fz0x1/HPI
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

smscalls: allow large XML files as input

Browse source 
once XML files increase past a certain size
(was about 220MB for me), the parser just throws
an error because the tree is too large (iirc for
security reasons)

could maybe look at using iterparse in the future
to parse it without loading the whole file, but this
seems to fix it fine for me
This commit is contained in:
purarue 2024-12-22 21:50:03 -08:00 • committed by karlicoss
parent d8c53bde34
commit f1d23c5e96
1 changed files with 7 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
my/smscalls.py
View file

@ -57,9 +57,12 @@ class Call(NamedTuple):
# The '(Unknown)' is just what my android phone does, not sure if there are others # The '(Unknown)' is just what my android phone does, not sure if there are others
UNKNOWN: set[str] = {'(Unknown)'} UNKNOWN: set[str] = {'(Unknown)'}
def _parse_xml(xml: Path) -> Any:
return etree.parse(str(xml), parser=etree.XMLParser(huge_tree=True))
def _extract_calls(path: Path) -> Iterator[Res[Call]]: def _extract_calls(path: Path) -> Iterator[Res[Call]]:
tr = etree.parse(str(path)) tr = _parse_xml(path)
for cxml in tr.findall('call'): for cxml in tr.findall('call'):
dt = cxml.get('date') dt = cxml.get('date')
dt_readable = cxml.get('readable_date') dt_readable = cxml.get('readable_date')
@ -133,7 +136,7 @@ def messages() -> Iterator[Res[Message]]:
def _extract_messages(path: Path) -> Iterator[Res[Message]]: def _extract_messages(path: Path) -> Iterator[Res[Message]]:
tr = etree.parse(str(path)) tr = _parse_xml(path)
for mxml in tr.findall('sms'): for mxml in tr.findall('sms'):
dt = mxml.get('date') dt = mxml.get('date')
dt_readable = mxml.get('readable_date') dt_readable = mxml.get('readable_date')
@ -225,8 +228,7 @@ def _resolve_null_str(value: str | None) -> str | None:
def _extract_mms(path: Path) -> Iterator[Res[MMS]]: def _extract_mms(path: Path) -> Iterator[Res[MMS]]:
tr = etree.parse(str(path)) tr = _parse_xml(path)
for mxml in tr.findall('mms'): for mxml in tr.findall('mms'):
dt = mxml.get('date') dt = mxml.get('date')
dt_readable = mxml.get('readable_date') dt_readable = mxml.get('readable_date')
@ -271,10 +273,7 @@ def _extract_mms(path: Path) -> Iterator[Res[MMS]]:
# #
# This seems pretty useless, so we should try and skip it, and just return the # This seems pretty useless, so we should try and skip it, and just return the
# text/images/data # text/images/data
# part_data: dict[str, Any] = part.attrib
# man, attrib is some internal cpython ._Attrib type which can't
# be typed by any sort of mappingproxy. maybe a protocol could work..?
part_data: dict[str, Any] = part.attrib # type: ignore
seq: str | None = part_data.get('seq') seq: str | None = part_data.get('seq')
if seq == '-1': if seq == '-1':
continue continue