I am using the SDWebImage component in my project. Recently, I also decided to add MWPhotoBrowser (which uses SDWebImage). After installing MWPhotoBrowser I can no longer compile without receiving numerous errors. I am certain this is because the SDWebImage version is different across components I use.
As soon as I add the location of the header files for MWPhotoBrowser (which include SDWebImage) to my user search path I start getting these errors.
Any guidance / help as to how I can sandbox the libraries that MWPhotoBrowser needs so it doesn't affect the rest of my app ?
Rather than modifying any of the browser code I decided to eliminate my copies of SdWebimage and leverage those included in MWPhotobrowser instead. Not ideal, but I think the proper solution is to modify MWPhotobrowser to use these libraries as sub modules.
Related
When you have a server side blazor app and you reference a razor class library directly (at the moment I am using dotnet5) it seems to automatically add the /_content/.bundle.scp.css by adding an #import to the .styles.css. This appears to work well enough at least in debug just by adding the reference and hitting F5.
I see conflicting documentation adding the links manually in many places and they are generally being added in the head as links, not as an #import. I am guessing this is maybe something that has changed since core3. I also have not figured out where the bundled files are supposed to "live" while you are still developing, when you publish you end up with a \wwwroot_content folder, but a development that appears not to exist on the filesystem and actually be mapped onto the \obj folders of the individual RCLs.
In the app I am currently working on the razor class libraries are loaded at run time, so this does not work automatically, for the time being I have the bundles referenced as individual links programmatically added in the head of the _host.cshtml after I copy them from the individual RCLs' obj<BuildFolder>\net5.0\ as that seems to be the only place they exist on disk. I am also serving them with an app.UseStaticFiles out of another folder since I am still unclear on how the inbuilt folders are meant to operate.
When I first figured out that the runtime load was an issue I was able to get confirmation there was not inbuilt support beyond direct references here https://github.com/dotnet/aspnetcore/issues/33284 and I started looking for how to re-implement the existing work-around more properly.
I would prefer this to work as similar to a directly referenced RCL as possible, but I am not finding any information on if the bundles are meant to be referenced globally at all times or conditionally, I do not see any obvious way to get them to be added as an #import programmatically, and I have no idea what the implications of trying to manually create the _content folder from the bundle files found in the various \obj folders would be.
How this is really supposed to work in development and when publishing? Is what i am doing even close or is there some glaring issue I will be encountering down the road?
I am using Xcode 4.6.2 and i believe i am facing circular import issue in my project which i am unable to troubleshoot. Due to which i am not able to access few methods from another class using the class method. See my earlier question here.
Although i am using #class instead of #import in my header files, still i am unable to fine where exactly the problem is. Members of the StackOverflow have suggested me to use Xcode's built in functionality found under Product > Generate Output > Preprocessed File.
I have used this functionality and it generated a file that is too long to follow and i don't know what to do with it.
I have tried searching how to use it to troubleshoot the issue but couldn't find much help. Can anyone help me point out how to use "Preprocessed File" to troubleshoot my issue. Thanks!
I also went through your earlier question you have mentioned inside the question. You have said that you are using #class instead of #import in your header files, the methods that you are trying to access are declared in the header files and there are no typos of any kind.
In such cases, usually no body points this issue but i am going to do it anyway because i have faced such issues many times. You have probably created many copies of your project to work on each functionality and also keeping a working project.
When you do this, sometimes Xcode is still using the older copies of few files. That means it is still using the older copy of the TheFeedStore.h when the methods you are trying to access were not declared by you.
How to solve this problem is very simple. Go to the file from which you are trying to access the methods and the files in which these methods are declared.
In the Utilities section on the right hand side, check the location and full path under "Identity and Type" area.
First check the names of the project, if it is different from the project name that you are working on, that means Xcode is still pulling the old copies of the files from the previous revision of your project. See the blue arrows where the project name is 13SampleMoreRequests in my case.
If this name is same as your project name, then my answer does not solve your problem. If its different, you should use the new copies of the file by browsing the new location using the sign that is pointed out by red arrow.
Once you browse and use the new files, your problem will be solved and you will be able to access the methods. If you still can't, copy these files, delete from the project and then add them again and you won't face this problem.
Hope this helps!
There is code that I want to include in most of my projects. Things like AFNetworking, categories for CoreData and unit testing, etc.
It seems logical to include all of these in a static library, and then use it in each project. I've noticed though, that many third-party libraries (like AFNetworking, and it's predecessor ASIHTTP) are included in projects by copying over all of their source files and then manually linking the necessary libraries to the project target.
This seems to me like the easiest way. It took a fair amount of time to figure out how to include an existing static library into a project. Even after I knew how, it still seems like a pain to do it for every new project. Also, the header search paths that you specify are to a local directory with the static library's files. Wouldn't it be easier, and is there a way, to copy the static library's files into the project? This is the same idea as including the class files directly like most libraries seem to do already, but it would be more organized because everything would be lumped into one library project, instead of having class files everywhere and having to include every one of them.
Static libraries feel like they should be the right way to go. Make a library that can be used with all projects that includes classes that every project will need. Makes sense. I am just conflicted because it seems like the right way to go is to leave everything out of a 'formal' library, and just copy over all of the class files instead.
I guess I am just looking for what experienced developers find to be the best option.
I would be among the first to admit that the process of referencing a static library in Xcode is not entirely intuitive. However, using a static library is the best option, without a doubt.
