Write `htmlDocPtr` to String

I'm trying to parse an HTML string, and write it out again (for the purposes of brevity, I've left out the transforms I wish to perform).

#include <iostream>
#include <libxml/HTMLparser.h>

static const char *html = "<!DOCTYPE html><html><head></head><body><div></div></body></html>";

int main(int argc, const char * argv[]) {
    xmlChar *buff;
    int buffersize;

    htmlDocPtr doc = htmlReadMemory(html, (unsigned)strlen(html), "noname.html", NULL, 0);
    xmlDocDumpFormatMemory(doc, &buff, &buffersize, 1);

    printf("%s", (char *) buff);

    xmlFree(buff);
    xmlFreeDoc(doc);

    return 0;
}

(I am almost 100% green in C++ land - please excuse any outlaying errors)

This all works (in that it doesn't error out), but xmlDocDumpFormatMemory treats the tree as XML and outputs accordingly:

<?xml version="1.0" standalone="yes"?>
<!DOCTYPE html>
<html>
  <head/>
  <body>
    <div/>
  </body>
</html>

As you can see, there's an XML declaration added above the doctype, and empty tags have been self-closed. As far as I can tell, there is no htmlDocDumpFormatMemory - is there an option or alternate function I can use?

Solution

Try the following code. It makes use of the xmlSaveToXXXX-functions (xmlSaveToBuffer, in this case), which allow to pass options like, for example, XML_SAVE_NO_DECL:

#include <iostream>
#include <libxml/HTMLparser.h>
#include <libxml/xmlsave.h>

static const char *html = "<!DOCTYPE html><html><head></head><body><div></div></body></html>";

int main(int argc, const char * argv[]) {

    htmlDocPtr doc = htmlReadMemory(html, (unsigned)strlen(html), "noname.html", NULL, 0);

    xmlBufferPtr buffer = xmlBufferCreate();
    if (buffer ==  NULL)
        return 1;  // Add error handling...

    xmlSaveCtxtPtr saveCtxtPtr = xmlSaveToBuffer(buffer,NULL, XML_SAVE_NO_DECL);
    if (xmlSaveDoc(saveCtxtPtr, doc) < 0)
        return 1;  // Add error handling

    xmlSaveClose(saveCtxtPtr);

    const xmlChar *xmlCharBuffer = xmlBufferContent(buffer);

    printf("%s", xmlCharBuffer);

    xmlBufferFree(buffer);
    xmlFreeDoc(doc);

    return 0;
}

It produces the following output:

<!DOCTYPE html>
<html><head></head><body><div></div></body></html>