More html symbols

src/calibre/ebooks/htmlsymbols.py

@@ -119,6 +119,97 @@ HTML_SYMBOLS = {
                 u'♥' : ['♥', '♥'], # heart
                 u'♦' : ['♦', '♦'], # diamond
                 # Extra http://www.ascii.cl/htmlcodes.htm
+                u' ' : [' '], # space
+                u'!' : ['!'], # exclamation point
+                u'#' : ['#'], # number sign
+                u'$' : ['$'], # dollar sign
+                u'%' : ['%'], # percent sign
+                u'\'' : ['''], # single quote
+                u'(' : ['('], # opening parenthesis
+                u')' : [')'], # closing parenthesis
+                u'*' : ['*'], # asterisk
+                u'+' : ['+'], # plus sign
+                u',' : [','], # comma
+                u'-' : ['-'], # minus sign - hyphen
+                u'.' : ['.'], # period
+                u'/' : ['/'], # slash
+                u'0' : ['0'], # zero
+                u'1' : ['1'], # one
+                u'2' : ['2'], # two
+                u'3' : ['3'], # three
+                u'4' : ['4'], # four
+                u'5' : ['5'], # five
+                u'6' : ['6'], # six
+                u'7' : ['7'], # seven
+                u'8' : ['8'], # eight
+                u'9' : ['9'], # nine
+                u':' : [':'], # colon
+                u';' : ['&#59;'], # semicolon
+                u'=' : ['='], # equal sign
+                u'?' : ['?'], # question mark
+                u'@' : ['@'], # at symbol
+                u'A' : ['A'], # 
+                u'B' : ['B'], # 
+                u'C' : ['C'], # 
+                u'D' : ['D'], # 
+                u'E' : ['E'], # 
+                u'F' : ['F'], # 
+                u'G' : ['G'], # 
+                u'H' : ['H'], # 
+                u'I' : ['I'], # 
+                u'J' : ['J'], # 
+                u'K' : ['K'], # 
+                u'L' : ['L'], # 
+                u'M' : ['M'], # 
+                u'N' : ['N'], # 
+                u'O' : ['O'], # 
+                u'P' : ['P'], # 
+                u'Q' : ['Q'], # 
+                u'R' : ['R'], # 
+                u'S' : ['S'], # 
+                u'T' : ['T'], # 
+                u'U' : ['U'], # 
+                u'V' : ['V'], # 
+                u'W' : ['W'], # 
+                u'X' : ['X'], # 
+                u'Y' : ['Y'], # 
+                u'Z' : ['Z'], # 
+                u'[' : ['['], # opening bracket
+                u'\\' : ['\'], # backslash
+                u']' : [']'], # closing bracket
+                u'^' : ['^'], # caret - circumflex
+                u'_' : ['_'], # underscore
+                u'`' : ['`'], # grave accent
+                u'a' : ['a'], # 
+                u'b' : ['b'], # 
+                u'c' : ['c'], # 
+                u'd' : ['d'], # 
+                u'e' : ['e'], # 
+                u'f' : ['f'], # 
+                u'g' : ['g'], # 
+                u'h' : ['h'], # 
+                u'i' : ['i'], # 
+                u'j' : ['j'], # 
+                u'k' : ['k'], # 
+                u'l' : ['l'], # 
+                u'm' : ['m'], # 
+                u'n' : ['n'], # 
+                u'o' : ['o'], # 
+                u'p' : ['p'], # 
+                u'q' : ['q'], # 
+                u'r' : ['r'], # 
+                u's' : ['s'], # 
+                u't' : ['t'], # 
+                u'u' : ['u'], # 
+                u'v' : ['v'], # 
+                u'w' : ['w'], # 
+                u'x' : ['x'], # 
+                u'y' : ['y'], # 
+                u'z' : ['z'], # 
+                u'{' : ['{'], # opening brace
+                u'|' : ['|'], # vertical bar
+                u'}' : ['}'], # closing brace
+                u'~' : ['~'], # equivalency sign - tilde
                 u'<' : ['&#60;', '&lt;'], # less than sign
                 u'>' : ['&#62;', '&gt;'], # greater than sign
                 u'¡' : ['&#161;', '&iexcl;'], # inverted exclamation mark