Shopify Product Replacement Parts - api

Question

Looking for some advice on an interesting situation using Shopify. I'm building a site for a client that has Products that have free replacement parts available. Each replacement part has variant color options.
So far I have had the users at the company add all the replacement parts as products in the store. I have filtered the search results and the catalog results so the replacment parts are not show.
The only place they want the replacement parts to show is when a user visits a PRODUCT, they can click a button that says order replacement parts. Then a screen will show with all the replacement parts for that product.
A single replacement part may belong to several products and may have different color variants.
So what I have done thus far was had the client tag all parts with at least two tags. A tag called "part" that identifies the product as a part. And one or more tags like "link:SKU123" that links the part to one or more products. Then on my Product page I was using liquid to loop all parts and display the ones that matched the products SKU. Then I found out that the for loop has a 50 item limit...
So I looked at the product API, which would be ok, except that it has no way to filter by tags. Tags seem so handy and yet I don't see many ways to use them... So ultimately I'm looking for a way to display all replacement parts for a particular product. It doesn't have to involve tags, although the client has already tagged all the parts and I would hate to tell them it was a waste of time. But really any thoughts on how to accomplish this would be greatly appreciated.
The only way I can think to do it, is to return all of the replacement parts and then filter through them on the page. I see the API is limited to 250 products, but I suppose I could make multiple calls until I get them all. I'm not sure how many replacment parts there are total, but I suspect there could be upwards of 1000. Seems like a waste of network resources to have to pull them all down and then filter through them...
P.S. - the replacement parts are free, can they be run through the checkout with a zero dollar amount?

Answer 1

Ok so I have a couple different answers to this question. Not sure which one I will use yet.
Method 1: Filter Collection by Tag using URL
The first one was provided by Shawn Rudolph on the Shopify forums. It involves filtering a collection by tag using the URL. Shawn's post here explains it well: https://ecommerce.shopify.com/c/shopify-discussion/t/product-replacement-parts-270174
Method 2: Use paginate to get all products from collection using the AJAX API
This method is pretty cool. Yes it is more work than method one, but this maybe useful in a lot of scenarios. Out of the box Shopify does not allow you to retrieve all products from a given collection using the AJAX API. It can be done with the admin API but not the AJAX one to my knowledge. However, you can access all products from a collection with a for loop, but the for loop only allows up to 50 items to be looped at a time. That's where the paginate trick comes in. Basically I adapted the technique outlined by davecap here: http://www.davecap.com/post/9675189741/infinite-scroll-for-shopify-collections
So first you need your HTML/Liquid layout:
{% paginate collections.mycollectionname.products by 50 %}
{% for product in collections.mycollectionname.products %}
<div class="clone-node" id="product-{{ forloop.index | plus:paginate.current_offset }}">
{{ product.title }}
</div>
{% endfor %}
{% if paginate.next %}
<div class="clone-node next" title="{{ paginate.next.url }}"></div>
{% endif %}
<div id="insertion-point"></div>
{% endpaginate %}
So let's break it down a bit. First we are paginating are products by 50. This is the max amount a for loop will allow, so that's what we are going to use:
{% paginate collections.mycollectionname.products by 50 %}
Next we begin to loop our products. Every product is given a wrapper div with a class of "clone-node" this is very important. I also assign the div a unique ID, which isn't necessary for this to work, but may come in handy when trying to identify the product for later operations.
{% for product in collections.mycollectionname.products %}
<div class="clone-node" id="product-{{ forloop.index | plus:paginate.current_offset }}">
{{ product.title }}
</div>
{% endfor %}
We have to make sure to include the paginate.next URL. We also give this a "clone-node" class and we add a "next" class. I assign the paginate.next.url to the title attribute, but you could assign it to any number of attributes. You just need to be able to fetch it with jQuery.
{% if paginate.next %}
<div class="clone-node next" title="{{ paginate.next.url }}"></div>
{% endif %}
Then lastly we assign an insertion point. This is where we want our next set of 50 products to be inserted once we fetch them:
<div id="insertion-point"></div>
OK so now let's look at the JS code:
<script>
var prevUrl = ""; //this helps us know when we are done receiving products
function getParts() {
//get the last instance of the .next node. This will give us the next URL to query
var nextNode= $(".next").last(),
url = nextNode.attr("title"); //nab the URL
//send a get request to our next URL
$.ajax({
type: 'GET',
url: url,
dataType:'text',
success: function (data) {
//use a dummy div to convert the text to HTML, then find all of our clone-nodes, including our new "next" div which contains our next URL
var cloneNodes = $("<div>").html(data).find(".clone-node");
//insert our new clone-nodes on the page
cloneNodes.insertBefore("#insertion-point");
//if the URL's don't match let's grab the next 50!
if (prevUrl != url) {
prevUrl = url;
getParts();
}
}
});
}
//Call getParts for the first time to get the party started.
getParts();
</script>
What this will basically do, is get the URL for the next page of products from the title attribute of the div that's holding the paginate.next.url. Then using the jQuery ajax function we call that URL and it returns a page of HTML to use formatted just like our existing page, with the same "clone-node" classes we assigned, only it has the next 50 products embedded in it.
In davecap's example he used a dataType of HTML on his Ajax call, but that gave me some troubles. So instead, I used dataType text and used a dummy div created by jQuery to convert the text into HTML. Then jQuery grabs all of the divs with the "clone-node" class on them and inserts them on the page before our insertion-point. Remember the clone-nodes now hold the next 50 products so we just added the next 50 products to our page.
Lastly, we check if the previous URL is not equal to the current one. If it's not equal, that means it's a new URL and thus there must be more products to fetch, so we recursively call our getParts() function, which starts the process over and grabs the next 50. This continues until finally the URLs match, which means no more products to fetch, and the process stops.
There you have it! Of course if you have to fetch thousands and thousands of products this may be less then ideal because you are calling them 50 at a time. But for smaller numbers (maybe hundreds and hundreds...) it should work just fine.

Answer 2

I would create the parts as standalone products. Then each part will have variants (colours). Then you create metafields for this parts (product) which have a field with a list of all product ids which are 'mother' for this parts. This way you don't need to have strange tags and you keep concepts more separated, & everything cleaner. As per the zero amount, yes, you can go to checkout with zero amount (if your shipping settings allow you to do so).