Yield does not return data

Question

When calling the method, an HTML component enumerable should be returned.

I’m using the HTML Agility Pack to read an HTML file. The same method works as expected when removing the yield and add manually to a list

    HtmlNode slideCineAll = GetNodeById(cinema, "slide-cine-all");
    HtmlNode section = GetNodeByName(slideCineAll, "section");
    IEnumerable<HtmlNode> articles = GetNodesByName(section, "article");

    private static IEnumerable<HtmlNode> GetNodesByName(HtmlNode root, string node)
    {
        foreach (HtmlNode link in root.ChildNodes)
        {
            if (link.Name.Equals(node))
            {
                yield return link;
            }
        }
    }

    private static List<HtmlNode> GetNodesByNameList(HtmlNode root, string node)
    {
        List<HtmlNode> nodes = new List<HtmlNode>();
        foreach (HtmlNode link in root.ChildNodes)
        {
            if (link.Name.Equals(node))
            {
                nodes.Add(link);
            }
        }
        return nodes;
    }

This is the result stored in the variable when executing the method

{ConsoleApplication1.Program.GetNodesByName}
node: null
root: null
System.Collections.Generic.IEnumerator<HtmlAgilityPack.HtmlNode>.Current: null
System.Collections.IEnumerator.Current: null

Expected result

values
Count = 20
[0]: Name: "article"}
.
.
.
values[0]
_attributes: {HtmlAgilityPack.HtmlAttributeCollection}
_childnodes: {HtmlAgilityPack.HtmlNodeCollection}
_endnode: Name: "article"}
.
.
.

This is the structure I’m going through, through the method GetNodesByName or GetNodesByNameList i can retrieve a list of any structure node html

<div id="slide-cine-all">
<section>
    <article>
        <!--mais elementos-->
    </article>
    <article>
        <!--mais elementos-->
    </article>
    <article>
        <!--mais elementos-->
    </article>
    <article>
        <!--mais elementos-->
    </article>
    <article>
        <!--mais elementos-->
    </article>
    <article>
        <!--mais elementos-->
    </article>
</section>
</div>

As described at the beginning, the Getnodesbynamelist method returns all items, in this case type article found in the file structure, but the same doesn’t happen when I use Yield.

Answer 1

See the documentation of yield return. He doesn’t do what seems to be what you expect of him. The part of return is important. When it arrives on this line, it finishes the execution of the method. So your code only returns an element of the existing nodes in your XML.

You can return more. You need to call the method more often. Each call will run a new line from where it stopped. The yield creates something called Generator (in English). It controls execution through a hidden state that determines at which point of an enumeration the program is running, so the call can continue from where it left off. Note that it returns a enumerable type and not the type of what you want in itself. This enumerable structure is what controls the continuity of execution from where it left off.

So your problem is that you get a single link of the knot and then tries to search inside it as if there were other elements of this knot. Of course there are no other elements, you haven’t read them yet. That’s where the problem is.

In the other method that works by not having the yield, the loop runs completely and scans all nodes and returns a more complete tree that can then be searched without problem. All you need is there.

I didn’t understand the problem as a whole but I think in this case the yield is getting in the way. And I advise using it only when you fully understand how it works. It is excellent but not a solution to all problems. I am not saying that this problem cannot benefit from it (generate efficiency by not sweeping the whole structure but only what is needed at the moment) but it would need to change some things in the code that consumes this method. In practice when you use one yield, roughly speaking, you will have another loop external to scan the entire structure you are searching for (of course you can also do it manually again again).

I suggest to inspect the data and follow the execution on debug to better understand what is happening with the code. It can help to learn about the functioning of the yield and see clearly your problem, and who knows until you find a better solution. Or explain otherwise the problem that I couldn’t see better than this.

To help understand execute the following code taken from this response in the OS:

public void Consumer() {
    foreach(int i in Integers()) {
        Console.WriteLine(i.ToString());
    }
}

public IEnumerable<int> Integers() {
    yield return 1;
    yield return 2;
    yield return 4;
    yield return 8;
    yield return 16;
    yield return 16777216;
}

Or this one response from the OS:

// Display powers of 2 up to the exponent 8:
foreach (int i in Power(2, 8)) {
    Console.Write("{0} ", i);
}

public static IEnumerable<int> Power(int number, int exponent) {
    int counter = 0;
    int result = 1;
    while (counter++ < exponent) {
        result = result * number;
        yield return result;
    }
}

I put in the Github for future reference.

Noted that consumer code always end up having to repeat the calls, in a certain way the repeats existing in the generator code (of yield)? The great advantage of yield for most situations is to create better abstractions.

See a quick explanation of what the Internals of command. And a more complete explanation.