I would just like to clarify that by 'design', I mean software design, not UI design.
I have an application similar to the native settings app. The problem I have with it is it doesn't follow the same clear-cut MVC style. Other apps tend to focus around displaying one kind of thing. In the case of a periodic table app for example, it's elements. The elements clearly comprise the model, and they share similar properties and behaviours, meaning they can be displayed and interacted with identically. An app like this almost designs itself!
My app, like the settings apps, consists of an arbitrary selection of rows displaying dissimilar data in dissimilar ways. One row might contain a switch, the other might modally present a very specific view when tapped. They're all very different.
How do you design something like this?
At the moment, I'm doing it all in the view controller, and the relevant rows are being tracked via an enum:
enum {
kNameRow,
kGenderRow,
kJobTypeRow,
kLevelOfExerciseRow,
kEmailAddressRow,
kTelephoneNumberRow
};
As I described, these cells are all very different, so displaying cells is handled like this:
// - tableView:cellForRowAtIndexPath pseudocode.
switch (indexPath.row) {
case kNameRow: // create name cell.
case kGenderRow: // create gender cell.
case kJobTypeRow: // create job type cell.
case kLevelOfExerciseRow: // create level of exercise cell.
case kEmailAddressRow: // create email address cell.
case kTelephoneNumberRow: // create telephone number cell.
}
And interacting with cells is handled similarly:
// - tableView:didSelectRowAtIndexPath pseudocode.
switch (indexPath.row) {
case kNameRow: // do name-specific stuff.
case kGenderRow: // do gender-specific stuff.
case kJobTypeRow: // do job type-specific stuff.
case kLevelOfExerciseRow: // do level of exercise-specific stuff.
case kEmailAddressRow: // do email address-specific stuff.
case kTelephoneNumberRow: // do telephone number-specific stuff.
}
This seems hugely unwieldy, and has the added of problem of not working when the table is broken down into multiple sections.
Is there a better way to do this? Are there any design patterns I would benefit from using when working with big tables of largely unrelated data?
Any tips at all are hugely appreciated.
I've become fond of implementing section controllers that pull the logic out of you UITableViewController subclass (or other hosting controller) and move them into self-contained classes.
I ended up implementing a base protocol that defines what a section controller needs to do - for me, that includes the number of rows in a section and a cell for the row (don't need the whole index path since the controller deals with a single section). I've got optional method for returning a section name and row height. That's all I've implemented so far since that's all I've actually needed.
It works for me because my individual sections tend to be homogeneous, but you could easily use the idea to return heterogeneous cells within the same section or refactor the idea to have cell type controllers instead of section controllers. In the end, my UITableViewDelegate and UITableViewDataSource methods just need to figure out which section controller to call instead of embedded all the logic within the UITableViewController subclass.
I think I got the idea from this article, but I also saw a more recent article that describes the same idea.
you might want to look at coreyfloyds project http://github.com/coreyfloyd/Generic-Heterogeneous-Table-Views i think this might have the functionality you need.
Here's my suggestion - handle each cell as a member of the view.
lol, it's been a while since I've used a table, so I could just be talkin' crap here but give it a try.
instead of an enum use:
NSThingyCell *nameRow;
NSThingyCell *genderRow;
#property IBOutlet NSThingyCell *nameRow;
#property IBOutlet NSThingyCell *genderRow;
- (IBAction) nameRowChanged:(id)sender;
- (IBAction) genderRowChanged:(id)sender;
and then instead of a table call with a switch, just wire each individual cell up in Interface Builder.
This has the added benefit of being row-independent, so if you have to put "ageRow" in between name and gender, nothing gets screwed up.
This will also get pretty big, so if your view has several tables, you may want to consider splitting those tables out into separate nibs/controllers and loading the views at run-time.
Have you ever thought of simply having an array of objects for a class which contains a UI element and some other identifiable data?
