Extracting collection/document path from document name

[juliusn]

I often find myself in a situation where I have a Firestore document that I want to update or delete, but no other reference to the collection/name than Document.name, which doesn't work as is, so I wrote custom functions to extract these parts from the name string. Is this really necessary? Is there a built-in solution in the library for doing this, or if not, should there be?

function pathToRef_(path) {
  return path.match(/(?:[^\/]*\/|^)[^\/]*$/)[0];
}

const docs = fs.getDocuments("collection");

for (const doc in docs) {
  
  const data = {
    "key": value
  }
  
  fs.updateDocument(pathToRef_(docs[doc].name), data, true);
}

Library Version: 32

[LaughDonor]

I'm not sure I'm following accurately... Can you provide an example name and walk through the problem?

We have Util_.regexPath with the expression that is used by the method Util_.getDocumentFromPath(). Does this help at all?

[juliusn]

For example if I have a collection of documents and I want to update a common property, calling updateDocument like so:

const docs = fs.getDocuments("collection");

docs.forEach(doc => fs.updateDocument(doc.name, { "key": "newVal" }, true));

throws an error saying Document name "[...]" lacks "/" at index [...]. If I want to get a working path for each document, I need to extract it from the Document.name property myself using string methods or regex match. I was just wondering if this is really necessary.

[LaughDonor]

Oh, I get it. Wouldn't you be able to do:

const collectionPath = "collection";
const docs = fs.getDocuments(collectionPath);
docs.forEach(doc => fs.updateDocument(`${collectionPath}/${doc.name}`, { "key": "newVal" }, true));

to accomplish the same without the extraction process?

[juliusn]

Oh, I get it. Wouldn't you be able to do:

const collectionPath = "collection";
const docs = fs.getDocuments(collectionPath);
docs.forEach(doc => fs.updateDocument(`${collectionPath}/${doc.name}`, { "key": "newVal" }, true));

to accomplish the same without the extraction process?

That would create a completely new collection hierarchy under collection, because Document.name returns with the base path, like projects/project_id/databases/(default)/collection_name/document_name.

[LaughDonor]

Ok, I'll be adding a property to the Document that lets you get this data. In the meantime, you can use:

docs.forEach(doc => {
    const name = doc.name.match(/^projects\/.+?\/databases\/\(default\)\/documents\/(.+\/.+)$/)[1];
    fs.updateDocument(name, { "key": "newVal" }, true));
});

Where the RegEx is from Util_.regexPath.