The main reason is maintainability: when you copy source code of a library to many places, you must remember to update all of them to the latest code when you upgrade to the next version of the library. This may be a rather error-prone process, especially when the underlying library source changes significantly (e.g. new files are added, old files are renamed, etc.)
There's a halfway solution - make an XCode project that builds your static library from source and put that into a shared repository (ie.. git submodule etc) which is included from each project's main repository.
Each of your projects would include this submodule and project. Then they get the latest source code each time they pull that submodule. If you set this up as a build dependency it will build a static library the first time you build and then XCode is smart enough just to include it each subsequent build so you get the benefit of fast build times.
You also get the advantage of having the source right there for stepping though / debugging.
If it's in a separate XCode project and a new version of a library adds or removes a source file you would only need to change that shared project - all your individual projects wouldn't change at all.
What about using CocoaPods? This tool does exactly what you want in a declarative way: you have a file (Podfile) where you declare your dependencies, and the tool downloads all the dependencies and builds a static library that gets added to your project.
I would agree that static libraries feel like they might be the correct way to go for a number of reasons, but can also introduce some issues.
The positives would be creating an easy way to add a library to a project. Although not completely intuitive, it is rather trivial to add a static library to a project after one does it a few times. Add the files, add the search path, done. This could also be useful in certain source control situations. Also, updating a library may be easier.
I think the real problem here is for the open source community. By including, say AFNetworking, for example, as a static library, you lose all access to the implementation files. This is a great feature of including source rather than a library. It lets you change code to how you see fit, and hopefully give back.
We're planning to launch a serie of applications in AppStore. They will be for some kind of different journals, showing different contents downloaded from a server via XML. So these applications will be made from exactly the same code (It's an universal application, so It'll work both in iPhone/iPad).
My initial idea was, in order to upload the application, compile just changing the images, logos and configurations (plist) that makes the application react as a particular journal. The compressed file would be uploaded to the AppStore.
However, this has resulted a horrible method, which promotes failures and mistakes. If I forget to change some image, as you can't see them in the compiled file (as it is included) they will end up in the store (and I will need four or five days in order to get the application changed).
I'm trying to look up for a better approach, wich keep the projects as independent as possible. I would like to be able to share the entire codebase: views, classes and nibs and create different projects for every journal.
Which is the best method to achieve that?. What structure would let me group both logic (controllers, classes) and UI and use it in the different projects?.
I hope I've explained.
As always, thank you very much.
You should keep most of your common code as a library project. Each final project should link with this project and provide images/assets along with code to mention these assets to common code. In my day job, I write a common library too, which gets used by 2 products/apps at my employer.
An Xcode project can have multiple Targets, all the Targets sharing code, but each Target getting its own resources (icons, images, text, plists, etc.) from a different subdirectory/folder within the same project directory/folder. Then you can check the whole thing, or just the shared source, into your source control repository.
You should also be testing each of your apps, built exactly the same way as any submission except for the codesigning, on a device before uploading to the store.
You can have a single Xcode project that creates multiple applications. You'll need to create a separate Info.plist with a different bundle identifier for each app.
If you are using a git repository you can just branch for each different app you want and that would keep track of all the differences and if you need to switch which you are working on you just have to checkout that branch. This would allow for the exact same structure just minor differences between the actual code for each.
My application has a plugin system that allows my users to write their own plugins that get loaded at runtime. Usually this is fine but in some cases two plugins use the same libraries that will cause a collision between those two.
Example:
Plugin A wants to use TouchJSON for working with JSON and thus the creator adds the TouchJSON code to the plugin source and it gets compiled and linked into the plugin binary. Later Plugin B also wants to use that same library and does exactly the same. Now when my app loads these two different plugins it detects this and spits out an warning like this:
Class CJSONScanner is implemented in
both [path_to_plugin_a] and
[path_to_plugin_b]. One of the two
will be used. Which one is undefined.
Since my app just loads plugins and makes sure they conform to a certain protocol I have no control over which plugins are loaded and if two or more use the same library.
As long as both plugins use the exact same version of the library this will probably work but as soon as the API changes in one plugin a bunch of problems will arise.
Is there anything I can do about this?
The bundle loading system provides no mean to pacifically resolve name conflicts. In fact, we're told to ensure ourselves that the problem doesn't happen, rather than what to do if it happens. (Obviously, in your case, that's not possible).
You could file a bug report with this issue.
If this is absolutely critical to your application, you may want to have bundles live in separate processes and use some kind of IPC, possibly NSDistantObject, to pass the data from your program to the plugin hosts. However, I'm fairly sure this is a bag of hurt, so if you don't have very clearly-defined interfaces that allow for distribution into different processes, it might be quite an undertaking.
In a single-process model, the only way to deal with this is to ensure that the shared code (more precisely, the shared Objective-C classes) is loaded once. There are two ways to do this:
Put the shared code in a framework.
Put the shared code in a loadable bundle, and load the bundle when the plug-in is loaded if the relevant classes aren’t already available (check using NSClassFromString()). The client code would also have to use NSClassFromString() rather than referring to classes directly.
Of course, if you aren’t in control of the plug-ins you can’t enforce either of these schemes. The best you can do is provide appropriate guidelines and possibly infrastructure; for instance, in the second case the loading could be handled by the application, perhaps by specifying a class to check for and the name of an embedded bundle to load if it isn’t available in the plug-in’s Info.plist.