#interface settingsOption {
NSString *key;
UIView *displayElement;
}
+ (settingsOption *)optionWithKey:(NSString *)key andDisplayElement:(UIView *)displayElement;
#property (nonatomic, retain) UIView *displayElement;
#property (nonatomic, retain) NSString *key;
#end
Where the class method would look like
+ (settingsOption *)optionWithKey:(NSString *)key andDisplayElement:(UIView *)displayElement;
settingsOption *option = [[settingsOption alloc] init];
option.key = key;
option.displayElement = displayElement;
return [option autorelease];
}
Your settings class would have an array of settingsOption instances.
- (void)somewhereInMySettingsClass
mySettings = [[NSMutableArray alloc] init];
[mySettings addObject:[settingsOption optionWithKey:#"age" andDisplayElement:[UIButton buttonWithStyle:UIButtonStyleRect]]];
[mySettings addObject:...];
}
The table's cellForRowAtIndexPath would just do
[cell addSubview:[[mySettings objectAtIndex:indexPath.row] displayElement]];
You were talking about sections, though, which would add another layer to the data. This might simply be a matter of splitting mySettings into an array of arrays instead, where each array in the array is one section.
Not sure if I missed anything above. Feel free to point and poke.
You might simplify the settingsOption class further by adding more helper classes for various types of elements, e.g.
+ (settingsOption *)buttonWithKey:(NSString *)key;
+ (settingsOption *)switchWithKey:(NSString *)key;
+ (settingsOption *)pickerWithKey:(NSString *)key withDataSource:(id <UIPickerViewDataSource>)source withDelegate:(id <UIPickerViewDelegate>)delegate;
etc etc.
Related
In the app I'm making I need to be able to store input from textfields, then show them in a table view with the ability, for the user, to delete or edit the input, and add more.
I have already made it possible to press a button -> next view, insert numbers and choose whatever option you want, and then press ADD. But I have no idea how to store and show the input.
The app is supposed to work this way:
You open the app, and you can press a PLUS sign. Then you are taken to a modal-view where get the option to choose from a slider, and input two different numbers in two textfields. Then press ADD. After the ADD-button is pressed, you are then taken to the first view, and the numbers and option you chose from the slider are then shown in a table view, with yet another plus sign beneath. So you can repeat the whole thing over and over.
After the above, there will be a "Calculate" button, where it calculates the different inputs, and gives an end result, but I will try and figure that one out myself, when I get the basic stuff working.
I'm looking forward to your answers, and I hope you will explain things thoroughly, as I am new at this field.
And please, don't just redirect me to , I have read a lot, but nothing that covered my problem. And I'm not that good at figuring out things myself, based on e.g. Apple's own reference library .
You are going to need someway to store all that info. If you need to sort or query against it in any way, use SQLite or you could use Core Data, but I never have.
More simply, you could instantiate an NSMutableArray, and make it property of the parent view, and when you push the child view to add the info, add a custom object that holds your info, and add it to the array. Tables are quite easy to use with array's, then implement the table methods to rearrange and delete them.
- (void)moveRowAtIndexPath:(NSIndexPath *)indexPath toIndexPath:(NSIndexPath *)newIndexPath
As far as calculating out totals or whatever, just loop through all the objects in that array, and do whatever you need to.
EDIT:
Example of a custom class:
MyClass
#property (nonatomic, retain) NSString *infoString1;
#property (nonatomic, retain) NSString *infoString2;
#property (nonatomic) int sliderValue;
How you could store the info
MyClass *object1 = [[MyClass alloc] init];
object1.infoString1 = textField.text;
object1.infoString2 = textField2.text;
object1.sliderValue = [theSlider.value intValue] //Dont think thats correct code, but you get the idea
[parentView.theArray addObject:object1];
// then dismiss the view
If I am using a custom UITableViewCell I can use the following methods to change the cell's appearance when transitioning state:
- (void)willTransitionToState:(UITableViewCellStateMask)state
- (void)didTransitionToState:(UITableViewCellStateMask)state
Is there a way to achieve this if I'm not using a custom tableview cell?
