[Python-checkins] CVS: python/dist/src/Lib shlex.py,1.5,1.6

Guido van Rossum python-dev@python.org
Mon, 1 May 2000 16:08:49 -0400 (EDT)


Update of /projects/cvsroot/python/dist/src/Lib
In directory eric:/projects/python/develop/guido/src/Lib

Modified Files:
	shlex.py 
Log Message:
Eric Raymond:

Added and documented the capability for shlex to handle lexical-level
inclusion and a stack of input sources.  Also, the input stream member
is now documented, and the constructor takes an optional source-filename.
The class provides facilities to generate error messages that track
file and line number.

[GvR: I changed the __main__ code so that it actually stops at EOF, as
Eric surely intended -- however it returned '' instead of the None he
was testing for.]


Index: shlex.py
===================================================================
RCS file: /projects/cvsroot/python/dist/src/Lib/shlex.py,v
retrieving revision 1.5
retrieving revision 1.6
diff -C2 -r1.5 -r1.6
*** shlex.py	2000/02/04 15:28:40	1.5
--- shlex.py	2000/05/01 20:08:46	1.6
***************
*** 2,5 ****
--- 2,6 ----
  
  # Module and documentation by Eric S. Raymond, 21 Dec 1998 
+ # Input stacking and error message cleanup added by ESR, March 2000
  
  import sys
***************
*** 7,15 ****
  class shlex:
      "A lexical analyzer class for simple shell-like syntaxes." 
!     def __init__(self, instream=None):
          if instream:
              self.instream = instream
          else:
              self.instream = sys.stdin
          self.commenters = '#'
          self.wordchars = 'abcdfeghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789_'
--- 8,18 ----
  class shlex:
      "A lexical analyzer class for simple shell-like syntaxes." 
!     def __init__(self, instream=None, infile=None):
          if instream:
              self.instream = instream
+             self.infile = infile
          else:
              self.instream = sys.stdin
+             self.infile = None
          self.commenters = '#'
          self.wordchars = 'abcdfeghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789_'
***************
*** 21,39 ****
          self.debug = 0
          self.token = ''
  
      def push_token(self, tok):
          "Push a token onto the stack popped by the get_token method"
!         if (self.debug >= 1):
!             print "Pushing " + tok
          self.pushback = [tok] + self.pushback;
  
      def get_token(self):
!         "Get a token from the input stream (or from stack if it's monempty)"
          if self.pushback:
              tok = self.pushback[0]
              self.pushback = self.pushback[1:]
!             if (self.debug >= 1):
!                 print "Popping " + tok
              return tok
          tok = ''
          while 1:
--- 24,80 ----
          self.debug = 0
          self.token = ''
+ 	self.filestack = []
+         self.source = None
+         if self.debug:
+             print 'shlex: reading from %s, line %d' % (self.instream,self.lineno)
  
      def push_token(self, tok):
          "Push a token onto the stack popped by the get_token method"
!         if self.debug >= 1:
!             print "shlex: pushing token " + `tok`
          self.pushback = [tok] + self.pushback;
  
      def get_token(self):
!         "Get a token from the input stream (or from stack if it's nonempty)"
          if self.pushback:
              tok = self.pushback[0]
              self.pushback = self.pushback[1:]
!             if self.debug >= 1:
!                 print "shlex: popping token " + `tok`
              return tok
+ 	# No pushback.  Get a token.
+         raw = self.read_token()
+         # Handle inclusions
+         while raw == self.source:
+             (newfile, newstream) = self.sourcehook(self.read_token())
+             self.filestack = [(self.infile,self.instream,self.lineno)] + self.filestack
+             self.infile = newfile
+             self.instream = newstream
+             self.lineno = 1
+             if self.debug:
+                 print 'shlex: pushing to file %s' % (self.infile,)
+             raw = self.get_token()
+         # Maybe we got EOF instead?
+         while raw == "":
+             if len(self.filestack) == 0:
+                 return ""
+             else:
+                 self.instream.close()
+                 (self.infile, self.instream, self.lineno) = self.filestack[0]
+                 self.filestack = self.filestack[1:]
+                 if self.debug:
+                     print 'shlex: popping to %s, line %d' % (self.instream, self.lineno)
+                 self.state = ' '
+                 raw = self.get_token()
+          # Neither inclusion nor EOF
+         if self.debug >= 1:
+             if raw:
+                 print "shlex: token=" + `raw`
+             else:
+                 print "shlex: token=EOF"
+         return raw
+ 
+     def read_token(self):
+         "Read a token from the input stream (no pushback or inclusions)"
          tok = ''
          while 1:
***************
*** 42,48 ****
                  self.lineno = self.lineno + 1
              if self.debug >= 3:
!                 print "In state " + repr(self.state) + " I see character: " + repr(nextchar) 
              if self.state == None:
!                 return ''
              elif self.state == ' ':
                  if not nextchar:
--- 83,90 ----
                  self.lineno = self.lineno + 1
              if self.debug >= 3:
!                 print "shlex: in state " + repr(self.state) + " I see character: " + repr(nextchar) 
              if self.state == None:
!                 self.token = '';	# past end of file
!                 break
              elif self.state == ' ':
                  if not nextchar:
***************
*** 51,55 ****
                  elif nextchar in self.whitespace:
                      if self.debug >= 2:
!                         print "I see whitespace in whitespace state"
                      if self.token:
                          break	# emit current token
--- 93,97 ----
                  elif nextchar in self.whitespace:
                      if self.debug >= 2:
!                         print "shlex: I see whitespace in whitespace state"
                      if self.token:
                          break	# emit current token
***************
*** 82,86 ****
                  elif nextchar in self.whitespace:
                      if self.debug >= 2:
!                         print "I see whitespace in word state"
                      self.state = ' '
                      if self.token:
--- 124,128 ----
                  elif nextchar in self.whitespace:
                      if self.debug >= 2:
!                         print "shlex: I see whitespace in word state"
                      self.state = ' '
                      if self.token:
***************
*** 96,100 ****
                      self.pushback = [nextchar] + self.pushback
                      if self.debug >= 2:
!                         print "I see punctuation in word state"
                      self.state = ' '
                      if self.token:
--- 138,142 ----
                      self.pushback = [nextchar] + self.pushback
                      if self.debug >= 2:
!                         print "shlex: I see punctuation in word state"
                      self.state = ' '
                      if self.token:
***************
*** 102,112 ****
                      else:
                          continue
-                 
          result = self.token
          self.token = ''
!         if self.debug >= 1:
!             print "Token: " + result
          return result
  
  if __name__ == '__main__': 
  
--- 144,162 ----
                      else:
                          continue
          result = self.token
          self.token = ''
!         if self.debug > 1:
!             if result:
!                 print "shlex: raw token=" + `result`
!             else:
!                 print "shlex: raw token=EOF"
          return result
  
+     def sourcehook(self, newfile):
+         "Hook called on a filename to be sourced."
+         if newfile[0] == '"':
+             newfile = newfile[1:-1]
+         return (newfile, open(newfile, "r"))
+ 
  if __name__ == '__main__': 
  
***************
*** 114,120 ****
      while 1:
          tt = lexer.get_token()
!         if tt != None:
!             print "Token: " + repr(tt)
!         else:
              break
  
--- 164,169 ----
      while 1:
          tt = lexer.get_token()
!         print "Token: " + repr(tt)
!         if not tt:
              break