ENH: Accept hierarchical fields

pypdf/_reader.py

@@ -585,6 +585,20 @@ def getFields(
  deprecation_with_replacement("getFields", "get_fields", "3.0.0")
  return self.get_fields(tree, retval, fileobj)
 
+ def _get_qualified_field_name(self, parent: DictionaryObject) -> str:
+ if "/TM" in parent:
+ return cast(str, parent["/TM"])
+ elif "/Parent" in parent:
+ return (
+ self._get_qualified_field_name(
+ cast(DictionaryObject, parent["/Parent"])
+ )
+ + "."
+ + cast(str, parent["/T"])
+ )
+ else:
+ return cast(str, parent["/T"])
+
  def _build_field(
  self,
  field: Union[TreeObject, DictionaryObject],
@@ -594,10 +608,19 @@ def _build_field(
  ) -> None:
  self._check_kids(field, retval, fileobj)
  try:
- key = field["/TM"]
+ key = cast(str, field["/TM"])
  except KeyError:
  try:
- key = field["/T"]
+ if "/Parent" in field:
+ key = (
+ self._get_qualified_field_name(
+ cast(DictionaryObject, field["/Parent"])
+ )
+ + "."
+ )
+ else:
+ key = ""
+ key += cast(str, field["/T"])
  except KeyError:
  # Ignore no-name field for now
  return
@@ -651,25 +674,48 @@ def _write_field(self, fileobj: Any, field: Any, field_attributes: Any) -> None:
  # Field attribute is N/A or unknown, so don't write anything
  pass
 
- def get_form_text_fields(self) -> Dict[str, Any]:
+ def get_form_text_fields(self, full_qualified_name: bool = False) -> Dict[str, Any]:
  """
  Retrieve form fields from the document with textual data.
 
  The key is the name of the form field, the value is the content of the
  field.
 
  If the document contains multiple form fields with the same name, the
- second and following will get the suffix _2, _3, ...
+ second and following will get the suffix .2, .3, ...
+
+ full_qualified_name should be used to get full name
  """
+
+ def indexed_key(k: str, fields: dict) -> str:
+ if k not in fields:
+ return k
+ else:
+ return (
+ k
+ + "."
+ + str(sum([1 for kk in fields if kk.startswith(k + ".")]) + 2)
+ )
+
  # Retrieve document form fields
  formfields = self.get_fields()
  if formfields is None:
  return {}
- return {
- formfields[field]["/T"]: formfields[field].get("/V")
+ ff = {}
+ for field, value in formfields.items():
+ if value.get("/FT") == "/Tx":
+ if full_qualified_name:
+ ff[field] = value.get("/V")
+ else:
+ ff[indexed_key(cast(str, value["/T"]), ff)] = value.get("/V")
+ return ff
+ """return {
+ (field if full_qualified_name else formfields[field]["/T"]): formfields[
+ field
+ ].get("/V")
  for field in formfields
  if formfields[field].get("/FT") == "/Tx"
- }
+ }"""
 
  def getFormTextFields(self) -> Dict[str, Any]: # deprecated
  """

pypdf/generic/_data_structures.py

@@ -1138,12 +1138,13 @@ class Field(TreeObject):
  :meth:`get_fields()<pypdf.PdfReader.get_fields>`
  """
 
- def __init__(self, data: Dict[str, Any]) -> None:
+ def __init__(self, data: DictionaryObject) -> None:
  DictionaryObject.__init__(self)
  field_attributes = (
  FieldDictionaryAttributes.attributes()
  + CheckboxRadioButtonAttributes.attributes()
  )
+ self.indirect_reference = data.indirect_reference
  for attr in field_attributes:
  try:
  self[NameObject(attr)] = data[attr]

tests/test_reader.py

@@ -866,6 +866,25 @@ def test_get_fields():
  assert dict(fields["c1-1"]) == ({"/FT": "/Btn", "/T": "c1-1"})
 
 
+def test_get_full_qualified_fields():
+ url = "https://github.com/py-pdf/PyPDF2/files/10142389/fields_with_dots.pdf"
+ name = "fields_with_dots.pdf"
+ reader = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
+ fields = reader.get_form_text_fields(True)
+ assert fields is not None
+ assert "customer.name" in fields
+
+ fields = reader.get_form_text_fields(False)
+ assert fields is not None
+ assert "customer.name" not in fields
+ assert "name" in fields
+
+ fields = reader.get_fields(True)
+ assert fields is not None
+ assert "customer.name" in fields
+ assert fields["customer.name"]["/T"] == "name"
+
+
 @pytest.mark.external
 @pytest.mark.filterwarnings("ignore::pypdf.errors.PdfReadWarning")
 def test_get_fields_read_else_block():