Edit: Please see Daniel Hanly's comment. Categories may be selectively applied using #import. My apologies to anyone that may have been misled by this answer. Relevant sections will be redacted for future reference.
Okay, attempt number two. As far as I am aware, there is no other documented way to implement the functionality you require without subclassing UITableViewCell. It's worth noting that Apple's docs on UITableViewCell specifically mention that the state transition methods are meant to be implemented by subclasses. That having been said, If you absolutely need to implement them without a subclass, there are a couple of less conventional solutions. Each comes with its own issues, and it may end up being unfeasible for you to implement them, but it's an interesting question nonetheless.
Disclaimer
If you only want a sane and relatively simple explanation, then consider the answer to your question to be "no, there is no way to do what you want." I only present the options below with the assertion that they will work. In no way do I endorse actually using them. Consider this my penance for providing my first answer with such an obvious flaw.
Option One - Categories
It is possible to get the functionality you're looking for by overriding the methods you listed in a custom UITableViewCell category.
The problem is that this approach would be a pretty bad idea 99% of the time. Once you define the category on UITableViewCell, those methods would be defined for all UITableViewCell objects throughout the app. Unless you want the exact same state transition functionality for every single table cell in the app, this approach isn't very helpful.
Option Two - Runtime magic
You can use the low-level Objective-C runtime functions to change the implementation of any method on the fly. Unlike the categories option, this approach is flexible enough to redefine the intended behavior whenever you need to, instead of being a one-shot deal.
For example, if you're trying to manage state transitions from a UITableViewController, you could do this:
CustomTableViewController.m
#import <objc/runtime.h>
- (void) customStateWillChange:(UITableViewCellStateMask)state
{
//custom UITableViewCell code
}
- (void) viewDidAppear:(BOOL)animated
{
//Store the original implementation
Method originalStateWillChangeMethod = class_getInstanceMethod([UITableViewCell class], #selector(willTransitionToState:));
originalStateWillChangeImplementation = method_getImplementation(originalStateWillChangeMethod); //variable declared in header file as type IMP
//Get the new implementation
Method newStateWillChangeMethod = class_getInstanceMethod([self class], #selector(customStateWillChange:));
IMP newStateWillChangeImplementation = method_getImplementation(newStateWillChangeMethod);
//Replace implementation
method_setImplementation(originalStateWillChangeMethod, newStateWillChangeImplementation);
//the rest of your viewDidAppear code
[super viewDidAppear:animated];
}
- (void) viewDidDisappear:(BOOL)animated
{
//restore the original implementation
Method originalStateWillChangeMethod = class_getInstanceMethod([UITableViewCell class], #selector(willTransitionToState:));
method_setImplementation(originalStateWillChangeMethod, originalStateWillChangeImplementation);
//rest of viewDidDisappear code
[super viewDidDisappear:animated];
}
This code may not suit your exact purposes, but I think it provides a useful example.
It's incredibly ugly though because the customStateWillChange: method defined here is only intended to be run as a part of the UITableViewCell class, but in this example it will be compiled as though it were part of the CustomTableController class. Among other annoyances, you would have to eschew the property dot notation, ignore compiler warnings and give up most if not all compile-time checks for that method's body.
Option 3 - Category with runtime magic
Exactly what it sounds like. Define any custom state change methods you like within a category (or several categories) on UITableViewCell. Be sure that each one has a separate name - adding two categories that each have a method of the same name will result in undefined behavior. Also, each one needs to have the same return type and argument types as the method it is intended to replace.
Then the references to [self class] in the above code would be replaced with [UITableViewCell class], and the customStateWillChange: method would be moved to the custom category. While still ugly, you can at least rely on the compiler to interpret the method bodies properly.
