This seems like an odd thing to state with such certainty. I'm not sure I see how it would be good for competition either - having worked for a grocery store before, management was generally aware of competitor's prices, at least at a "we send someone to every other competitor every few weeks to check prices on major important stuff". We were very open internally that we had the prices of a few items pegged to always be cheaper than any price Walmart had for them.
As stated other places in the thread, most grocery stores are super low margin overall. Especially on basic food items. Things like canned goods, basic cuts of meat, eggs, milk and the like were often sold at cost or even at a loss. Things that we actually had good margins on were mostly products where we had vertical integrations or special contracts with farms or factories - we were matching our competitors' prices, but at a better margin because we'd worked out how to pay less for the raw goods.
With all that in mind, I highly doubt how "disastrous" it would be to have a public price database. Sure, you could probably shave your pennies by spreading your shopping over 4-5 different stores because they're each slightly cheaper than each other, but you'd almost assuredly spend more in time and transportation costs to do that than you'd save.
What would cause the stores to lobby against it is that many stores, especially now, change prices on things pretty often. The CSV you imagine publishing to a government site is something that, as I recall, takes something like 5 hours to generate every night due to the vast variety of products across the company, and the fact that different stores in different areas may be pricing products differently or offering different coupons, or simply not stock the product at all. There is no one price per barcode universally across the company, so the outputted pricing list ends up being gigabytes of data.