Removing items by id

I’ve got documents that look like this:

const tasks = [
{ id: 1, title: "clean the house" }, 
{ id: 2, title: "eat food" }
]

If the title of the task with id = 1 changes, I’d like to update that change in the index. However, in my current application, I don’t have access to the entire old version of the document. I just know the id and the new values for the title field.

In order to remove an item in Minisearch, it looks like I need to pass the whole document that I originally added. Is there a way I can remove an item by id? If so, I can just remove by id and then add the new document.

[priyadarshy]

Thank you for the detailed help and explanation of why this isn’t supported. Your decision making makes total sense to me. I’m happy to implement this within my application in order support removals, I was just avoiding it incase there was a way to remove items by id.

Thank you for building this awesome library. Before I implemented minisearch, I experimented with fuse, lunr and flexsearch. Minisearch had the best documentation and the it’s search UX felt like it worked nicely out of the box, like a user would expect in most web apps. I found the rest of libraries to be very academic (lunr), hard to get started with (flexsearch), or slow (fuse). Thank you!

[lucaong]

@mashpie @clibu @lukevp @rfox12 @priyadarshy I revive this old thread to announce that the newly released v6.0.0-beta.1 (soon to be v6.0.0) includes a new method discard (and related methods discardAll and replace) which removes a document by ID.

The reasons given in this thread for why remove takes a full document are still valid, but discard takes a different approach, made possible by the internal design of MiniSearch. Instead of immediately removing a document from the index (which would only be feasible given the original document), discard marks the current document version as discarded, so it is immediately ignored by searches, but does not mutate the index. Instead, that is left to a vacuuming process, ran by default after a certain number of documents are discarded. Vacuuming traverses the index and removes references to discarded documents, allowing memory to be released. See the documentation here: https://lucaong.github.io/minisearch/classes/_minisearch_.minisearch.html#discard

This provides the feature requested by this issue, while avoiding the drawbacks of a naive implementation: the original documents are not stored, and use cases that do not need to remove documents incur no overhead.