I'm trying to implement a simple "insert or update" (so-called 'upsert') method in Grails / GORM / mongodb plug-in / MongoDB.
The approach I used with Hibernate (using merge) fails with a duplicate key error. I presume perhaps merge() isn't a supported operation in mongodb GORM, and tried to get to the native upsert method through GMongo.
I finally have a version that works (as posted below), but it is probably not the best way, as adding any fields to the object being saved will break the code silently.
public void upsertPrefix(p) {
def o = new BasicDBObject()
o.put("_id", p.id)
o.put("someValue", p.someValue)
o.put("otherValue", p.otherValue)
// DBObject o = p as DBObject // No signature of method: mypackage.Prefix.keySet() is applicable for argument types: () values: []
db.prefix.update([_id : p.id], o, true, false)
// I actually would want to pass p instead of o here, but that fails with:
// No signature of method: com.gmongo.internal.DBCollectionPatcher$__clinit__closure2.doCall() is applicable for argument types: (java.util.ArrayList) values: [[[_id:keyvalue], mypackage.Prefix : keyvalue, ...]]
/* All of these other more "Hibernatesque" approaches fail:
def existing = Prefix.get(p.id)
if (existing != null) {
p.merge(flush:true) // E11000 duplicate key error
// existing.merge(p) // Invocation failed: Message: null
// Prefix.merge(p) // Invocation failed: Message: null
} else {
p.save(flush:true)
}
*/
}
I guess I could introduce another POJO-DbObject mapping framework to the mix, but that would complicate things even more, duplicate what GORM is already doing and may introduce additional meta-data.
Any ideas how to solve this in the simplest fashion?
Edit #1: I now tried something else:
def existing = Prefix.get(p.id)
if (existing != null) {
// existing.properties = p.properties // E11000 duplicate key error...
existing.someValue = p.someValue
existing.otherValue = p.otherValue
existing.save(flush:true)
} else {
p.save(flush:true)
}
Once again the non-commented version works, but is not well maintainable. The commented version which I'd like to make work fails.
Edit #2:
Version which works:
public void upsertPrefix(p) {
def o = new BasicDBObject()
p.properties.each {
if (! (it.key in ['dbo'])) {
o[it.key] = p.properties[it.key]
}
}
o['_id'] = p.id
db.prefix.update([_id : p.id], o, true, false)
}
Version which never seems to insert anything:
def upsertPrefix(Prefix updatedPrefix) {
Prefix existingPrefix = Prefix.findOrCreateById(updatedPrefix.id)
updatedPrefix.properties.each { prop ->
if (! prop.key in ['dbo', 'id']) { // You don't want to re-set the id, and dbo is r/o
existingPrefix.properties[prop.key] = prop.value
}
}
existingPrefix.save() // Never seems to insert anything
}
Version which still fails with duplicate key error:
def upsertPrefix(p) {
def existing = Prefix.get(p.id)
if (existing != null) {
p.properties.each { prop ->
print prop.key
if (! prop.key in ['dbo', 'id']) {
existingPrefix.properties[prop.key] = prop.value
}
}
existing.save(flush:true) // Still fails with duplicate key error
} else {
p.save(flush:true)
}
}
Assuming you have either an updated version of the object, or a map of the properties you need to update with their new values, you could loop over those and apply the updates for each property.
Something like this:
def upsert(Prefix updatedPrefix) {
Prefix existingPrefix = Prefix .findOrCreateById(updatedPrefix.id)
updatedPrefix.properties.each { prop ->
if (prop.key != 'id') { // You don't want to re-set the id
existingPrefix.properties[prop.key] = prop.value
}
}
existingPrefix.save()
}
How to exclude updating the ID may not be quite correct, so you might have to play with it a bit. You also might consider only updating a property if it's corresponding new value is different from the existing one, but that's essentially just an optimization.
If you have a map, you might also consider doing the update the way the default controller scaffolding does:
prefixInstance.properties = params
MongoDB has native support for upsert. See the findAndModify Command with upsert parameter true.
Related
I have a simple association table whose PK isn't referenced anywhere but when I am trying to delete a record from it in the following way, I get an error. I am using EF code-first. Any help would be very very helpful. Thanks in advance.
List<ViolationTypeNOV> novRels = UnitOfWork.Context.ViolationTypeNOVs.Where(x => x.NOVId == nov.NOVId).Include("ViolationType").Include("NOV").ToList();
foreach (ViolationTypeNOV o in novRels)
{
UnitOfWork.Context.ViolationTypeNOVs.Remove(o);
}
UnitOfWork.Context.SaveChanges();
Here is the error message I am getting. If the table's PK isn't referenced in any way, why is it failing with this error? Just not able to understand:
The operation failed: The relationship could not be changed because one or more of the foreign-key properties is non-nullable. When a change is made to a relationship, the related foreign-key property is set to a null value. If the foreign-key does not support null values, a new relationship must be defined, the foreign-key property must be assigned another non-null value, or the unrelated object must be deleted.
The same thing if I am running through SSMS same thing is working though:
DELETE ViolationTypeNOVs
WHERE ViolationTypeNOVId = 2
Why?
But again if I am running the same query through context as below, I get the same error at the calling SaveChanges:
foreach (ViolationTypeNOV o in novRels)
{
string str = string.Format("Delete ViolationTypeNOVs where ViolationTypeNOVId = {0}", new object[] { o.ViolationTypeNOVId });
UnitOfWork.Context.Database.ExecuteSqlCommand(str);
}
UnitOfWork.Context.SaveChanges();
It seems like some of the objects within the context aren't nulling or getting deleted, is there anyway to clear them all in one go? Because these ids/objects are used in multiple places in the code - please let me know how to clear them all - thanks a lot.
Any help please?
Fixed it - the problem is we need to clear up all the objects and its links that the parent object is using then only we can save the changes thanks here is my solution
public bool Delete(NOV nov, bool performCommit = true)
{
System.Data.Entity.DbContextTransaction dbOperation = null;
if (performCommit)
dbOperation = UnitOfWork.BeginTransaction();
try
{
//-- Remove the Items - "foreach" approach was a problem
// http://weblogs.asp.net/ricardoperes/entity-framework-pitfalls-deleting-orphans
//------------------------------------------------------
// Remove the Violations that are in this NOV
//------------------------------------------------------
List<Violation> violationIdlist = new List<Violation>();
foreach (var v in nov.ViolationNOVs)
{
var a = UnitOfWork.ViolationRepository.GetAll().Where(z => z.ViolationId == v.ViolationId).FirstOrDefault();
violationIdlist.Add(a);
}
foreach (var v in violationIdlist)
{
var a = nov.ViolationNOVs.Where(x => x.NOVId == nov.NOVId && x.ViolationId == v.ViolationId)?.FirstOrDefault();
nov.ViolationNOVs.Remove(a);
}
nov.IssuedBy.Clear();
//deleting all OneToMany references to NOV
List<ViolationTypeNOV> novRels = UnitOfWork.Context.ViolationTypeNOVs.Where(x => x.NOVId == nov.NOVId).Include("ViolationType").Include("NOV").ToList();
nov?.ViolationTypeNOVs?.Clear();
//foreach (ViolationTypeNOV o in novRels)
//{
// UnitOfWork.Context.ViolationTypeNOVs.Remove(o);
// o?.ViolationType?.ViolationTypeNOVs?.Remove(o);
// nov?.ViolationTypeNOVs?.Remove(o);
//}
UnitOfWork.Context.ViolationTypeNOVs.RemoveRange(novRels);
List<ViolationNOV> violationNOVs = UnitOfWork.Context.ViolationNOVs.Where(x => x.NOVId == nov.NOVId).Include("Violation").Include("NOV").ToList();
nov?.ViolationNOVs?.Clear();
UnitOfWork.Context.ViolationNOVs.RemoveRange(violationNOVs);
List<CaseNOV> caseNOVs = UnitOfWork.Context.CaseNOVs.Where(x => x.NOVId == nov.NOVId).Include("Case").Include("NOV").ToList();
nov?.CaseNOVs?.Clear();
UnitOfWork.Context.CaseNOVs.RemoveRange(caseNOVs);
UnitOfWork.Context.SaveChanges();
if (dbOperation != null)
dbOperation.Commit();
LogHandler.LogInfo(2521, "Deleted NOV " + nov.NOVNumber);
return true;
}
catch (Exception ex)
{
LogHandler.LogError(2523, "Commit Fail in NOV Delete", ex);
if (dbOperation != null)
dbOperation.Rollback();
throw ex;
}
}
This statement has fixed the problem: UnitOfWork.Context.ViolationTypeNOVs.RemoveRange(novRels); thanks a lot for everybody who tried to help me
I am trying to look up record using if I have the key then use Find if not use Where
private ApplicationDbContext db = new ApplicationDbContext();
public bool DeactivatePrice(int priceId = 0, string sponsorUserName = "")
{
var prices = db.BeveragePrices;
// if we have an id then find
if (priceId != 0)
{
prices = prices.Find(priceId);
}
else
{
prices = prices.Where(b => b.UserCreated == sponsorUserName);
}
if (prices != null)
{
// do something
}
return true;
I get the following error for
prices = prices.Find(priceId);
Cannot convert app.Model.BeveragePrices from system.data.entity.dbset
I am copying the pattern from this answer but something must be different.
Seems you forgot to put a predicate inside the Find function call. Also you need to do ToList on the collection. The second option is a lot more efficient. The first one gets the whole collection before selection.
Another note commented by #Alla is that the find returns a single element. So I assume another declaration had been made for 'price' in the first option I state down here.
price = prices.ToList.Find(b => b.PriceId == priceId);
Or
prices = prices.Select(b => b.PriceId == priceId);
I assume the field name is PriceId.
I want to copy object properties to another object in a generic way (if a property exists on target object, I copy it from the source object).
My code works fine using ExpandoMetaClass, but I don't like the solution. Are there any other ways to do this?
class User {
String name = 'Arturo'
String city = 'Madrid'
Integer age = 27
}
class AdminUser {
String name
String city
Integer age
}
def copyProperties(source, target) {
target.properties.each { key, value ->
if (source.metaClass.hasProperty(source, key) && key != 'class' && key != 'metaClass') {
target.setProperty(key, source.metaClass.getProperty(source, key))
}
}
}
def (user, adminUser) = [new User(), new AdminUser()]
assert adminUser.name == null
assert adminUser.city == null
assert adminUser.age == null
copyProperties(user, adminUser)
assert adminUser.name == 'Arturo'
assert adminUser.city == 'Madrid'
assert adminUser.age == 27
I think the best and clear way is to use InvokerHelper.setProperties method
Example:
import groovy.transform.ToString
import org.codehaus.groovy.runtime.InvokerHelper
#ToString
class User {
String name = 'Arturo'
String city = 'Madrid'
Integer age = 27
}
#ToString
class AdminUser {
String name
String city
Integer age
}
def user = new User()
def adminUser = new AdminUser()
println "before: $user $adminUser"
InvokerHelper.setProperties(adminUser, user.properties)
println "after : $user $adminUser"
Output:
before: User(Arturo, Madrid, 27) AdminUser(null, null, null)
after : User(Arturo, Madrid, 27) AdminUser(Arturo, Madrid, 27)
Note: If you want more readability you can use category
use(InvokerHelper) {
adminUser.setProperties(user.properties)
}
I think your solution is quite good and is in the right track. At least I find it quite understandable.
A more succint version of that solution could be...
def copyProperties(source, target) {
source.properties.each { key, value ->
if (target.hasProperty(key) && !(key in ['class', 'metaClass']))
target[key] = value
}
}
... but it's not fundamentally different. I'm iterating over the source properties so I can then use the values to assign to the target :). It may be less robust than your original solution though, as I think it would break if the target object defines a getAt(String) method.
If you want to get fancy, you might do something like this:
def copyProperties(source, target) {
def (sProps, tProps) = [source, target]*.properties*.keySet()
def commonProps = sProps.intersect(tProps) - ['class', 'metaClass']
commonProps.each { target[it] = source[it] }
}
Basically, it first computes the common properties between the two objects and then copies them. It also works, but I think the first one is more straightforward and easier to understand :)
Sometimes less is more.
Another way is to do:
def copyProperties( source, target ) {
[source,target]*.getClass().declaredFields*.grep { !it.synthetic }.name.with { a, b ->
a.intersect( b ).each {
target."$it" = source."$it"
}
}
}
Which gets the common properties (that are not synthetic fields), and then assigns them to the target
You could also (using this method) do something like:
def user = new User()
def propCopy( src, clazz ) {
[src.getClass(), clazz].declaredFields*.grep { !it.synthetic }.name.with { a, b ->
clazz.newInstance().with { tgt ->
a.intersect( b ).each {
tgt[ it ] = src[ it ]
}
tgt
}
}
}
def admin = propCopy( user, AdminUser )
assert admin.name == 'Arturo'
assert admin.city == 'Madrid'
assert admin.age == 27
So you pass the method an object to copy the properties from, and the class of the returned object. The method then creates a new instance of this class (assuming a no-args constructor), sets the properties and returns it.
Edit 2
Assuming these are Groovy classes, you can invoke the Map constructor and set all the common properties like so:
def propCopy( src, clazz ) {
[src.getClass(), clazz].declaredFields*.grep { !it.synthetic }.name.with { a, b ->
clazz.metaClass.invokeConstructor( a.intersect( b ).collectEntries { [ (it):src[ it ] ] } )
}
}
Spring BeanUtils.copyProperties will work even if source/target classes are different types. http://docs.spring.io/autorepo/docs/spring/3.2.3.RELEASE/javadoc-api/org/springframework/beans/BeanUtils.html
I'm writing a tinyMce plugin which contains a section of code, replacing one element for another. I'm using the editor's dom instance to create the node I want to insert, and I'm using the same instance to do the replacement.
My code is as follows:
var nodeData =
{
"data-widgetId": data.widget.widgetKey(),
"data-instanceKey": "instance1",
src: "/content/images/icon48/cog.png",
class: "widgetPlaceholder",
title: data.widget.getInfo().name
};
var nodeToInsert = ed.dom.create("img", nodeData);
// Insert this content into the editor window
if (data.mode == 'add') {
tinymce.DOM.add(ed.getBody(), nodeToInsert);
}
else if (data.mode == 'edit' && data.selected != null) {
var instanceKey = $(data.selected).attr("data-instancekey");
var elementToReplace = tinymce.DOM.select("[data-instancekey=" + instanceKey + "]");
if (elementToReplace.length === 1) {
ed.dom.replace(elementToReplace[0], nodeToInsert);
}
else {
throw new "No element to replace with that instance key";
}
}
TinyMCE breaks during the replace, here:
replace : function(n, o, k) {
var t = this;
if (is(o, 'array'))
n = n.cloneNode(true);
return t.run(o, function(o) {
if (k) {
each(tinymce.grep(o.childNodes), function(c) {
n.appendChild(c);
});
}
return o.parentNode.replaceChild(n, o);
});
},
..with the error Cannot call method 'replaceChild' of null.
I've verified that the two argument's being passed into replace() are not null and that their parentNode fields are instantiated. I've also taken care to make sure that the elements are being created and replace using the same document instance (I understand I.E has an issue with this).
I've done all this development in Google Chrome, but I receive the same errors in Firefox 4 and IE8 also. Has anyone else come across this?
Thanks in advance
As it turns out, I was simply passing in the arguments in the wrong order. I should have been passing the node I wanted to insert first, and the node I wanted to replace second.
So how does one obtain the previous value of a custom field in a Jira IssueEventListener? I am writing a custom handler for the issueUpdated(IssueEvent) event and I would like to alter the handler's behavior if a certain custom field has changed. To detect the type of change I would like to compare the previous and current values.
(I'm am not asking about how to obtain its current value - I know how to get that from the related Issue)
I am developing against Jira 4.0.2 on Windows.
Is the best way to scan the change history for the last known value?
List changes = changeHistoryManager.getChangeHistoriesForUser(issue, user);
I'm assuming the original poster is writing a JIRA plugin with Java. I cannot be certain of how to accomplish this task in JIRA v4.0.2, but here is how I managed to do so with JIRA v5.0.2 (the solutions may very well be the same):
public void workflowEvent( IssueEvent event )
{
Long eventTypeId = event.getEventTypeId();
if( eventTypeId.equals( EventType.ISSUE_UPDATED_ID ) )
{
List<GenericValue> changeItemList = null;
try
{
changeItemList = event.getChangeLog().getRelated( "ChildChangeItem" );
}
catch( GenericEntityException e )
{
// Error or do what you need to do here.
e.printStackTrace();
}
if( changeItemList == null )
{
// Same deal here.
return;
}
Iterator<GenericValue> changeItemListIterator = changeItemList.iterator();
while( changeItemListIterator.hasNext() )
{
GenericValue changeItem = ( GenericValue )changeItemListIterator.next();
String fieldName = changeItem.get( "field" ).toString();
if( fieldName.equals( customFieldName ) ) // Name of custom field.
{
Object oldValue = changeItem.get( "oldvalue" );
Object newValue = changeItem.get( "newvalue" );
}
}
}
}
Some possible key values for changeItem are:
newvalue
oldstring
field
id
fieldtype
newstring
oldvalue
group
For many of the custom field types Object oldValue is probably just a String. But I don't think that's true for every case.
Try this example :
String codeProjetOldValue= "";
List<GenericValue> changeItemList = issueEvent.getChangeLog().getRelated("ChildChangeItem");
for (GenericValue genericValue : changeItemList) {
if(champCodeProjet.equals(genericValue.get("field"))){
codeProjetOldValue=genericValue.getString("oldstring");
break;
}
}
Note that : champCodeProjet is the name of customfield.