I have a couple of writes that I want to be done together.
So I used a write batch.
Since the write batch requires a document reference, I have been creating the document before that write batch operation.
DocumentReference accountHistoryDoc = await queryResult.reference.collection('accountHistory').add({});
This led to many empty documents. Since I'm still testing and debugging the app, I assume because of an exception after the creation of the document.
How can I make sure that an empty document isn't created in the case of failure?
I'm thinking of changing this line
wb.set(
accountHistoryDoc, // Change this line
{
'account': newAccount,
'serverTimestamp': FieldValue.serverTimestamp(),
'type': 'hisab',
},
);
to
wb.set(
await subscriberDoc.collection('accountHistory').add({}),// new line
{
'account': newAccount,
'serverTimestamp': FieldValue.serverTimestamp(),
'type': 'hisab',
},
);
Is this my thinking correct?
Rest of code:
QuerySnapshot query = await FirebaseFirestore.instance
.collection(CurrentUser.getCurrentUser().uid)
.where('mobile', isEqualTo: mobile)
.get();
QueryDocumentSnapshot queryResult = query.docs.first;
DocumentReference subscriberDoc = queryResult.reference;
DocumentReference accountHistoryDoc = await queryResult.reference.collection('accountHistory').add({}); // < -- new empty document here
WriteBatch wb = FirebaseFirestore.instance.batch();
// update the total account
wb.update(
subscriberDoc,
{
'totalAccount': subscriber.totalAccount + newAccount,
},
);
// add new document in account history
wb.set(
accountHistoryDoc,
await subscriberDoc.collection('accountHistory').add({}),
{
'account': newAccount,
'serverTimestamp': FieldValue.serverTimestamp(),
'type': 'hisab',
},
);
If you call CollectionReference.doc() without an argument, it generates a new unique DocumentReference without already creating that document in the database. You can then use this DocumentReference to create the new document inside the batched write.
Also see the documentation for the FlutterFire doc() method.
Related
I am trying to delete a field from my firebase array. I am using this code but it deletes then whole activity array.
FirebaseFirestore.instance
.collection(
widget.user.user.uid)
.doc(documentName)
.update({
"activities":
FieldValue.delete()
Instead I want to delete a specific field from 'activities' like activities[2]. How can I do to fix this?
To remove an item at a specific index you can use the arrayRemove method from FieldValue
FirebaseFirestore.instance.collection(widget.user.user.uid).doc(documentName).update({
"activities": FieldValue.arrayRemove("itemToRemove")
})
You should try using FieldValue.arrayRemove() method and then pass the value that you want to remove using this :
var val=[]; //blank list for add elements which you want to delete
val.add('$addDeletedElements');
FirebaseFirestore.instance
.collection(
widget.user.user.uid)
.doc(documentName)
.update({
"activities":
FieldValue.arrayRemove(val])
You have to tell firebase full path like this
FirebaseFirestore.instance
.collection("chat_dialog")
.doc("India")
.collection("dialog_details")
.where(['firestore_id'].contains('your_value'))
.get()
.then((value) {
print(value.docs.length);
for (var element in value.docs) {
print(element.id);
}
});
I want to run a transaction to update data in the Cloud Firestore using cloud_firestore_odm.
This code works fine:
usersRef
.doc('foo_id')
.update(
name: 'John',
);
But this one doesn't. I'm doing something wrong, can anyone tell me how to properly do it?
final transaction = await FirebaseFirestore.instance.runTransaction((_) async => _);
usersRef
.doc('foo_id')
.transactionUpdate(
transaction,
name: 'John',
);
Due to how the ODM works, the syntax for using transactions using the Firestore ODM is slightly different.
Instead of:
await FirebaseFirestore.instance.runTransaction((transaction) async {
transaction.update(usersRef.doc('id'), {'age': 42});
});
You should do:
await FirebaseFirestore.instance.runTransaction((transaction) async {
usersRef.doc('id').transactionUpdate(transaction, age: 42);
});
Basically, the transaction vs "reference" are swapped. But as a benefit, the transaction object is fully typed.
