Relation between Search results and Inventory data / settings? Some things work, some don't

Question

I have a ton of items I imported via REST API. They don't show up in search. For this test I'm just using the LIKE operator. Magento 1.8

What works:

• Manually edit a single product
• Set qty=1 and is_in_stock to yes/1
• If prompted, redo search index

The problem is this doesn't scale; I have tens of thousands of products.

What doesn't work, attempt 1:

• System -> Configuration -> Catalog / Inventory -> Stock Options
• Display Out of Stock Products = Yes
• Redo search index (all indexes, just to be safe)

This is the purported fix that I found on Google a while back, but annoyingly doesn't work.

What doesn't work, attempt 2:

• Run SQL command: UPDATE magento.cataloginventory_stock_item SET qty = 1, is_in_stock = 1;
• Redo search index (all indexes, just to be safe)

I know you're not supposed to do this, but this is a dev system and I wanted to try.

Use REST API?

• I believe I tried this some time back and it didn't work.
• Even if it did work, it'd be too slow. We're moving away from the REST API because it takes days / weeks to complete for a large data-set.

Questions:

• When I manually edit a product and set qty=1 / in-stock=yes, it's clearly doing something "special", above and beyond tweaking the cataloginventory_stock_item table. What else is it doing?

• Is there some way to simulate whatever that is?

• What the heck is the point of the "Display Out of Stock Products = Yes" option if it doesn't affect search?

Magento seems to do a lot of internal filtering... I'm guessing that something in my data doesn't quite match.

Answer 1

The fastest method you're going to find for fixing this issue would be to use an extension such as uRapidFlow. I've seen and had similar issues before with various methods of importing products, normally opening and saving the product repairs the issues because something is not set during the import. This is normally the attribute set, sometimes the stock, but the end result is the fix is always tedious and not adequate. I've had a lot of success using uRapidFlow since then for importing mass products and having them work exactly as intended. You can also use it to export your current set, add any missing data and update all of your current products very rapidly. Just make sure to enable the indexes when setting up the profile.

The alternative to all of this is trying to find exactly what piece of information is missing from the import you did through the API. This would require more information to troubleshoot and answer, such as the exact API method used for your import.