Of course, messing with the runtime adds a whole lot of complexity to keep track of. It could work fine, but it's not good design, it would take serious effort to ensure it worked safely and correctly, and it would be likely to bring anguish and despair to anyone maintaining it.
References
The Objective-C Programming Language - Categories and Extensions
Objective-C Runtime Reference
Absolutely. The UITableViewDelegate protocol specifies a number of methods to manage state transitions for the table view's cells. Take a look at the UITableViewDelegate Class Reference, specifically the methods listed under the heading "Editing Table Rows".
Edit
Sorry, you're right. The UITableViewDelegate methods don't respond to direct changes to the cell's properties. I've found a way that does work, but I'm going to put it in a different answer to avoid confusion.
I have a model that essentially stores contact info: phone, twitter, address, etc.
I need to show that data in a UITableView, so I need essentially 3 pieces of information: a label ("Phone"), the actual data ("(123) 456-789") and a URL to handle that data with the right application ("tel://whatever").
I was thinking of using 3 arrays, one for each piece of info, and going through each attribute, see if it's given (could be optional), and if so, add stuff to the 3 arrays for that attribute.
Then, in the data source for the table view, each row can reference an index in the 3 arrays to populate the cell.
The order is important so I can show address before phone, etc.
Can anyone come up with a more elegant way to design the model so that I may not need 3 arrays, or lots of:
if(self.phone != nil)
{
[labels addObject:#"foo"];
[descs addObject:self.phone];
[urls addObject:#"baz"];
}
There will only be one "Contact" that can have a varying number of attributes, depending on some outside condition. So each attribute is optional. The table view will show the attributes of that one contact, in a certain order.
Thanks!
I believe you should create a model class, with properties for each attribute. Otherwise this will come back and bite your ass some day.
Storing the values in an array, means you have stored presentational meaning into the model. Breaking the MVC pattern.
This also means you can have some actual logic in your model object, which will reduce potential defects allot. As an example:
// Contact.h
#interface Contact : NSObject {
}
#property(nonatomic, copy) NSString* phoneNumber;
#property(nonatomic, readonly) NSURL phoneNumberURL;
#end
// Contact.m
#implementation Contact
#synthesize phoneNumber;
-(NSURL*)phoneNumberURL {
return [NSURL URLWithString:
[#"tel:" stringByAppendingString:self.phoneNumber]];
}
#end
This way you do not have to update the URLs from outside the model object. The model object should be able to derive this information itself. This will ensure that your application logic is not spread out into dozens of different places, but are kept in well defined pockets of logic that are easy to maintain.
This seems to be a common problem, but I can't figure out anything from the answers I've seen so far. I have an iPhone app that uses a subclass of NSMutableArray to store objects, plus some additional properties. The subclass is skhCustomArray. The subclass initializes fine, with no objects in the skhCustomArray, and I assign it to the the property of my view controller, which is a pointer to an skhCustomArray.
prescriptionListVC* newPrescList = [[prescriptionListVC alloc] initWithNibName:#"PrescriptionList" bundle:nil];
newPrescList.curPersonPrescriptions = [personDetails objectAtIndex:0];
That works fine. Yet when I push my view managed by my view controller onto the navigation controller stack, the count method in the numberOfRowsInSection method crashes the app, see below.
- (NSInteger)tableView:(UITableView *)tableView numberOfRowsInSection:(NSInteger)section {
// Return the number of rows in the section.
return [curPersonPrescriptions count];
}
What could be causing this? How can a valid custom array, with no objects, not return a valid count? Where am I going wrong? Thanks.
Subclass of NSArray? You're aware that NSArray is a class cluster, and is therefore somewhat difficult to subclass, right? In fact, it's so fraught with danger that the NSArray documentation has a whole section dedicated to what you need to do in order to subclass it.
I'll bet that that's the source of your woes.
You almost certainly don't need to subclass NSMutableArray in this situation. Instead, make a new class which has an array as a property, along with the extra properties you desire.
