Using CocoaPods with Parse is giving me error ”duplicate symbol _PFConfigParametersRESTKey"? - duplicates

Currently, I am in the middle of migrating my iOS app from api.parse.com to my own server. In the guide I am following, I am at the point where I need to test the app's functionality with a local Parse Server. Although, setting up a custom Parse Server requires having the latest Parse-SDK, and I am running an older version. I am trying to update my frameworks via CocoaPods. My Podfile is as follows:
# Uncomment this line to define a global platform for your project
# platform :ios, '9.0'
target 'MYAPP' do
# Uncomment this line if you're using Swift or would like to use dynamic frameworks
# use_frameworks!
# Pods for MYAPP
pod 'ParseFacebookUtilsV4'
pod 'Parse'
#pod 'ParseTwitterUtils'
pod 'ParseCrashReporting'
pod 'ParseUI'
target 'MYAPPTests' do
inherit! :search_paths
# Pods for testing
end
end
When I try running the app, I get the following error:
duplicate symbol _PFConfigParametersRESTKey in:
/Users/ME/Library/Developer/Xcode/DerivedData/MYAPP/Build/Products/Debug-iphoneos/Parse/libParse.a(PFConfig.o)
/Users/ME/Library/Developer/Xcode/DerivedData/MYAPP/Build/Products/Debug-iphoneos/Parse/libParse.a(PFConfigController.o)
duplicate symbol _PFConfigParametersRESTKey in:
/Users/ME/Library/Developer/Xcode/DerivedData/MYAPP/Build/Products/Debug-iphoneos/Parse/libParse.a(PFConfig.o)
/Users/ME/Library/Developer/Xcode/DerivedData/MYAPP/Build/Products/Debug-iphoneos/Parse/libParse.a(PFCurrentConfigController.o)
ld: 2 duplicate symbols for architecture arm64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
I've been searching for something online to help, but nobody seems to be experiencing this problem. I think my case is unique, as I have yet to see someone who has two duplicate errors on the same symbol pointing to the same archive(libParse.a) with one shared file (PFConfig.o) and two differing ones (PFConfigController.o and PFCurrentConfigController.o). I've implemented a variety of solutions that would generally solve this "duplicate symbol" error, but I haven't had any success.
Things I have done:
Ensured that all manually added versions of these frameworks have been removed from the project.
Scanned the project directory up and down multiple times via Finder/Command Line/grep/find and could not find any duplicated frameworks.
Ensured I did not add any "import *.m" files accidentally.
Checked for red files/duplicates in the Frameworks folder as well as the "Link Library With Libraries" section of "Build Phases."
Checked my framework, header, and other linker paths and they seem to be alright. My "Other Linker Flags" contains "$(inherited)" and a -force_load call to a third party ".a" file for analytics.
Cleared ~/Library/Developer/Xcode/DerivedData as well as removed Pods/ and ran "pod install" multiple times.
I went on to investigate the problem in the Parse files. The only place where PFConfigParametersRESTKey is defined is here and here. This seems alright since one of them is preceded by the extern keyword(reference here). I tried messing with the source files a little bit by making this variable not static and also trying to rename one of them. Nothing worked. I cannot figure out where to look to fix this.. If anybody can shed some light here I would greatly appreciate it! Thank you.

The solution to my problem was to remove an -ObjC flag from the linker config. For some reason, -ObjC was not one of the entries in the Linking section of Build Settings. The way I found it was by going to Pods/Pods-MYAPP.debug.xcconfig, and manually removing the -ObjC flag from the OTHER_LDFLAGS variable.

Related

Docstrings are not generated on Read the Docs with Sphinx autodoc and napoleon extensions

