Hi,
In my webtest, I use a storeXPath statement to extract information from the web page. The information in the web-page looks as follows: <TD class="verdanaBlue">74 200 kg </TD> The information to be extracted must be 74200 without leading and trailing spaces This looks quite straight-forward, except for filtering the sequence. >From this topic http://stackoverflow.com/questions/247135/using-xpath-to-search-text-contain ing-, I learned that it can be matched by typing Alt+0160 on the NUMERIC keypad. And indeed, the following xpath expression works in the xpath section of Mozilla Webtest Recorder: translate(normalize-space(//td/text()), 'kg ', '') where the blank right after 'kg' is the Alt+0160 and after that a normal space is added to remove the space between the 74 and the 200. When you apply the string-length command on this result, it returns 5, ensuring that I have exactly the result I want (74200 without leading and trailing spaces). However, when I use this xpath in the storeXPath webtest command, there is a parsing error for that line containing the xpath: Invalid byte 1 of 1-byte UTF-8 sequence. Does anybody have experience in the character and xpath? Is there a workaround for avoiding the error message in Webtest? Is it maybe a bug? Thanks for the help! David