When you subclass NSMutableArray, you need to implement some mandatory methods like count, addObject:, insertObjectAtIndex etc. This is what we call as class cluster.
If you want to add some more feature/behavior to already implemented object then you can write a "category" instead of "subclassing" it.
If you want to subclass it, then you have to implement all those methods which your are going to use so better write a category of NSMutableArray and extend the feature what you want and use the NSMutableArray object only. This will solve your problem and also this is the easy and almost right way to add new behavior to already existing class.
Is there any way to access the owning UITableView from within a UITableViewCell?
Store a weak reference to the tableView in the cell, which you'd set in -tableView:cellForRowAtIndexPath: of your table's dataSource.
This is better than relying on self.superview to always be exactly the tableView is fragile. Who knows how Apple might re-organize the view hierarchy of UITableView in the future.
Here's a nicer way to do it, which does not rely on any particular UITableView hierarchy. It will work with any future iOS version, provided that UITableView does not change classname altogether. Not only this is extremely unlikely, but if it does happen you will have to retouch your code anyway.
Just import the category below and get your reference with [myCell parentTableView]
#implementation UIView (FindUITableView)
-(UITableView *) parentTableView {
// iterate up the view hierarchy to find the table containing this cell/view
UIView *aView = self.superview;
while(aView != nil) {
if([aView isKindOfClass:[UITableView class]]) {
return (UITableView *)aView;
}
aView = aView.superview;
}
return nil; // this view is not within a tableView
}
#end
// To use it, just import the category and invoke it like so:
UITableView *myTable = [myTableCell parentTableView];
// It can also be used from any subview within a cell, from example
// if you have a UILabel within your cell, you can also do:
UITableView *myTable = [myCellLabel parentTableView];
// NOTE:
// If you invoke this on a cell that is not part of a UITableView yet
// (i.e., on a cell that you just created with [[MyCell alloc] init]),
// then you will obviously get nil in return. You need to invoke this on cells/subviews
// that are already part of a UITableView.
UPDATE
There is some discussion in the comments about whether keeping a weak reference is a better approach. It depends on your circumstances. Traversing the view hierarchy has some small runtime penalty as you are looping until the target UIView is identified. How deep are your views? On the other hand, keeping a reference on every cell has a minimal memory penalty (a weak reference is a pointer after all), and generally adding object relationships where they are not needed is considered a bad OO design practice for many reasons, and should be avoided (see details in the comments below).
More importantly, keeping table references inside cells adds code complexity and can lead to errors, because UITableViewCells are reusable. It is no coincidence that UIKit does not include a cell.parentTable property. If you define your own you must add code to manage it, and if you fail to do so effectively you can introduce memory leaks (i.e., cells live past the lifetime of their table).
Because typically you'll be using the category above when a user interacts with a cell (execute for a single cell), and not when laying-out the table in [tableView:cellForRowAtIndexPath:] (execute for all visible cells), the runtime cost should be insignificant.
Xcode 7 beta, Swift 2.0
This works fine for me, in my opinion it has nothing to do with the hierarchy or whatever. I had no trouble with this approach so far. I've used this for many async callbacks (ex. when an API request is done).
TableViewCell class
class ItemCell: UITableViewCell {
var updateCallback : ((updateList: Bool)-> Void)? //add this extra var
#IBAction func btnDelete_Click(sender: AnyObject) {
let localStorage = LocalStorage()
if let description = lblItemDescription.text
{
//I delete it here, but could be done at other class as well.
localStorage.DeleteItem(description)
}
updateCallback?(updateList : true)
}
}
Inside table view class that implements the DataSource and Delegate
func tableView(tableView: UITableView, cellForRowAtIndexPath indexPath: NSIndexPath) -> UITableViewCell {
let cell: ItemCell = self.ItemTableView.dequeueReusableCellWithIdentifier("ItemCell") as! ItemCell!
cell.updateCallback = UpdateCallback //add this extra line
cell.lblItemDescription?.text = self.SomeList[indexPath.row].Description
return cell
}
func UpdateCallback(updateTable : Bool) //add this extra method
{
licensePlatesList = localStorage.LoadNotificationPlates()
LicenseTableView.reloadData()
}
Ofcourse you can put any variable in the updateCallback and change it's function in the tableView accordingly.