I am using the Sphinx autodoc and napoleon extensions to generate the documentation for my project (Qtools). This works well on my local machines. I am using Sphinx 3.1.2 (or higher). However, when I build the documentation on Read the Docs (RTD), only text added directly to the reStructuredText files that form the source of the documentation is processed. The docstrings that are supposed to be pulled in by autodoc do not appear in the HTML documentation generated by RTD. So for example in docs\source\section2_rsdoc.rst I have:
Response spectra
================
The response spectrum class
---------------------------
.. autoclass:: qtools.ResponseSpectrum
:members:
Response spectrum creation
--------------------------
.. autofunction:: qtools.calcrs
.. autofunction:: qtools.calcrs_cmp
.. autofunction:: qtools.loadrs
See also :func:`qtools.convert2rs` (converts a power spectrum into a response spectrum).
This results in:
Response spectra
The response spectrum class
Response spectrum creation
See also qtools.convert2rs (converts a power spectrum into a response spectrum).
In other words, all directives are apparently ignored, and hyperlinks to other functions are not added. I have examined several basic guidance documents such as this one, but I cannot figure out what I am doing wrong. RTD builds the documentation without any errors or warnings. In RTD advanced settings I have:
Documentation type: Sphinx HTML
Requirements file: requirements.txt
Python interpreter: CPython 3.x
Install Project: no
Use system packages: no
Python configuration file: blank
Enable PDF build: no
Enable EPUB build: no
I haven't touched any other settings.
In conf.py I have tried the following variations of line 15: sys.path.insert(0, os.path.abspath('.')), sys.path.insert(0, os.path.abspath('../..')) and the current sys.path.insert(0, os.path.abspath('../../..')). None of those made any difference.
I would be grateful for any help!
RTD builds the documentation without any errors or warnings
This is slightly incorrect. As you can see in the build logs, autodoc is emitting numerous warnings like this one:
WARNING: autodoc: failed to import class 'ResponseSpectrum' from module 'qtools'; the following exception was raised:
No module named 'qtools'
This has happened for all your variations of sys.path.insert, as you can see in some past builds.
Trying to make it work this way is tricky, since Read the Docs does some magic to guess the directory where your documentation is located, and also the working directory changes between commands.
Instead, there are two options:
Locate where the conf.py is located (see How do you properly determine the current script directory?) and work out a relative package from there.
Invest some time into making your code installable using up-to-date Python packaging standards, for example putting all your sources inside a qtools directory, and creating an appropriate pyproject.toml file using flit.

gcloudignore doesn't allow standard wildcard whitelist

Gcloudignore works like gitignore in that you can exclude certain files from uploading to GCF. Sometimes when you have really large projects with lots of generated files, it can be useful to exclude all files except a few.
.gcloudignore
# Ignore everything
# Or /*
*
# Except the Cloud Function files we want to deploy
!/package.json
!/index.js
The following gcloudignore file gives us: File index.js or function.js that is expected to define function doesn't exist in the root directory. meaning index.js is ignored and can not be read.
However the following ignore file syntax works just fine to deploy:
# Ignore everything
/[!.]*
/.?*
# Except the Cloud Function files we want to deploy
!/package.json
!/index.js
I tried peering into the gcloud program code, but I was wondering if anyone knows why this is the case?
I ran into this problem as well.
I found a workaround -- explicitly allow the current dir:
# Ignore everything by default
*
# Allow files explicitly
!foo.bar
!*.baz
# Explicitly allow current dir. `gcloud deploy` fails without it.
!.
Works for me. Don't know why.
OperationError: code=13, message=Error setting up the execution environment for your function. Please try deploying again after a few minutes.
I just got this problem in recent. The error code 13 means, in my view point, the code/function failed to initialize (with in time).
In my case, the problem is mostly caused by either Provided module can't be loaded. (Dependencies failed) or Provided code is not a loadable module. (Main module is missing) after checking Stackdriver Logs, which point to the deploy isn't complete.
Below is how I got it to work:
# Ignore everything by default
/*
# Allow files explicitly
!config.js
!index.js
# Tried !app/ but does not work
!app/**
# Explicitly allow current folder, as Arne Jørgensen said.
!.
Be care the !app/ isn't enough. I have to use !app/** to including all files.
If it's your case, check the source tab from function console to see which files uploaded in actual.

Monodevelop fails to build, can't find Microsoft.DiaSymReader.Native.x86.dll'

I am trying to build a web api project using monodevelop on a mac. The thing is that after a few hiccups (explained in a question that turned out to be so messy I have just deleted) I get to the point of getting this error
/Users/myuser/git/LiveData/LiveData/CSC: Error CS0041: Unexpected error writing debug information -- 'Windows PDB writer is not available -- could not find Microsoft.DiaSymReader.Native.x86.dll' (CS0041) (LiveData)
In a windows machine the same project builds using visual studio targeting mono 4.5.
When I click on the error it tells me that /Users/myuser/git/LiveData/LiveData/CSC doesn't exist
Another thing is that in the folder structure of the solution there's a package folder (not the one inside the project) and inside this one I have a folder called Microsoft.Net.Compilers 1.3.2 that has inside another folder called "tools" that contains among other things csc.exe and the dll thta can't be found.
I have tried to install the dll directly in the project using nuget but even if it was installed the build showed me the same error
Thanks,
As for workaround for now you can just limit usage of Microsoft.Net.Compilers to Release configuration (edit *.csproj file):
<Import Project="..\packages\Microsoft.Net.Compilers.1.3.2\build\Microsoft.Net.Compilers.props" Condition="'$(Configuration)' == 'Release' And Exists('..\packages\Microsoft.Net.Compilers.1.3.2\build\Microsoft.Net.Compilers.props')" />
take a look at beginning of condition:
'$(Configuration)' == 'Release'
This way I can build and debug locally and build my project ie. in appharbor.
Building the project in release configuration should fix it!
This might not be completely related but may be helpful in some ways. Regarding the issue on 'could not find Microsoft.DiaSymReader.Native.x86.dll', have a look at this issue on GitHub: https://github.com/dotnet/cli/issues/3016
It seems like the solution is either:
Dependency to Microsoft.NETCore.Platforms needed for RID graph which
was missing. Any package which has transitive dependency on it (like
NETStandard.Library) could also make things work.
Adding dependency to "Microsoft.NETCore.Platforms": "1.0.1-" or
"NETStandard.Library":"1.5.0-" make it work.
adding Microsoft.NETCore.Platforms works as well