The same logic applies for any other transaction method.
Try this:
await FirebaseFirestore.instance((transaction) async {
await transaction.update(usersRef.doc('foo_id'),{
'name' : 'John'
});
});
I have two collections. One collection "User", who contains the user info (name...) And one collection "Post" who contains all posts of my flutter application. A post document contains many fields like a "Title", "Name" of the user. I add an option in my application to allow the user to change his name. But I must change the name in the "User" collection and in all posts it creates in the "Post" collection. How should I do it? Can anyone help me with an example?
There's nothing magical here. You'll need to get all post documents for that user with a query, and then update them one by one or in batches.
I'd also recommend checking out: What is the fastest way to write a lot of documents to Firestore?
Depends on which firestore are you using.
For Cloud Firestore:
You can update like this way, this is the case where you are updating just one field of your user.
final docRef = db.collection("users").doc("user_id");
final updates = <String, dynamic>{
"timestamp": FieldValue.serverTimestamp(),
};
docRef.update(updates).then(
(value) => print("DocumentSnapshot successfully updated!"),
onError: (e) => print("Error updating document $e"));
For updating a nested field:
// Assume the document contains:
// {
// Title: "Post Tittle",
// Name: "Post Name"
// user: { id: "1", name: "User Name" }
// date: "2022-12-08"
// }
db
.collection("posts")
.doc("post_id")
.update({"date": "2022-13-08", "user.name": "New name"});
You can see more details here: https://firebase.google.com/docs/firestore/manage-data/add-data
Assuming there is a user id attached to each post, you can query from the posts collection where the user (that is changing his/her name) id matches the id that is in the post (from the posts collection) and then modify the user property/attribute from the results.
A sample code would look like this,
To modify from the user collection
final docRef = FirebaseFirestore.instance
.collection('users')
.doc(id);
final response = await docRef
.update(updatedDateInJson)
.then((value) => value)
.onError((error, stackTrace) => error);
To modify from the posts collection where the user is found in a post
final response = FirebaseFirestore.instance
.collection('posts')
.where('name', isEqualTo: 'John')
.get()
.then((value) async {
// modify each doc
});
Quick Note: use a onError block to check for errors in updating docs.
_getLatestCompletedWorkout() async {
try {
QuerySnapshot workouts;
workouts = await FirebaseFirestore.instance
.collection('users')
.doc(FirebaseAuth.instance.currentUser!.uid)
.collection('workouts')
.get();
for (var workout in workouts.docs) {
print('WORKOUT = ');
print(workout);
}
.....
What I really need is to get the last document saved; but before that I am just trying to fetch the "workouts" collection; the workouts.docs list always has 0 items. There are 2 items in the DB. What is wrong with this code? Also how to get the last saved item?
As mentioned by Frank :
You can refer Alex answer here :
The simplest way to achieve this is to add a date
property to each object in your collection, then simply query it
according to this new property descending and call limit(1) function.
This is the required query:
this.historyRef = afs.collection<History>('history', ref => ref.orderBy('date', 'desc').limit(1));
this.history = this.historyRef.snapshotChanges().map(actions => {
return actions.map(a => {
const data = a.payload.doc.data() as Hisotory;
const docId = a.payload.doc.id;
return { docId, ...data };
});
});
This has been found since cloud_firestore updates that prevent app that not regiter in App Check to take updates seriously "but it store it with lighter id color" which Protect your Cloud Firestore resources from abuse, such as billing fraud or phishing
Kindly check and ensure your app is registered inside App Check in Firebase console
The following transaction completely replaces the data in a Firestore Map when run:
DocumentSnapshot docSnapshot = await tx.get(docRef); // doc Snapshot
if (docSnapshot.exists) {
await tx.update(
docRef,
<String, dynamic>{
'Users': {
'user.uid': {'name': user.displayName, 'date': DateTime.now()},
}
},
);
What would be the correct way to add to a map instead of replacing the existing data, the same way FieldValue.arrayUnion works?
Since you already fetched the data you could take the map out from the snapshot, replace the data there and call the update with the altered map.