Someone might want to tell me if it is save to use though, just to be sure.
You have to add a reference back to the UITableView when you construct the table view cell.
However, almost certainly what you really want is a reference to your UITableViewController... that requires the same thing, set it as a delegate of the cell when you build the cell and hand it to the table view.
An alternate approach if you are wiring up actions is to build the cells in IB, with the table view controller as the files owner - then wire up buttons in the cell to actions in the table view controller. When you load the cell xib with loadNibNamed, pass in the view controller as the owner and the button actions will be wired back to the table view controller.
If you have custom classes for your UITableViewCells, you can add an id type variable in your cell's header, and synthesize the variable. After you set the variable when you load the cell, you are free to do what you please with the tableview or any other higher view without much hassle or overhead.
cell.h
// interface
id root;
// propery
#property (nonatomic, retain) id root;
cell.m
#synthesize root;
tableviewcontroller.m
- (UITableViewCell *)tableView:(UITableView *)tableView cellForRowAtIndexPath:(NSIndexPath *)indexPath {
// blah blah, traditional cell declaration
// but before return cell;
cell.root = tableView;
}
Now you can call any of the tableview's methods from within your cell using the root variable. (e.g., [root reloadData]);
Ah, takes me back to the good old days of flash programming.
The two methods in other answers are: (A) store a reference to the table, or (B) walk up the superviews.
I'd always use something like (A) for model objects and (B) for table cells.
Cells
If you are dealing with a UITableViewCell, then AFAIK you must either have the UITableView at hand (say you are in a table delegate method), or are dealing with a visible cell that is in the view hierarchy. Otherwise, you may well be doing something wrong (please note the "may well").
Cells are liberally reused and if you happen to have one that is not visible then the only real reason that cell exists is because of iOS UITableView performance optimization (a slower iOS version would have released and hopefully dealloc'd the cell when it moved off screen) or because you have a specific reference to it.
I guess this is probably the reason that table cells are not endowed with a tableView instance method.
So (B) gives the right result for all iOS's so far, and all future ones until they radically change how views work.
Though in order to avoid writing generalizable code over and over, I'd use this:
+ (id)enclosingViewOfView:(UIView *)view withClass:(Class)returnKindOfClass {
while (view&&![view isKindOfClass:returnKindOfClass]) view=view.superview;
return(view);
}
and a convenience method:
+ (UITableView *)tableForCell:(UITableViewCell *)cell {
return([self enclosingViewOfView:cell.superview withClass:UITableView.class]);
}
(or categories if you like)
BTW, if you are concerned about the effect of a loop with 20 or so iterations of that size on your app performance,.. don't.
Models
If you are talking about the model object that is displayed in the cell, then definitely that model could/should know about its parent model, which may be used to find, or trigger changes in, the table(s) that the cell's model might be displayed in.
This is like (A), but less brittle with future iOS updates (eg one day they might make the UITableViewCell reuse cache exist per reuseidentifier, rather than per reuseidentifier per tableview, on that day all the implementations that use the weak reference method will break).
Th model method would be used for changes to the data displayed in the cell (i.e. model changes) since changes will propagate wherever the model is displayed (eg. some other UIViewController somewhere else in the app, logging, ...)
The cell method would be used for tableview actions, which would likely always be a bad idea if the cell isn't even a subview of a table (though it's your code, go nuts).
Either way, use a unit test rather than assuming that seemingly cleaner code just works when they update iOS.
UITableView *tv = (UITableView *) self.superview.superview;
UITableViewController *vc = (UITableViewController *) tv.dataSource;