1    #! /usr/bin/env python
   2    # Time-stamp: <2001-10-17 22:05:33 cymbala>
   3    
   4    
   5    # SYNTAX:  python noncurlq.py FILENAME
   6    # ====================================
   7    
   8    # Here's an example:
   9    #
  10    # cymbala@debian:~$ python /floppy/noncurlq.py 04.htm
  11    # non-curly quote: 'industrialists"[71]:  \012    '
  12    # cymbala@debian:~$
  13    # =======================================================
  14    
  15    
  16    # Why?
  17    
  18    # FIND straight (non-curly) quotes in HTML data (such as
  19    # words quoted by the author)...
  20    #
  21    # ... while IGNORING straight quotes that are part of
  22    # mark-up (such as HREF values that are quoted inside an A
  23    # anchor).
  24    
  25    
  26    # Once located, you may manually change '"' to either
  27    # '&#8220;' or '&#8221;' depending on whether the straight
  28    # quote is an opener or a closer.  Likewise for "'" (and its
  29    # corresponding numeric values for open-left-single-quote
  30    # and                              open-right-single-quote.
  31    
  32    
  33    # Ideally, one would use "&ldquo;"   "&lsquo;"
  34    #                        "&rdquo;"   "&rdquo;"
  35    # ...instead of numeric values.
  36    # Symbols are better (more portable) than numbers.
  37    
  38    
  39    # Besides finding straight quotes, this script is an example
  40    # of how to use Python's sgmllib (SGML library) to look for
  41    # data inside an HTML file.  It's as simple as possible.  In
  42    # fact, most of this script was COPIED from sgmllib.py (located in
  43    # /usr/lib/python1.5).  All I had to do was make a few changes based
  44    # on a general understanding of how it works.
  45    
  46    
  47    # NOTE OF WARNING: Due to ``file = args[0]'' this script can
  48    # only process one file at a time.  This is dangerous,
  49    # because some people might ignore SYNTAX found above and
  50    # specify multiple file names for input... they would think
  51    # all files are being processed, whereas only the first
  52    # would be processed!!!!!  And that would be bad (or "evil"
  53    # to use a popular Bushism).
  54    
  55    
  56    
  57    
  58    # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  59    # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  60    # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  61    
  62    
  63    from sgmllib import *
  64    import sys
  65    
  66    
  67    # Copied from: /usr/lib/python1.5/sgmllib.py
  68    # (Added "pass" statements.)
  69    # -------------------------------------------------------
  70    def test(args = None):
  71        import sys
  72    
  73        if not args:
  74            args = sys.argv[1:]
  75            pass
  76    
  77        if args and args[0] == '-s':
  78            args = args[1:]
  79            klass = SGMLParser
  80            pass
  81        else:
  82            klass = TestSGMLParser
  83            pass
  84    
  85        if args:
  86            file = args[0]
  87            pass
  88        else:
  89            file = 'test.html'
  90            pass
  91    
  92        if file == '-':
  93            f = sys.stdin
  94            pass
  95        else:
  96            try:
  97                f = open(file, 'r')
  98                pass
  99            except IOError, msg:
 100                print file, ":", msg
 101                sys.exit(1)
 102                pass
 103            pass
 104    
 105        data = f.read()
 106        if f is not sys.stdin:
 107            f.close()
 108            pass
 109    
 110        x = klass()
 111        for c in data:
 112            x.feed(c)
 113            pass
 114        x.close()
 115        pass
 116    # -------------------------------------------------------
 117    
 118    
 119    
 120    
 121    # Copied from: /usr/lib/python1.5/sgmllib.py
 122    #   (Added "pass" statements.)
 123    #   (Commented most, if not all, print statements.)
 124    #
 125    #   (Added condition to look for '"' or '"' in data.)    <--- !
 126    #
 127    # =======================================================
 128    class TestSGMLParser(SGMLParser):
 129    
 130        def __init__(self, verbose=0):
 131            self.testdata = ""
 132            SGMLParser.__init__(self, verbose)
 133            pass
 134    
 135        def handle_data(self, data):
 136            self.testdata = self.testdata + data
 137            if len(`self.testdata`) >= 70:
 138                self.flush()
 139                pass
 140            pass
 141    
 142        def flush(self):
 143            data = self.testdata
 144            if data:
 145                self.testdata = ""
 146                # print 'data:', `data`
 147    
 148                # 2001.10.12 -RjC
 149                #
 150                if ((string.find(data, '"') > -1) or (string.find(data, "'") > -1)):
 151                    print 'non-curly quote:', `data`
 152                    pass
 153                
 154                pass
 155            pass
 156    
 157        def handle_comment(self, data):
 158            self.flush()
 159            r = `data`
 160            if len(r) > 68:
 161                r = r[:32] + '...' + r[-32:]
 162                pass
 163            # print 'comment:', r
 164            pass
 165    
 166        def unknown_starttag(self, tag, attrs):
 167            self.flush()
 168            if not attrs:
 169                # print 'start tag: <' + tag + '>'
 170                pass
 171            else:
 172                # print 'start tag: <' + tag,
 173                for name, value in attrs:
 174                    # print name + '=' + '"' + value + '"',
 175                    pass
 176                # print '>'
 177                pass
 178            pass
 179    
 180        def unknown_endtag(self, tag):
 181            self.flush()
 182            # print 'end tag: </' + tag + '>'
 183            pass
 184    
 185        def unknown_entityref(self, ref):
 186            self.flush()
 187            # print '*** unknown entity ref: &' + ref + ';'
 188            pass
 189    
 190        def unknown_charref(self, ref):
 191            self.flush()
 192            # print '*** unknown char ref: &#' + ref + ';'
 193            pass
 194    
 195        def close(self):
 196            SGMLParser.close(self)
 197            self.flush()
 198            pass
 199    # =======================================================
 200    
 201    
 202    
 203    test()
 204    
 205    ###
 206    #