Georg Brandl | 1f7fffb | 2010-10-15 15:57:45 +0000 | [diff] [blame] | 1 | """ |
| 2 | General functions for HTML manipulation. |
| 3 | """ |
| 4 | |
Ezio Melotti | 4a9ee26 | 2013-11-19 20:28:45 +0200 | [diff] [blame] | 5 | import re as _re |
| 6 | from html.entities import html5 as _html5 |
| 7 | |
| 8 | |
| 9 | __all__ = ['escape', 'unescape'] |
| 10 | |
Georg Brandl | 1f7fffb | 2010-10-15 15:57:45 +0000 | [diff] [blame] | 11 | |
| 12 | def escape(s, quote=True): |
| 13 | """ |
| 14 | Replace special characters "&", "<" and ">" to HTML-safe sequences. |
| 15 | If the optional flag quote is true (the default), the quotation mark |
Senthil Kumaran | d71bbf9 | 2011-09-13 07:14:13 +0800 | [diff] [blame] | 16 | characters, both double quote (") and single quote (') characters are also |
| 17 | translated. |
Georg Brandl | 1f7fffb | 2010-10-15 15:57:45 +0000 | [diff] [blame] | 18 | """ |
Ezio Melotti | 4603487 | 2013-07-07 11:11:24 +0200 | [diff] [blame] | 19 | s = s.replace("&", "&") # Must be done first! |
| 20 | s = s.replace("<", "<") |
| 21 | s = s.replace(">", ">") |
Georg Brandl | 1f7fffb | 2010-10-15 15:57:45 +0000 | [diff] [blame] | 22 | if quote: |
Ezio Melotti | 4603487 | 2013-07-07 11:11:24 +0200 | [diff] [blame] | 23 | s = s.replace('"', """) |
| 24 | s = s.replace('\'', "'") |
| 25 | return s |
Ezio Melotti | 4a9ee26 | 2013-11-19 20:28:45 +0200 | [diff] [blame] | 26 | |
| 27 | |
| 28 | # see http://www.w3.org/TR/html5/syntax.html#tokenizing-character-references |
| 29 | |
| 30 | _invalid_charrefs = { |
| 31 | 0x00: '\ufffd', # REPLACEMENT CHARACTER |
| 32 | 0x0d: '\r', # CARRIAGE RETURN |
| 33 | 0x80: '\u20ac', # EURO SIGN |
| 34 | 0x81: '\x81', # <control> |
| 35 | 0x82: '\u201a', # SINGLE LOW-9 QUOTATION MARK |
| 36 | 0x83: '\u0192', # LATIN SMALL LETTER F WITH HOOK |
| 37 | 0x84: '\u201e', # DOUBLE LOW-9 QUOTATION MARK |
| 38 | 0x85: '\u2026', # HORIZONTAL ELLIPSIS |
| 39 | 0x86: '\u2020', # DAGGER |
| 40 | 0x87: '\u2021', # DOUBLE DAGGER |
| 41 | 0x88: '\u02c6', # MODIFIER LETTER CIRCUMFLEX ACCENT |
| 42 | 0x89: '\u2030', # PER MILLE SIGN |
| 43 | 0x8a: '\u0160', # LATIN CAPITAL LETTER S WITH CARON |
| 44 | 0x8b: '\u2039', # SINGLE LEFT-POINTING ANGLE QUOTATION MARK |
| 45 | 0x8c: '\u0152', # LATIN CAPITAL LIGATURE OE |
| 46 | 0x8d: '\x8d', # <control> |
| 47 | 0x8e: '\u017d', # LATIN CAPITAL LETTER Z WITH CARON |
| 48 | 0x8f: '\x8f', # <control> |
| 49 | 0x90: '\x90', # <control> |
| 50 | 0x91: '\u2018', # LEFT SINGLE QUOTATION MARK |
| 51 | 0x92: '\u2019', # RIGHT SINGLE QUOTATION MARK |
| 52 | 0x93: '\u201c', # LEFT DOUBLE QUOTATION MARK |
| 53 | 0x94: '\u201d', # RIGHT DOUBLE QUOTATION MARK |
| 54 | 0x95: '\u2022', # BULLET |
| 55 | 0x96: '\u2013', # EN DASH |
| 56 | 0x97: '\u2014', # EM DASH |
| 57 | 0x98: '\u02dc', # SMALL TILDE |
| 58 | 0x99: '\u2122', # TRADE MARK SIGN |
| 59 | 0x9a: '\u0161', # LATIN SMALL LETTER S WITH CARON |
| 60 | 0x9b: '\u203a', # SINGLE RIGHT-POINTING ANGLE QUOTATION MARK |
