I'm trying to programmatically read Spring-style properties files, including YAML-formatted files.
What I have somewhat works, but it ends up flattening the YAML and so I end up with keys that are all mixed up and when I try to map it to a specific object, it gets confused.
Here is what I have (in Kotlin):
import my.project.Properties
import java.util.Properties as JavaProperties
fun loadConfig(configFiles: List<String>): Properties =
JavaProperties().apply {
putAll(appProperties.defaultConfig)
(if (configFiles.isEmpty()) appProperties.defaultConfigFiles else configFiles)
.forEach { PropertiesLoaderUtils.fillProperties(this, FileSystemResource(it)) }
}.let {
// throws exception because it was flattened
objectMapper.convertValue(it, Properties::class.java)
}
If I have a config file like this:
a:
b:
c: 1
I would want it to end up something equivalent to this in the Java Properties:
mapOf("a" to mapOf("b" to mapOf("c" to 1)))
Instead, I just get something like this:
mapOf("c" to 1)
It totally flattens it out and throws away the hierarchy entirely.
I'd prefer not to just use a YAML parser and do it myself, because I would like to allow the "normal" Spring style property files as well.
Note: These have to be loaded at run-time because the application is actually for running against something else, so I can't just let Spring handle it automatically like a normally would.
Related
I have an application.yml file with the following:
topics:
input:
- name: topic1
partitions: 3
replicas: 1
- name: topic2
partitions: 6
replicas: 2
I would like to be able to update or add a new topic object at runtime.
I have tried the following for updating an existing object:
java -jar app.jar --topics.input[0].name="topicX"
and the following for adding another object to the list:
java -jar app.jar --topics.input[2].name="topicX" --topics.input[2].partitions=6 --topics.input[2].replicas=2
I am accessing the properties in the following way:
#Component
#ConfigurationProperties(prefix = "topics")
#Validated
public class TopicsConfiguration {
#NotEmpty
#NotNull
private List<TopicConfiguration> input = new ArrayList<>();
public List<TopicConfiguration> getInputTopics() {
return input;
}
public void setFacts(List<TopicConfiguration> input) {
this.input = input;
}
}
Where TopicConfiguration is just a POJO with the 3 fields listed.
When I don't try and modify any of the property objects at runtime this works exactly as I expect, however I can not update the property list at all. When I try and update an existing object I get an NPE. When I try and add a new object to the list I get:
Property: topics.input[2].name
Value: lmnop
Origin: "topics.input[2].name" from property source "commandLineArgs"
Reason: The elements [topics.input[2].name,topics.input[2].partitions,topics.input[2].replicas] were left unbound.
I would just like to know if there is any way to update or add an element to the list at runtime so that users of my project don't have to modify application.yml if they want to update this configuration.
okey so i did some research and some testing and came up with the following:
you cannot update a defined list in application.yml
if you run:
java -jar myApp.jar -my-list[2].name=foo
it will fail, because as soon as you want to pass the list the list will override the current list in application.yml and you are trying to pass in the 3rd item in the list when there is nothing at index 0 and 1.
it is not a dynamic list, it as an array.
You need to pass the entire list:
java -jar myApp.jar -my-list[0].name=foo -my-list[1].name=bar -my-list[2].name=foobar
So if you are going to pass a list in from cmd, you must always define the list from scratch.
So you want to reload your application.properties or application.yml in your Spring Boot application on the fly without having to make another code commit to another deployment? If I understood that right then here is how you can do it.
Read the these links and also google more on Spring Boot Actuator Reload Properties or Spring Scheduling. We can't provide you an out-written code to get this working but there is plenty of examples to examine and try out.
Links:
Actuator Refresh
Spring Scheduling
If I were you, I would prefer Actuator method as it avoids confusion and is cleaner.
I want to create API, which takes part of its input from environment (urls etc..). Basically a Jar file.
I also wan't the values being auto updated when application.properties changes. When there is change, this is called:
org.springframework.cloud.endpoint.RefreshEndpoint#refresh
However, I consider it bad practice to have magic env variable keys like 'server.x.url' in the Api contract between client application and the jar. (Problem A)
That's why I'd like to use Api like this. But there's problem of old values.
public class MyC {
TheAPI theApi=null;
void MyC(){
theApi = new TheApi();
theApi.setUrl( env.get("server.x.url") );
}
doStuff() {
theApi.doStuff(); // fails as theApi has obsolete value of server.x.url, Problem B
}
So either I have ugly API contract or I get obsolete values in the API calls.
I'm sure there must be Spring way of solving this, but I cant get it to my head just now.
Overview.
My camel setup calls two service methods. the response of the first one is passed into the second one and then output the final response as json webpage. Fairly simple nothing too complicated.
Further breakdown to give some more context.
