/*****************************************************************************
 *
 * Download a document and use libtidy to parse and extract <A href=> values.
 * Written by Jeff Pohlmeyer
 *
 * LibTidy => http://tidy.sourceforge.net
 *
 * gcc -Wall -I/usr/local/include tidycurl.c -lcurl -ltidy -o hrefs
 *
 */

#include <stdio.h>
#include <tidy/tidy.h>
#include <tidy/buffio.h>
#include <curl/curl.h>

/* curl write callback, to fill tidy's input buffer...  */
uint write_cb(char *in, uint size, uint nmemb, TidyBuffer *out)
{
  uint r;
  r = size * nmemb;
  tidyBufAppend( out, in, r );
  return(r);
}

/* Traverse the document tree */
void walk_node(TidyDoc doc, TidyNode tnod, int indent )
{
  TidyNode child;
  for ( child = tidyGetChild(tnod); child; child = tidyGetNext(child) )
  {
    if (tidyNodeGetId(child)==TidyTag_A)
    {
      TidyAttr attr;
      /* walk the attribute list */
      for ( attr=tidyAttrFirst(child); attr; attr=tidyAttrNext(attr) ) {
        if (tidyAttrGetId(attr)==TidyAttr_HREF) {
         ctmbstr url=tidyAttrValue(attr);
          if (url) { printf("%s\n", url);}
        }
      }
    }
    walk_node( doc, child, indent + 4 ); /* recursive */
  }
}


int main(int argc, char **argv )
{
  CURL *curl;
  char curl_errbuf[CURL_ERROR_SIZE];
  TidyDoc tdoc;
  TidyBuffer docbuf = {0};
  TidyBuffer tidy_errbuf = {0};
  int err;
  if (argc == 2) {
    curl = curl_easy_init();
    curl_easy_setopt(curl, CURLOPT_URL, argv[1]);
    curl_easy_setopt(curl, CURLOPT_ERRORBUFFER, curl_errbuf);
    curl_easy_setopt(curl, CURLOPT_NOPROGRESS, yes);
    curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_cb);
    curl_easy_setopt(curl, CURLOPT_FOLLOWLOCATION, yes);
    tdoc = tidyCreate();
    tidyOptSetBool(tdoc, TidyForceOutput, yes); /* try harder */
    tidyOptSetInt(tdoc, TidyWrapLen, 4096);
    tidySetErrorBuffer( tdoc, &tidy_errbuf );
    tidyBufInit(&docbuf);

    curl_easy_setopt(curl, CURLOPT_WRITEDATA, &docbuf);
    err=curl_easy_perform(curl);
    if ( !err ) {
      err = tidyParseBuffer(tdoc, &docbuf); /* parse the input */
      if ( err >= 0 ) {
        err = tidyCleanAndRepair(tdoc); /* fix any problems */
        if ( err >= 0 ) {
          err = tidyRunDiagnostics(tdoc); /* load tidy error buffer */
          if ( err >= 0 ) {
            walk_node( tdoc, tidyGetRoot(tdoc), 0 ); /* walk the tree */
          }
        }
      }
    }
    else { fprintf(stderr, "%s\n", curl_errbuf); }
    curl_easy_cleanup(curl);
    tidyBufFree(&docbuf);
    tidyBufFree(&tidy_errbuf);
    tidyRelease(tdoc);
    return(err);

  }
  else
    printf( "usage: %s <url>\n", argv[0] );

  return(0);
}