| 61 | 0x9c: '\u0153', # LATIN SMALL LIGATURE OE |
| 62 | 0x9d: '\x9d', # <control> |
| 63 | 0x9e: '\u017e', # LATIN SMALL LETTER Z WITH CARON |
| 64 | 0x9f: '\u0178', # LATIN CAPITAL LETTER Y WITH DIAERESIS |
| 65 | } |
| 66 | |
| 67 | _invalid_codepoints = { |
| 68 | # 0x0001 to 0x0008 |
| 69 | 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, 0x8, |
| 70 | # 0x000E to 0x001F |
| 71 | 0xe, 0xf, 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, |
| 72 | 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, |
| 73 | # 0x007F to 0x009F |
| 74 | 0x7f, 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, |
| 75 | 0x8b, 0x8c, 0x8d, 0x8e, 0x8f, 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, |
| 76 | 0x97, 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, |
| 77 | # 0xFDD0 to 0xFDEF |
| 78 | 0xfdd0, 0xfdd1, 0xfdd2, 0xfdd3, 0xfdd4, 0xfdd5, 0xfdd6, 0xfdd7, 0xfdd8, |
| 79 | 0xfdd9, 0xfdda, 0xfddb, 0xfddc, 0xfddd, 0xfdde, 0xfddf, 0xfde0, 0xfde1, |
| 80 | 0xfde2, 0xfde3, 0xfde4, 0xfde5, 0xfde6, 0xfde7, 0xfde8, 0xfde9, 0xfdea, |
| 81 | 0xfdeb, 0xfdec, 0xfded, 0xfdee, 0xfdef, |
| 82 | # others |
| 83 | 0xb, 0xfffe, 0xffff, 0x1fffe, 0x1ffff, 0x2fffe, 0x2ffff, 0x3fffe, 0x3ffff, |
| 84 | 0x4fffe, 0x4ffff, 0x5fffe, 0x5ffff, 0x6fffe, 0x6ffff, 0x7fffe, 0x7ffff, |
| 85 | 0x8fffe, 0x8ffff, 0x9fffe, 0x9ffff, 0xafffe, 0xaffff, 0xbfffe, 0xbffff, |
| 86 | 0xcfffe, 0xcffff, 0xdfffe, 0xdffff, 0xefffe, 0xeffff, 0xffffe, 0xfffff, |
| 87 | 0x10fffe, 0x10ffff |
| 88 | } |
| 89 | |
| 90 | |
| 91 | def _replace_charref(s): |
| 92 | s = s.group(1) |
| 93 | if s[0] == '#': |
| 94 | # numeric charref |
| 95 | if s[1] in 'xX': |
| 96 | num = int(s[2:].rstrip(';'), 16) |
| 97 | else: |
| 98 | num = int(s[1:].rstrip(';')) |
| 99 | if num in _invalid_charrefs: |
| 100 | return _invalid_charrefs[num] |
| 101 | if 0xD800 <= num <= 0xDFFF or num > 0x10FFFF: |
| 102 | return '\uFFFD' |
| 103 | if num in _invalid_codepoints: |
| 104 | return '' |
| 105 | return chr(num) |
| 106 | else: |
| 107 | # named charref |
| 108 | if s in _html5: |
| 109 | return _html5[s] |
| 110 | # find the longest matching name (as defined by the standard) |
| 111 | for x in range(len(s)-1, 1, -1): |
| 112 | if s[:x] in _html5: |
| 113 | return _html5[s[:x]] + s[x:] |
| 114 | else: |
| 115 | return '&' + s |
| 116 | |
| 117 | |
| 118 | _charref = _re.compile(r'&(#[0-9]+;?' |
| 119 | r'|#[xX][0-9a-fA-F]+;?' |
| 120 | r'|[^\t\n\f <&#;]{1,32};?)') |
| 121 | |
| 122 | def unescape(s): |
| 123 | """ |
| 124 | Convert all named and numeric character references (e.g. >, >, |
| 125 | &x3e;) in the string s to the corresponding unicode characters. |
| 126 | This function uses the rules defined by the HTML 5 standard |
| 127 | for both valid and invalid character references, and the list of |
| 128 | HTML 5 named character references defined in html.entities.html5. |
| 129 | """ |
| 130 | if '&' not in s: |
| 131 | return s |
| 132 | return _charref.sub(_replace_charref, s) |