Method_1. Takes in scanId. This works ok. It produces an object called ScheduledScan .class
Method_2. Takes in object previous instance of ScheduledScan .class and returns a list of ConvertedScans scans. Then would like to display said list
Description of the code
#Override
public void configure() throws Exception {
restConfiguration().bindingMode(RestBindingMode.json);
rest("/publish")
.get("/scheduled-scan/{scanId}")
.to("bean:SentinelImportService?method=getScheduledScan").outType(ScheduledScan .class)
.to("bean:SentinelImportService?method=convertScheduledScan");
}
The methods that are called look like the following
ScheduledScan getScheduledScan(#Header("scanId") long scanId);
List<ConvertedScans > convertScheduledScan(#Body ScheduledScan scheduledScans);
It is returning the the following error
No body available of type: path. .ScheduledScan but has value:
of type: java.lang.String on: HttpMessage#0x63c2fd04. Caused by: No type converter available
The following runs without the error, i.e. without method 2. So I think im almost there.
rest("/publish")
.get("/scheduled-scan/{scanId}")
.to("bean:SentinelImportService?method=getScheduledScan");
Now from reading the error it looks like im passing in a HttpMessage not the java object? I'm a bit confused about what to do next? Any advice much appreciated.
I have found some similar questions to this message. However I am looking to pass the java object directly into the service method.
camel-rest-bean-chaining
how-to-share-an-object-between-methods-on-different-camel-routes
You should setup the outType as the last output, eg what the REST response is, that is a List/Array and not a single pojo. So use .outTypeList(ConvertedScans.class) instead.
I can't determine how to add custom properties or search for them.
Everything I have tried is giving me a Error - #<Google::Apis::ClientError: invalid: Invalid query> when I attempt to search for them. I can successfully complete other queries but I don't know if the client is setup to work with appProperties (or even properties at all).
Basically I just need the correct syntax for searching and adding since it doesn't appear to be in the documentation.
Assuming you already have a reference to an authorized DriveService, you can search based on appProperties using a q-parameter (documented here), like this:
file_list = drive.list_files(
q: "appProperties has { key='my_app_key' and value='my_val' }",
fields: 'files(id, name, appProperties)',
spaces: 'drive')
If you omit the fields parameter then the search will still work but the properties themselves won't be returned.
Updating appProperties is definitely arcane and the documentation is opaque. What you need is the ID of the file, and a File value object as a container for the attributes to update. Something like this:
new_app_properties = { 'my_app_key' => 'my_val' }
update_f = Google::Apis::DriveV3::File.new(app_properties: new_app_properties)
drive.update_file(file_id, update_f)
I'm currently using MvvmCross DownloadCache -- and it's working alright -- especially nice when I just need to drop in an Image URL and it automagically downloads / caches the image and serves up a UIImage.
I was hoping to leverage the code for one other use case -- which is I'd like to grab source images from URL's and cache the files on the local file system, but what I really want for this other use case is the image path on the local file system instead of the UIImage itself.
What would help me most if I could get an example of how I might accomplish that. Is it possible to make that happen in a PCL, or does it need to go into the platform specific code?
Thanks -- that works, but just in case anyone else is following along, I wanted to document how I got the Mvx.Resolve<IMvxFileDownloadCache>() to work. In my setup.cs (in the touch project), I had:
protected override void InitializeLastChance ()
{
Cirrious.MvvmCross.Plugins.DownloadCache.PluginLoader.Instance.EnsureLoaded();
Cirrious.MvvmCross.Plugins.File.PluginLoader.Instance.EnsureLoaded();
Cirrious.MvvmCross.Plugins.Json.PluginLoader.Instance.EnsureLoaded();
...
}
But that wasn't enough, because nothing actually registers IMvxFileDownloadCache inside the DownloadCache plugin (which I was expecting, but it's just not the case).
So then I tried adding this line here:
Mvx.LazyConstructAndRegisterSingleton<IMvxFileDownloadCache, MvxFileDownloadCache>();
But that failed because MvxFileDownloadCache constructor takes a few arguments. So I ended up with this:
protected override void InitializeLastChance ()
{
...
var configuration = MvxDownloadCacheConfiguration.Default;
var fileDownloadCache = new MvxFileDownloadCache(
configuration.CacheName,
configuration.CacheFolderPath,
configuration.MaxFiles,
configuration.MaxFileAge);
Mvx.RegisterSingleton<IMvxFileDownloadCache>(fileDownloadCache);
...
}
And the resolve works okay now.
Question:
I do wonder what happens if two MvxFileDownloadCache objects that are configured in exactly the same way will cause issues by stepping on each other. I could avoid that question by changing the cache name on the one I'm constructing by hand, but I do want it to be a single cache (the assets will be the same).
If you look at the source for the plugin, you'll find https://github.com/MvvmCross/MvvmCross/blob/3.2/Plugins/Cirrious/DownloadCache/Cirrious.MvvmCross.Plugins.DownloadCache/IMvxFileDownloadCache.cs - that will give you a local file path for a cached file:
public interface IMvxFileDownloadCache
{
void RequestLocalFilePath(string httpSource, Action<string> success, Action<Exception> error);
}
You can get hold of a service implementing this interface using Mvx.Resolve<IMvxFileDownloadCache>()
To then convert that into a system-wide file path, try NativePath in https://github.com/MvvmCross/MvvmCross/blob/3.2/Plugins/Cirrious/File/Cirrious.MvvmCross.Plugins.File/IMvxFileStore.cs#L27