A temp buffer for every characters call, push the terminating null and then cat to a longer lived buffer. The text is complete when you hit the endElement and you get it out of the longer lived buffer, which is then ready for the next characters field.
... at least that has worked for me, but I am only a beginner here. If you want pretty, you probably need another answer :-) BJ On Tue, 2004-02-24 at 07:03, Vincent Finn wrote: > so what is the best strategy for getting a complete text? > > Do I have to do multiple calls to parseNext whenever I hit a text node until I hit > an endElement? > > > > -----Original Message----- > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > > Sent: 23 February 2004 17:50 > > To: [EMAIL PROTECTED] > > Subject: Re: splits in character data using SAX > > > > > > > > > > > > > > Hi, > > > > The parser will also make multiple calls to the characters > > function when it > > encounters an entity reference, so your plan is likely > > unworkable. Also, > > it's quite possible the buffer size could change in the future, or the > > parser's behavior could change for some other reason. > > > > Dave > > > > > > > > > > > > > > "Vincent Finn" > > > > > > <[EMAIL PROTECTED] To: > > <[EMAIL PROTECTED]> > > > > omsoft.com> cc: > > (bcc: David N Bertoni/Cambridge/IBM) > > > > Subject: > > splits in character data using SAX > > > > 02/23/2004 09:29 > > > > > > AM > > > > > > Please respond to > > > > > > xerces-c-dev > > > > > > > > > > > > > > > > > > Hi, > > > > I have a question about the way DefaultHandler::characters() > > can send to > > chunks for a single text element > > > > I found an answer of sorts in the mail archive but I am > > hoping for a bit > > more certainty > > > > The question is when will the characters function split data? > > Are there specific rules? > > > > What I am hoping for is to be able to make sure this will never happen > > so I can ignore it rathger than having to write extra code. > > > > The archive link is this > > http://marc.theaimsgroup.com/?l=xerces-c-dev&m=100076161808793&w=2 > > > > essentially that there may be a split if the buffer is full > > in which case can anyone tell me the buffer size? > > > > then and the second reply mentions running into child nodes > > which has me > > slightly confused > > it mentions that a split may happen if a child node is found, but text > > nodes don't have child nodes > > have I missed the point of the function? > > I was working with it as a method of getting the value of text nodes > > > > Vin > > > > -------------------------------------------------------------- > > ---------- > > [EMAIL PROTECTED] > > -------------------------------------------------------------- > > ---------- > > I will find humour in my everyday life > > by looking for people I can laugh at. > > -------------------------------------------------------------- > > ---------- > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
