I use GreenDao in my project and it does a good job for mapping my server model to the android device. What I have been struggling with for a while now is that dao.update and/or dao.updateInTx methods are not updating any rows in the database.
What I do:
/**
* The first approach -> All in one runnable
*
*/
// list definitions
ArrayList<Country> countryList;
ArrayList<Country> countryListDetail;
final DaoSession daoSession = daoMaster.newSession();
// execute everything in runnable
// in order to optimize insert time
daoSession.runInTx(new Runnable() {
#Override
public void run() {
CountryDao countryDao = new CaountryDao();
// delete all records first
countryDao.deleteAll();
// insert countries with less data
size = countryList.size();
for (int i=0; i < size; i++) {
countryDao.insert(countryList.get(i));
}
// update countries with more data
// ID's of the Objects in countryListDetail match those from
// the countryList, so the primary key mathces
size = countryListDetail.size();
for (int i=0; i < size; i++) {
countryDao.update(countryListDetail.get(i));
}
}
}
/**
* The second approach -> updateInTx()
*
*/
// list definitions
ArrayList<Country> countryList;
ArrayList<Country> countryListDetail;
// insert & update logic
final DaoSession daoSession = daoMaster.newSession();
CountryDao countryDao = new CaountryDao();
countryDao.insertInTx(countryList);
countryDao.updateInTx(countryListDetail);
In both cases, when I pull the database from the device and inspect it, the Country table has only the base insert data, but no detail data which should come from the update statements. When debugging, the GreenDao logic seems to execute the updateInsideSynchronized() method and stmt.execute() is also called. Does anybody know what I may be doing wrong?
Please enable this and check
dao.queryBuilder().LOG_VALUES=true;
dao.queryBuilder().LOG_SQL=true;
and make sure countryList and countryListDetail have the same Primarykey when you update
Probably your countyListDetail doesn't contain Country-objects, that have been queried using greendao. Hence the objects have "no idea about" the database, because there is no reference to the DaoSession. The update only works for objects already inserted or queried from the database.
If the data in countryList is included in the data of countryListDetail, I'd use insertOrReplace.
Otherwise you have to merge your Country-objects before insert or before update.
Related
I am inserting a new book into my book table and after trying to assign it to a many-to-many relation table. Imo this should run in a transaction.
(Because if the m2m insertion fails, the information about the realtionship is lost). My code now looks as follows and fails as i cannot access the BookUserXRefDao.insert(bookUser); query due to static context errors.
Is there an easy way to fix this?
#Transaction
public void insertBook(Book theBook, List<Integer> userIds){
long newBookId= insert(theBook);
//Insert into the m2m relation
BookUserXRef[] bookUser = new BookUserXRef[userIds.size()];
for (int i = 0; i < userIds.size(); i++) {
BookUserXRef[i] = new BookUserXRef(newBookId,userIds.get(i));
}
BookUserXRefDao.insert(bookUser);
}
Just realized that i can access the Singleton Database Instance from within my transaction.
Therefore i could just use
AppDb.getAppDb().BookUserXRefDao().insert(bookUser);
That solved the problem.
I'm a beginner in Realm.
I have a table with 3 columns which named Id, Name, Email,Address.
To get the data of Name column, we use a query like 'SELECT Name from table_name' for SQLite.
If we using Realm in Android, then which method do we have to use for fetching the data of only one column?
I searched alot on Google & documentation but to no avail.
Could anyone help me?
Update:
What I am tried:
RealmResults<User> results = query.findAll();
ArrayList<String> name = new Arraylist();
for(i=0; i<results.size; i++){
name.add(result.get(i).getName();
}
My problem:
results.size() > 10k. So I want to avoid 10k iteration
for(i=0; i<results.size; i++){
}
Look at queries section at the documentation:
All fetches (including queries) are lazy in Realm, and the data is never copied.
This mean, that data of particular column (property) will be fetched when you call getMyProperty() method. Not after call of finadAll() method of RealmQuery object
If we using Realm in Android, then which method do we have to use for fetching the data of only one column?
You can't, because Realm is an object store, it doesn't have concept of "columns".
My problem:
results.size() > 10k. So I want to avoid 10k iteration
for(i = 0; i < results.size(); i++){
}
Solution: don't iterate?
RealmResults<User> results = query.findAll();
//List<String> name = new ArrayList<>();
//for(i = 0; i < results.size(); i++){
// name.add(result.get(i).getName();
//}
return results;
// ...
String name = results.get(position).getName();
How to check if table is empty using parse , I'm having a problem with the code below :
private String[] getMaxDateMessage() throws ParseException {
final String[] msgData = new String[3];
ParseObject ob = null;
String[] userIds = {currentUserId, recipientId};
ParseQuery<ParseObject> query = ParseQuery.getQuery("ParseMessage");
query.whereContainedIn("senderId", Arrays.asList(userIds));
query.whereContainedIn("recipientId", Arrays.asList(userIds));
query.orderByDescending("createdAt");
if(query.hasCachedResult())
{
ob = query.getFirst();
if (ob.isDataAvailable()) {
//for (int i = 0; i < 1; i++) {
//createdDate[0] = messageList.get(i).get("createdAt").toString();
msgData[0] = ob.getCreatedAt().toString();
msgData[1] = ob.get("senderId").toString();
msgData[2] = ob.get("recipientId").toString();
// }
}
}
The thing is that the table is empty , so the query should return null , but no exception is been throwed , it just crashes the app .
