]> git.madduck.net Git - etc/vim.git/blobdiff - blib2to3/pgen2/tokenize.py

madduck's git repository

Every one of the projects in this repository is available at the canonical URL git://git.madduck.net/madduck/pub/<projectpath> — see each project's metadata for the exact URL.

All patches and comments are welcome. Please squash your changes to logical commits before using git-format-patch and git-send-email to patches@git.madduck.net. If you'd read over the Git project's submission guidelines and adhered to them, I'd be especially grateful.

SSH access, as well as push access can be individually arranged.

If you use my repositories frequently, consider adding the following snippet to ~/.gitconfig and using the third clone URL listed for each project:

[url "git://git.madduck.net/madduck/"]
  insteadOf = madduck:

Teach the Vim plugin to respect pyproject.toml (issue 414) (#1273)
[etc/vim.git] / blib2to3 / pgen2 / tokenize.py
index 8c1c4cfac680a2f4815d880c989d6e3b3d115bb0..26835702eb79ffbaf19889b70cb1b4118313d8a7 100644 (file)
@@ -1,6 +1,8 @@
 # Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006 Python Software Foundation.
 # All rights reserved.
 
+# mypy: allow-untyped-defs, allow-untyped-calls
+
 """Tokenization help for Python programs.
 
 generate_tokens(readline) is a generator that breaks a stream of
@@ -25,6 +27,21 @@ are the same, except instead of generating tokens, tokeneater is a callback
 function to which the 5 fields described above are passed as 5 arguments,
 each time a new token is found."""
 
+from typing import (
+    Callable,
+    Iterable,
+    Iterator,
+    List,
+    Optional,
+    Text,
+    Tuple,
+    Pattern,
+    Union,
+    cast,
+)
+from blib2to3.pgen2.token import *
+from blib2to3.pgen2.grammar import Grammar
+
 __author__ = "Ka-Ping Yee <ping@lfw.org>"
 __credits__ = "GvR, ESR, Tim Peters, Thomas Wouters, Fred Drake, Skip Montanaro"
 
@@ -41,13 +58,6 @@ __all__ = [x for x in dir(token) if x[0] != "_"] + [
 ]
 del token
 
-try:
-    bytes
-except NameError:
-    # Support bytes type in Python <= 2.5, so 2to3 turns itself into
-    # valid Python 3 code.
-    bytes = str
-
 
 def group(*choices):
     return "(" + "|".join(choices) + ")"
@@ -119,9 +129,6 @@ Bracket = "[][(){}]"
 Special = group(r"\r?\n", r"[:;.,`@]")
 Funny = group(Operator, Bracket, Special)
 
-PlainToken = group(Number, Funny, String, Name)
-Token = Ignore + PlainToken
-
 # First (or only) line of ' or " string.
 ContStr = group(
     _litprefix + r"'[^\n'\\]*(?:\\.[^\n'\\]*)*" + group("'", r"\\\r?\n"),
@@ -130,7 +137,6 @@ ContStr = group(
 PseudoExtras = group(r"\\\r?\n", Comment, Triple)
 PseudoToken = Whitespace + group(PseudoExtras, Number, Funny, ContStr, Name)
 
-tokenprog = re.compile(Token, re.UNICODE)
 pseudoprog = re.compile(PseudoToken, re.UNICODE)
 single3prog = re.compile(Single3)
 double3prog = re.compile(Double3)
@@ -181,7 +187,11 @@ def printtoken(type, token, xxx_todo_changeme, xxx_todo_changeme1, line):  # for
     )
 
 
-def tokenize(readline, tokeneater=printtoken):
+Coord = Tuple[int, int]
+TokenEater = Callable[[int, Text, Coord, Coord, Text], None]
+
+
+def tokenize(readline: Callable[[], Text], tokeneater: TokenEater = printtoken) -> None:
     """
     The tokenize() function accepts two parameters: one representing the
     input stream, and one providing an output mechanism for tokenize().
@@ -206,25 +216,36 @@ def tokenize_loop(readline, tokeneater):
         tokeneater(*token_info)
 
 
+GoodTokenInfo = Tuple[int, Text, Coord, Coord, Text]
+TokenInfo = Union[Tuple[int, str], GoodTokenInfo]
+
+
 class Untokenizer:
-    def __init__(self):
+
+    tokens: List[Text]
+    prev_row: int
+    prev_col: int
+
+    def __init__(self) -> None:
         self.tokens = []
         self.prev_row = 1
         self.prev_col = 0
 