Trying to compile libgdx to html, but it cant clear the assets

Having trouble compiling my libgdx project to html.
Looking at the log I think the problem is;
Copying resources from ../android/assets to war/
C:\TomsProjects\MeshExplorerV2\html\..\android\assets
C:\TomsProjects\MeshExplorerV2\html\assets
[ERROR] Generator 'com.badlogic.gdx.backends.gwt.preloader.PreloaderBundleGenerator' threw an exception while rebinding 'com.badlogic.gdx.backends.gwt.preloader.PreloaderBundle'
java.lang.RuntimeException: Couldn't clean target path 'war\assets'
Now it strikes me as slightly odd the second line; As there was no html\assets folder.
The assets folder was in html\war\assets, like the last line complains about.
However, after creating a empty assets folder in html, and manually deleting the contents of html\war\assets I still get the same error.
Theres also a "html\webapp" directory without any assets folder. Not sure if there should be one.
I have tried cleaning/rebuilding from eclipse, as well as randomly also trying Gradle>>Refresh All.
Additionally, the Java version runs just fine.
Any pointers, thanks :)
While not exactly reveling the route of the problem, I noticed it let me compile again as long as I restarted eclipse inbetween.
This infers some file-lock going on, possibly a server still running, allthough nothing but the gwt compile as triggered by me.
Still, restarting is a fine workaround for me as it only takes a minute or two.
ah
Pinned it down: Close your FTP program when compiling - at least CoreFTP was locking the directory. Closing it works every time now.
I had a similar issue where I was getting the following error
[ERROR] Generator 'com.badlogic.gdx.backends.gwt.preloader.PreloaderBundleGenerator' threw an exception while rebinding 'com.badlogic.gdx.backends.gwt.preloader.PreloaderBundle'
java.lang.RuntimeException: Couldn't clean target path 'war/assets'
All my files under the html/war/assets directory were read only, re-starting the IDE didn't help. To fix it I needed to change permissions chmod -R 777 html/war/assets (you probably dont need 777).

Msbuild running in Jenkins target calling HgPull fails with HgProcessException: The command <hg.exe> is not available

I am porting over an MSBuild script from CCnet to run in Jenkins. The MSBuild project is used to create a deployment package. I would rather have Jenkins drive this process itself but that's a longer term aim.
The problem i am having is (as in the title) when we try and use the HgPull target, from the MSBuild mecurial task (http://msbuildhg.codeplex.com/) we get the error message
HgProcessException: The command hg.exe is not available [Path to project]
I have seen on the project web site that someone solved it by adding the LibraryLocation proeprty to the target but that seems to have made no difference. My target currently looks like this:
Target Name="UpdateSources">
<HgPull
LibraryLocation="C:\Program Files\TortoiseHg\hg.exe"
Force="true"
Update="true"
LocalPath="$(SourcePath)"
/>
<HgUpdate
LibraryLocation="C:\Program Files\TortoiseHg\hg.exe"
Clean="true"
LocalPath="$(SourcePath)"
/>
I'm rather at a loss. Please let me know if you need any more information added to this post to solve this issue. I'm really quite new to MSBuild so really not sure where to start investigating this.
EDIT:
One thing i forgot to mention was that i have tried running the MSbuild command in a console window on the build server and still get the same result. This is really odd given it works fine in CCNet, what magic is CCNet doing to make this command work?
This is now resolved, unfortunately i'm not sure what changes i made to correct these. I believe it may have been down to path separators and whether they where trailing or not in another part of the config file. It does so annoy me with the lack of resilience/consistency between applications where you need to specify paths with/without trailing slashes.
Just a thought, but try adding 'C:\Program Files\TortoiseHg' to your system path. Maybe CCNet has it specified somewhere that Jenkins doesn't have access to.
Also, just for sanity's sake, verify that hg.exe actually exists at that location.