if E.QName(parent).localname=='form' or not name:continue if not any(E.QName(ch).localname=='instanceManager' and ch.get('name')=='_'+name for ch in parent if isinstance(ch.tag,str)): parent.insert(parent.index(n),E.Element('{'+ns+'}instanceManager',name='_'+name)) # Verify every submitted target before appending the incremental revision. if any(resolve(data,path).text!=value for path,value in fields.items()):raise ValueError('IMM1344 payload/XML mismatch') restore_on_open(form,pk['template'],c,fields,ns) datasets=replace_answer_data(pk['datasets'],data);form.set('checksum',form_checksum(pk['template'],datasets)) updates={'datasets':datasets,'form':E.tostring(form)} for i in range(0,len(x),2): name=str(x[i]) if name in updates: ref=x.raw_get(i+1);stream=generic.StreamObject(stream_data=updates[name]);stream.compress();writer.objects[(ref.generation,ref.idnum)]=stream with out.open('wb') as target:writer.write(target) return out