While creating an ecommerce site site using Shopify, we ran into the issue of having thousands of product images with usable information in the file names, but no way of easily turning that into useful Shopify data. Furthermore, we had no good way, other than the ever-faithful Microsoft Excel, of interacting with and modifying product.csv files provided by Shopify. This utility was born out of that need.

The Shopfiy Inventory Tool (henceforth known as SIT) is pretty straightforward in how it operates. After running the executable it compiles to, you get a basic Windows Forms GUI. SIT then allows you to either import from CSV or import from images. Importing from CSV assumes you're giving SIT a 'products.csv' exported from your shopify admin page. Once you import product data from shopify or from images, you will see a populated grid on the right, and various controls on the left for modifying values.

Importing from images is really the primary utility of SIT. In our case, we went from hours to generate a hundred or so products down to just a handful of minutes using the import from images functionality of SIT.

SIT takes two sets of data into account when generating a Shopify products.csv file. The first piece of data is the images' filenames themselves which should contain useful information. The second piece of data is a details.txt file that lives in subfolders with associated images.

Before you start importing images, you need to generate details.txt files that will live in each subfolder that contains images. When SIT starts its import from images, it will look for this details.txt file to better fill in the data for the products.csv you will eventually upload to Shopify.

Take a look at the sample-details.txt in this repo to see what an example details.txt file looks like. You'll see that the file is a simple list of keys followed by values with colon separation. Newline characters tell the program to look for a new key and subsequent value.

Once SIT finds the Desciption: key in the details.txt file, it stops looking for new keys and reads the remaining text in as the product's description.

One other aspect to using the import from images functionality is that you need to properly name your image files so that they're read by SIT.

Take the file Concerto+RU-201-40+Roll-Up_Jelly_Roll.jpg for example. The + characters designate splits between data. So we have 3 pieces of data in this image: Concerto, RU-201-40, and Roll-Up Jelly Roll. Notice that the _ character is replaced by a space.

When generating product descriptions, SIT lets you populate a description using bits from the product's image name by using curly braces around a number. The number corresponds to a place in the image's filename to look for the desired information.

Take the example image name listed above, Concerto+RU-201-40+Roll-Up_Jelly_Roll.jpg for the following details.txt file.

Vendor: somevendor
Type: sometype
Tags: atagofsomesort
Price: 123.45
Url: http://where-to-find-your-images
Description:
Title of Description Section
First Thing: {1}
Manufacturer: Some Vendor
Second Thing: {3}

After SIT process the image filename and the details.txt file, the description would become:

First Thing: Concerto
Manufacturer: Some Vendor
Second Thing: Roll-Up Jelly Roll

SIT automatically generates the proper HTML for a product description when you list out the description as I did in the example above. The first line after Description: would be wrapped in h4 tags and remaining lines would be placed into an unsorted list or ul.

To override the default HTML layout, wrap your description in double curly braces, {{ description }} like so:

Vendor: somevendor
Type: sometype
Tags: atagofsomesort
Price: 123.45
Url: http://where-to-find-your-images
Description:
{{
<div>
<table>
	<tr>
		<td>
			Foo
		</td>
	</tr>
</table>
</div>
}}

SIT will glady go through a folder and all of the folder's subfolders, looking for image files and details.txt files. Using a system of folder naming with descriptive image file names will yield best results.

I'm fully aware that I could generalize this utility even more. However, this implementation works very well for us, and I honestly don't know what generalizing the utility further would gain for people. That being said, I'd happy to be convinced that I'm wrong.

Shopify Inventory Tool icon source - Variations 1 by GuillenDesign