-    def add_whitespace(self, start):
+    def add_whitespace(self, start: Coord) -> None:
         row, col = start
         assert row <= self.prev_row
         col_offset = col - self.prev_col
         if col_offset:
             self.tokens.append(" " * col_offset)
 
-    def untokenize(self, iterable):
+    def untokenize(self, iterable: Iterable[TokenInfo]) -> Text:
         for t in iterable:
             if len(t) == 2:
-                self.compat(t, iterable)
+                self.compat(cast(Tuple[int, str], t), iterable)
                 break
-            tok_type, token, start, end, line = t
+            tok_type, token, start, end, line = cast(
+                Tuple[int, Text, Coord, Coord, Text], t
+            )
             self.add_whitespace(start)
             self.tokens.append(token)
             self.prev_row, self.prev_col = end
@@ -233,7 +254,7 @@ class Untokenizer:
                 self.prev_col = 0
         return "".join(self.tokens)
 
-    def compat(self, token, iterable):
+    def compat(self, token: Tuple[int, Text], iterable: Iterable[TokenInfo]) -> None:
         startline = False
         indents = []
         toks_append = self.tokens.append
@@ -266,7 +287,7 @@ cookie_re = re.compile(r"^[ \t\f]*#.*?coding[:=][ \t]*([-\w.]+)", re.ASCII)
 blank_re = re.compile(br"^[ \t\f]*(?:[#\r\n]|$)", re.ASCII)
 
 
-def _get_normal_name(orig_enc):
+def _get_normal_name(orig_enc: str) -> str:
     """Imitates get_normal_name in tokenizer.c."""
     # Only care about the first 12 characters.
     enc = orig_enc[:12].lower().replace("_", "-")
@@ -279,7 +300,7 @@ def _get_normal_name(orig_enc):
     return orig_enc
 
 
-def detect_encoding(readline):
+def detect_encoding(readline: Callable[[], bytes]) -> Tuple[str, List[bytes]]:
     """
     The detect_encoding() function is used to detect the encoding that should
     be used to decode a Python source file. It requires one argument, readline,
@@ -301,13 +322,13 @@ def detect_encoding(readline):
     encoding = None
     default = "utf-8"
 
-    def read_or_stop():
+    def read_or_stop() -> bytes:
         try:
             return readline()
         except StopIteration:
             return bytes()
 
-    def find_cookie(line):
+    def find_cookie(line: bytes) -> Optional[str]:
         try:
             line_string = line.decode("ascii")
         except UnicodeDecodeError:
@@ -354,7 +375,7 @@ def detect_encoding(readline):
     return default, [first, second]
 
 
-def untokenize(iterable):
+def untokenize(iterable: Iterable[TokenInfo]) -> Text:
     """Transform tokens back into Python source code.
 
     Each element returned by the iterable must be a token sequence
@@ -364,7 +385,7 @@ def untokenize(iterable):
     Round-trip invariant for full input:
         Untokenized source will match input source exactly
 
-    Round-trip invariant for limited intput:
+    Round-trip invariant for limited input:
         # Output text will tokenize the back to the input
         t1 = [tok[:2] for tok in generate_tokens(f.readline)]
         newcode = untokenize(t1)
@@ -376,7 +397,9 @@ def untokenize(iterable):
     return ut.untokenize(iterable)
 
 
-def generate_tokens(readline, grammar=None):
+def generate_tokens(
+    readline: Callable[[], Text], grammar: Optional[Grammar] = None
+) -> Iterator[GoodTokenInfo]:
     """
     The generate_tokens() generator requires one argument, readline, which
     must be a callable object which provides the same interface as the
@@ -395,7 +418,7 @@ def generate_tokens(readline, grammar=None):
     lnum = parenlev = continued = 0
     numchars = "0123456789"
     contstr, needcont = "", 0
-    contline = None
+    contline: Optional[str] = None
     indents = [0]
 
     # If we know we're parsing 3.7+, we can unconditionally parse `async` and
@@ -407,6 +430,9 @@ def generate_tokens(readline, grammar=None):
     async_def_indent = 0
     async_def_nl = False
 
+    strstart: Tuple[int, int]
+    endprog: Pattern[str]
+
     while 1:  # loop over lines in stream
         try:
             line = readline()
@@ -416,6 +442,7 @@ def generate_tokens(readline, grammar=None):
         pos, max = 0, len(line)
 
         if contstr:  # continued string
+            assert contline is not None
             if not line:
                 raise TokenError("EOF in multi-line string", strstart)
             endmatch = endprog.match(line)