refactor!: nexttoken redundant for lexer

playa/pdfdocument.py

@@ -794,6 +794,7 @@ def __init__(
         if self.catalog.get("Type") is not LITERAL_CATALOG:
             if settings.STRICT:
                 raise PDFSyntaxError("Catalog not found!")
+        self.parser.seek(0)
 
     def _initialize_password(self, password: str = "") -> None:
         """Initialize the decryption handler with a given password, if any.
@@ -823,6 +824,10 @@ def _initialize_password(self, password: str = "") -> None:
         assert self.parser is not None
         self.parser.fallback = False  # need to read streams with exact length
 
+    def __iter__(self) -> Iterator[Tuple[int, object]]:
+        """Iterate over positions and top-level PDF objects in the file."""
+        return self.parser
+
     def _getobj_objstm(self, stream: PDFStream, index: int, objid: int) -> object:
         if stream.objid in self._parsed_objs:
             (objs, n) = self._parsed_objs[stream.objid]

playa/pdfparser.py

@@ -50,6 +50,7 @@ def do_keyword(self, pos: int, token: PSKeyword) -> None:
             self.add_results(*self.pop(1))
 
         elif token is KEYWORD_ENDOBJ:
+            # objid genno "obj" ... and the object itself
             self.add_results(*self.pop(4))
 
         elif token is KEYWORD_NULL:

playa/psparser.py

@@ -285,10 +285,6 @@ def __iter__(self) -> Iterator[Tuple[int, PSBaseParserToken]]:
         """Iterate over tokens."""
         return self
 
-    def nexttoken(self) -> Tuple[int, PSBaseParserToken]:
-        """Get the next token in iteration, raising StopIteration when done."""
-        return self.__next__()
-
     def __next__(self) -> Tuple[int, PSBaseParserToken]:
         """Get the next token in iteration, raising StopIteration when
         done."""
@@ -455,7 +451,7 @@ def flush(self) -> None:
         return
 
     def __next__(self) -> PSStackEntry[ExtraT]:
-        """Return the next object, returning StopIteration at EOF.
+        """Return the next object, raising StopIteration at EOF.
 
         Arrays and dictionaries are represented as Python lists and
         dictionaries.
@@ -575,4 +571,4 @@ def get_inline_data(self, target: bytes = b"EI") -> Tuple[int, bytes]:
     def nexttoken(self) -> Tuple[int, PSBaseParserToken]:
         """Get the next token in iteration, raising StopIteration when
         done."""
-        return self._lexer.__next__()
+        return next(self._lexer)

tests/test_pdfdocument.py

@@ -34,6 +34,12 @@ def test_read_header():
     assert read_header(BytesIO(b"%PDF-1.7")) == "1.7"
 
 
+def test_objects():
+    with playa.open(TESTDIR / "simple1.pdf") as doc:
+        for obj in doc:
+            print(obj)
+
+
 def test_page_labels():
     with playa.open(TESTDIR / "contrib" / "pagelabels.pdf") as doc:
         labels = [label for _, label in zip(range(10), doc.page_labels)]

tests/test_psparser.py

@@ -296,7 +296,7 @@ def inline_parsers(
     bp = Lexer(data)
     assert bp.get_inline_data(target=target, blocksize=blocksize) == expected
     if nexttoken is not None:
-        assert bp.nexttoken() == nexttoken
+        assert next(bp) == nexttoken
 
 
 def test_get_inline_data() -> None: