I'm trying to add a string value and a customized horizontal field manager to a HashTable. String being the key and the custom horizontal field manager being the value. I do remove the pair from the Hashtable whenever the horizontal field managers are removed from the screen. Also, I'm not persisting the Hashtable.
I get the error JVM Error 525 : Bad persistent object after a few minutes on the simulator. When I try to run the same build on a device it restarts after a few minutes of creation of the Hashtable. Please note that everything works fine if I just don't push anything in the Hashtable.
So my question is, is it wrong to put a Horizontal Field Manager or something extended from it in a HashTable? If not, what could be the problem here?
Make sure that all of the objects being persisted (including those added to the HashTable) implement the Persistable interface.
Related
I have created a entity called #USER-NAME and have set that as a requirement.
Now, for the first time when the entity is detected in the conversation - say, "I am John" , then the memory is set to John. On subsequent encounter of the same entity with different value - "I am Dave", the memory remains unchanged.
I have seen the edit memory option, which provides 1. reset memory 2. set to a value . For the option 2, it does not provide a way to set to the value of #USER-NAME, instead only provides option to enter static values.
How can I update the memory every time the value of the entity changes ??
EDIT
Hi, I am attaching some screenshots to show what's exactly going wrong.
I have a Entity named '#USER_NAME' that saves the user name in a memory variable .
I make the following conversation -
The JSON payload after the conversation is as follows. This works perfectly-
I update the conversation again by providing a new user name.
This triggers the entity just fine. You can see the entity being detected properly.
However, the memory value remains the same.
What I wanted was the memory variable to replace 'Dev' with 'John'.
Remember that:
memory <> Intent
You can set memory in the message section or update automatically using for example a requirement in this case every time the skill is trigged it will replace the value in the memory ID
EDIT: Because the set memory field expect a JSON you can't use memory as you want, but if you reset that memory ID shomewhere relevant in the chat (in my sample I delete it right after saying Hi XXX) so when the skill is trigged again it will "replace" it with the new value
In the Requirement I set the golden entity #Person to variable "name" and if is missing I ask her name.
Sample Image
the memory is a persistent object so if you want to reset it you need either to have specific conditions within the builder or go through a webhook to have a backend code to reste the memory.
I know that UserDefaults are meant simply to save preferences, but these preferences are persistent - the values saved in UserDefaults are maintained over an unlimited number of app launches and are always available to be read as long as the app remains installed... right?
Is it possible that these values will be cleared or will not be correctly accessed at any point? After years of using UserDefaults and depending on the consistency of the values they hold, I have now seen twice in one day of work that when my app launched and checked a simple boolean value, the value was not correct.
if defaults.bool(forKey: "beenLaunched") {
This code runs each time the app launches. If the value is true, I do nothing more, but if it is false, I set a few values as this is the user's very first launch of the app and then I call defaults.set(true, forKey: "beenLaunched") and defaults.set(0, forKey: "eventsCompleted") and a few other values.
I found this thread on the Apple forums in which Eskimo said "For the central NSUserDefaults method, -objectForKey:, a result of nil means that the value is unavailable, but there’s no way to distinguish between this key is not present and this value can’t be fetched because the user defaults are offline." (This appears to be in reference to a specific case of background launching while a device is locked)
I can look into a more secure way of saving simple data such as a Bool value, an Int, or a String, but using UserDefaults for these types of values has always been simple, straightforward, and reliable. Can anybody chime in on the matter and if I was wrong to believe in UserDefaults' persistence?
Thank you!
UserDefaults isn't a "service"; it's never not available to your application. The file it writes to is a PLIST (and therefore all values are stored according to the PLIST standard). For example, all numbers (including booleans) are stored as an NSNumber to the file and can be retrieved either by object(forKey:) or bool(forKey:). If you use the object method and nothing is set for that value you get nil, whose boolean value is false (or 0). Same if you use the boolean method (you get false). This means no matter which way you go, you'll always get false if there's no value or a value of false. Design your logic around that (which you already have - "beenLaunched" will be empty and therefore false if it's never been launched) and you should be fine.
As for the suggestion of synchronize(), ignore it. Unless you're doing something really weird with threads and preference access or you've interrupted the application immediately after setting a value/object for the problem key, it's got nothing to do with this. Per the very first paragraph of the docs, synchronize() is called periodically as needed. In practice, it's called pretty much immediately after a change occurs.
For context, none of my apps have ever called synchronize() and some of them are old enough to drive. Never a single problem. If you don't have a very good justification for calling synchronize() yourself you almost certainly don't need it and attempts to explain why you do need to sprinkle it everywhere are ... often amusing.
In your specific case, the value stuck by after first run multiple times then suddenly didn't once. Is it possible you changed your app's bundle identifier or name? The defaults are stored by identifier+name so a change would effectively "reset" your app's defaults. Have you been running your app in the simulator and did you just reset-content-and-settings in the simulator? On your device and deleted the app before re-running it on-device?
If you are working in swift then returning nil means objectforkey has not been assigned any value at all . In other case it always returns proper value if you casted saved value properly.
And userdefaults is always available to use, it can never goes offline.
With help from the great posts here, I understand the error. But I need some clarification please.
Say my managed object context (schema) has 3 tables (entities), and say each entity had 3 attributes of which one attribute for each entry is NOT optional.
So now for the first time, my app creates a managed object for the first entity, filling its mandatory attribute; app has not created managed objects for the second and third entities yet - didn't have to yet. When I try to save the context at this point, I get error code 1570. Is it because I have not filled out values for the second and 3rd entities?
I am not sure if this helps you.
But The cocoa error 1570 means that mandatory fields are not filled in. So please make sure your mandatory fields are not nil.
iphone Core Data Unresolved error while saving
Yep it was a mandatory field that was not filled in. The post above showed me which field.
If you have previously run it on your device (or simulator) with the attribute set as mandatory and then changed it to optional, delete the app from the device before running again. That was the problem for my app.
Hope that helps someone! :-)
In my case, I was setting a mandatory BOOL property directly as YES or NO, but you should use
NSNumber numberWithBOOL
in order to make it work.
I'm encountering the following problem - I have simple GWT overlay types, and I'm trying to convert one to a JSON string on the client; I'm simply doing:
new JSONObject(this).toString();
The conversion works, but it adds an additional, incorrect key to the json string, such as:
{"key1":"value1", "key2":value2, "$H":1}
where "$H":1 doesn't correspond to anything in my overlay type.
Any idea why this is?
Any help is appreciated on this, thanks.
This issue is define in this link
The $H property comes from the
implementation of
JavaScriptObject#hashCode() (in
com.google.gwt.cire.client.impl.Impl#getHashCode(Object)).
In your case, this is due to
AbstractEditableCell maintaining a map
of value keys to their "view data",
and your use (I guess) of the default
ProvidesKey implementation
(SimpleProvidesKey) which directly
returns the item.
So, when rendering, the EditTextCell
calls getViewData, which looks up the
key in the map (and thus needs the
hashcode of the key, hence the call to
hashCode), and the key is your JSO
(hence the new $H property).
I believe that giving a ProvidesKey
implementation (in you case, returning
the name property for instance) to the
Celltable would solve your issue.
What is wrong in this code?
I was expected "titi" in person.name but I still have "toto"!
More explicitly, how to modify a record in a function?
init1()->
S=#person{name="toto"}, %record creation and field setting
fct(S),
io:format("~s~n",[S#person.name]).
fct(R)->
R#person{name="titi"}. %record updating
You need to get a result of fct():
init1()->
S=#person{name="toto"}, %record creation and field setting
S2 = fct(S), % Get updated record
io:format("~s~n",[S2#person.name]).
fct(R)->
R#person{name="titi"}. %record updating
Bertaud, I think you are getting ahead of yourself a bit. You really need to understand the basics of immutability before you write any more code. (i.e. "variables" do not vary : you can only assign a value to them once.) I suggest you read the free online guide "Learn You Some Erlang For Great Good", at http://learnyousomeerlang.com/. The section that covers the basics of variables is http://learnyousomeerlang.com/starting-out-for-real#invariable-variables.
It is impossible to stress too much that all data in Erlang is immutable. So to do something like in your original question you need to modify it like #hdima did. The record is not updated but rewritten. In the same way there is no global data in Erlang, all data belongs to a process. This is even true of ETS tables as they basically behave like a process, albeit a built-in one without explicit communication.
So if you use the process dictionary or an ETS table the data itself can never be updated, only the dictionary/table. This means that to modify some data in the dictionary/table you basically have to:
"Read" the data
Update the data making new data
"Write" the new back into the dictionary/table
Without writing the new data back into the dictionary/table it will be lost, as your new data was.
Within fct(), you're not mutating the record, but you're returning a new value for the record, which needs to be used further. If you're calling fct(S), without handling the return value, then you'll lose that new value ("titi").