<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>Hi all,<br><br>see also<br>http://gis.19327.n5.nabble.com/address-search-in-Canada-tp5856816.html<br><br>I see several ways to implement the preparation proposed by Andrzej.<br>I think about implementing a method that removes <br>a) trailing + leading spaces from tag values and<br>b) all doubled spaces within the tag value<br>Examples:<br>" abc" -> "abc"<br>"abc " -> "abc"<br>"abc def" -> "abc def"<br>" x y z " -> "x y z"<br><br>Now, this can be done <br>a) with all (evaluated) tags before style rules are processed<br>pro: style rules do not have to care about these cases<br>contra: extra spaces produced by the style actions are not catched<br>b) after style processing<br>pro: style can handle these special cases (e.g. use echotags),<br>extra spaces produced by style actions are handled<br>contra: style rules may not match, e.g. if an element <br>has the tag highway="primary " <br><br>c) two times a) + b)<br><br>I think option c) is the best, presuming that it doesn't involve <br>an unacceptable performance impact.<br><br>I am not sure if anybody uses these spaces willingly?<br>Do we need an option to configure this?<br><br>Gerd<br><br> </div></body>
</html>