So how can I check if the table is empty before trying to fetch any data ?
Update : The solution that I have found is to use query.count().
If the count returns a value that is not 0 then the table is not empty .
Using query.count() to determine if the table is empty is not an optimal solution. While this is perfectly fine when actually run against an empty table, using query.count() will almost always result in a sub-optimal query when there's more than one object in the table. The reason for this is quite clear: you only care about the first object matched by this query, yet a query.count() will scan the whole table in order to return the total of objects that match your query.
Therefore, the ideal solution is to simply use query.getFirst() and check if you get any results. You should be able to handle the case where ob is not a ParseObject, e.g. the collection is either empty or no objects match your query.
I have these tables in an Android based application where I'm using OrmLite for the database management.
What I want to have an x number of array list depending on how many of the product type FOLDER I have.
So in this case I want to a list of products where the productId equals parentId.
So I want a list where
if(productType = FOLDER) {
if(productId = parentId){
//add product
}
}
Basically what I want to end up with, in this case three lists with each containing a list of products where parentId is the same for every product.
I've tried many things, and some works better than others, but a code I want to run actually throws a nullpointer.
DatabaseHelper dbHelper = getHelper();
List<Product> productsParents = null;
try {
Dao<Product, Integer> dao = dbHelper.getDao();
PreparedQuery<Product> prepQu = dao.queryBuilder().where()
.eq("parentId", dao.queryBuilder().selectColumns("productId").where()
.eq("productType", ProductType.FOLDER).prepare()).prepare();
productsParents = dao.query(prepQu);
} catch (SQLException e) {
...
}
This code isn't working because productParents returns null, and it does not do what I want, even though it's a slight hint. If someone know how to do this in code that would be sufficient also, or more likely a mix of java and ormlite.
Have you had a chance to RTFM around building queries? The ORMLite docs are pretty extensive:
http://ormlite.com/docs/query-builder
Your problem is that a prepared query cannot be an argument to the eq(...) method. Not sure where you saw an example of that form.
So there are a couple ways you can do this. The easiest way is to do a different query for each productType:
Where<Product, Integer> where = dao.queryBuilder().where();
where.eq("parentId", parentId).and().eq("productType", ProductType.FOLDER);
productsParents = where.query();
// then do another similar query again with ProductType.PRODUCT, ...
If you want to do just one query then you can get all products that match the parentId and then separate them using code:
Where<Product, Integer> where = dao.queryBuilder().where();
where.eq("parentId", parentId);
productsParents = where.query();
List<Product> productFolders = new ArrayList<Product>();
List<Product> productProducts = new ArrayList<Product>();
...
for (Product product : productsParents) {
if (product.getProductType() == ProductType.FOLDER) {
productFolders.add(product);
} else if (product.getProductType() == ProductType.PRODUCT) {
productProducts.add(product);
} else ...
}
I receive data from a server using JSON and I want to order them alphabetically with alphabet indexed section and store them in a ListView.
Maybe something that will happen in :
for(int i=0;i<jArray.length();i++){
// here
}
I read that you can order elements like that only using a cursor. In my case would be very inefficient to store the elements from the server in the database and read them again. Waste of time and memory.
So, I am asking you if there could be any solution for my problem : order alphabetically with alphabet indexed section string received from JSON .
EDIT: I want my listview to look like this http://eshyu.wordpress.com/2010/08/15/cursoradapter-with-alphabet-indexed-section-headers/ . I mean with those sections . All tutorials I found said that you need to fetch information with a cursor. My question was if I could't do this wihout a cursor, because it would be a waste of memory to store them in the local database too.
You may need to parse the JSON Array :
List<Project> list = new ArrayList<Project>();
for (int i = 0; i < jArray.length(); i++) {
JSONObject obj = (JSONObject) jArray.get(i);
project = new Project();
project.setId( Long.parseLong(obj.get("id").toString()));
project.setKey(obj.get("key").toString());
project.setName(obj.get("name").toString());
list.add(project);
}
You can use the comparator class like this to sort them :
Collections.sort(list), new Comparator<Project>() {
public int compare(Project p1, Project p2) {
return p1.getKey().compareToIgnoreCase(p2.getKey());
}
});
You can also have Project class and implements Comparable:
public class Project implements Comparable<Project> {
private long id;
private String key;
private String name;
public int compareTo(Project p) {
if (this.key > p.key)
return -1;
else if (this.key < p.key)
return 1;
return 0;
}
}
And then sort the list by Collections.sort(list);
My suggestion is try to sort the data in the Server-side, because the memory of the phone is limited and it may make you application time consuming to show the data, but you do not have memory limitation problem in the Server-side.
use a comparator to sort the arraylist as described here . And then use an ArrayAdapter to show the items in Listview