icuSources/data/buildtool/comment_stripper.py

   1 # Copyright (C) 2018 and later: Unicode, Inc. and others.
   2 # License & terms of use: http://www.unicode.org/copyright.html
   3
   4 import io
   5
   6 class CommentStripper(object):
   7     """Removes lines starting with "//" from a file stream."""
   8
   9     def __init__(self, f):
  10         self.f = f
  11         self.state = 0
  12
  13     def read(self, size=-1):
  14         bytes = self.f.read(size)
  15         # TODO: Do we need to read more bytes if comments were stripped
  16         # in order to obey the size request?
  17         return "".join(self._strip_comments(bytes))
  18
  19     def _strip_comments(self, bytes):
  20         for byte in bytes:
  21             if self.state == 0:
  22                 # state 0: start of a line
  23                 if byte == "/":
  24                     self.state = 1
  25                 elif byte == "\n":
  26                     self.state = 0
  27                     yield byte
  28                 else:
  29                     self.state = 2
  30                     yield byte
  31             elif self.state == 1:
  32                 # state 1: read a single '/'
  33                 if byte == "/":
  34                     self.state = 3
  35                 elif byte == "\n":
  36                     self.state = 0
  37                     yield "/"  # the one that was skipped
  38                     yield "\n"
  39                 else:
  40                     self.state = 2
  41                     yield "/"  # the one that was skipped
  42                     yield byte
  43             elif self.state == 2:
  44                 # state 2: middle of a line, no comment
  45                 if byte == "\n":
  46                     self.state = 0
  47                 yield byte
  48             elif self.state == 3:
  49                 # state 3: inside a comment
  50                 if byte == "\n":
  51                     self.state = 0