Parsing unusual XML file

copytco

Hi there,

I am trying to parse a BeerXML file (specification can be found under www.beerxml.com), which is kind of unusual. The structure looks more or less like below.

<?xml version="1.0" encoding="ISO-8859-1"?>
<RECIPES>
   <RECIPE>
     <NAME>Wiezenbock</NAME>
     <VERSION>1</VERSION>
     <TYPE>All Grain</TYPE>
     <BREWER>Panin</BREWER>
     <ASST_BREWER></ASST_BREWER>
     <BATCH_SIZE>20.0000000</BATCH_SIZE>
     <BOIL_SIZE>25.3400000</BOIL_SIZE>
     <BOIL_TIME>70.0000000</BOIL_TIME>
     <EFFICIENCY>70.0000000</EFFICIENCY>
     <HOPS>
        <HOP>
        <NAME>Lublin</NAME>
        <VERSION>1</VERSION>
        <ORIGIN>Poland</ORIGIN>
        <ALPHA>5.0000000</ALPHA>
        </HOP>
        ...
      </HOPS>
       ...
   </RECIPE>
</RECIPES>

And so on. Despite wrong encoding (I will take care of this myself), here are many children, while built-in XML parser reads only one parent - "RECIPES" and only one child - "RECIPE". The XML file is automatically generated by BeerSmith software, so I do not have any possibility to modify the output. Is there any way to properly extract e.g. all values of "NAME" for all found "HOP" within "HOPS"?

PhiLho

I don't see how the XML is "unusual". Looks OK to me. Processing has the loadXML() method, have you looked at it?

GoToLoop

Like @PhiLho said, that's a valid XML file: :-\"

XML beer = loadXML("beer.xml");
println(beer);
exit();

However, if you're only looking for NAME values inside HOP entries, a custom parser can be written too: :-bd

// forum.processing.org/two/discussion/10016/parsing-ususual-xml-file

void setup() {
  String[] beers = loadStrings("beer.xml");

  printArray(beers);
  println();

  String[] names = findHopNames(beers);
  printArray(names);
  exit();
}

static final String[] findHopNames(String[] arr) {
  if (arr == null || arr.length == 0)  return new String[0];

  StringList sl = new StringList();
  boolean hopFlag = false;

  for (String s : arr) {
    if (s == null)  continue;

    if      (s.contains("<HOP>"))   hopFlag = true;
    else if (s.contains("</HOP>"))  hopFlag = false;
    else if (hopFlag) {
      String name = extractHopName(s);

      if (name != null) {
        sl.append(name);
        hopFlag = false;
      }
    }
  }

  return sl.array();
}

protected static final String extractHopName(String s) {
  int idx = s.indexOf("<NAME>") + 6, end = s.indexOf("</", idx);
  return idx != 5? s.substring(idx, end) : null;
}

PhiLho

Bad idea to use a custom parser for XML, in general... As shown in the loadXML page, the loaded XML can be walked to inspect the children, to an arbitrary depth.

copytco

PhiLho, GoToLoop, thank you both for the contribution.

GoToLoop, your custom parser is a great solution, but why write new functions, when the default library should deal with a problem right away. :) If we would not figure out what I am doing wrong here, so I cannot parse the file in normal way I will use your solution.

PhiLho, I am using loadXML() but somehow I am not able to extract information I want. I can only get "RECIPES" as parent and "RECIPE" as child. To write my code I used references from here - https://processing.org/reference/loadXML_.html . Maybe I just cannot see something that is clear, but I would be grateful if you could show me how to get names of each hop within the recipe.

copytco

Ok, I have figured it out. Below working code.

XML xml;

void setup() {
  xml = loadXML("recipe.xml");
  XML[] recipe = xml.getChildren("RECIPE");
  XML[] hops = recipe[0].getChildren("HOPS");
  XML[] hop = hops[0].getChildren("HOP");

  String[] hop_names = new String[hop.length];

  for (int i = 0; i < hop.length; i++)
  {
    XML[] x = hop[i].getChildren("NAME");
    hop_names[i] = x[0].getContent();
    println(hop_names[i]);
  }
}

PhiLho

Good! Thanks for sharing your solution.
Pro Tip when asking for help: show your attempt, so that we can see what went wrong... :-) Even better here as you solved the issue by yourself.