StringConvert | converting string between a various charset | Data Manipulation library

Exceptions aren't just a funny name. You shouldn't use an exception just because the name sounds vaguely relevant. CharConversionException is about issues with charset encoding and e.g. XML reading; the point is, it extends IOException, so it is a kind of 'problem with an Input/Output channel', and it just doesn't apply here. Make your own exception types, or use one of the more general exception types that are intentionally designed (by reading the javadoc and checking the inheritance) to be a bit more nebulous. For example, IllegalArgumentException could be used here (if the argument doesn't start with a D or d, it is, apparently, illegal? Bit of a weird method).

The difference is in the kind of exception it is.

In java, throwables are checked, meaning, a method needs to catch all exceptions it throws, unless it explicitly decrees that it, itself, throws this. Thus, this does not compile:

TimestampConverter expects Unix epoch time, not ticks.

You'll need to convert it, which would have to be a custom transform or a modification in your Producer (which shouldn't be a major problem because most languages have datetime epoch functions)

Convert ticks to unix timestamp

First StringConverter offers only one method to overwrite object ConvertFromString(..).
The converstion to string is handled by nothing because it's suppose to be a string.

Here I supposse that your Type is DateTime and you got it in multiple Exotique format. If you have only one format you can change the default format for that type.

A simple demo class and it's mapping:

You're using the JSON Schema converter (io.confluent.connect.json.JsonSchemaConverter), not the JSON converter (org.apache.kafka.connect.json.JsonConverter).

The JSON Schema converter uses the Schema Registry to store the schema, and puts information about it on the front few bytes of the message. That's what's tripping up your code (Could not read JSON: Invalid UTF-32 character 0x17a2241 (above 0x0010ffff) at char #1, byte #7)).

So either use the JSON Schema deserialiser in your code (better), or switch to using the org.apache.kafka.connect.json.JsonConverter converter (less preferable; you throw away the schema then).

More details: https://rmoff.net/2020/07/03/why-json-isnt-the-same-as-json-schema-in-kafka-connect-converters-and-ksqldb-viewing-kafka-messages-bytes-as-hex/

There are two ways to computes the width and height of the specified string of text. You can use neither GetTextExtentPoint32 and Graphics::MeasureString to get the actual size depending on how you drew your text.

In your case, you are drawing the text using GDI+ but you are measuring the width and height of it using the classic GDI which will give you a different result because of scaling. Although you can still use GetTextExtentPoint32 to measure the text. However, you will need to handle the DPI to get the width and height the same as how it is drawn using GDI+. GDI+ is an improvement on GDI and there are differences between the two. For example, the scaling on how they draw a string.

The following code shows how to draw the string using Graphics::DrawString and compute the width and height using Graphics::MeasureString. Drawing and computation are done using GDI+.

Static class members are shared between all instances of a class. There's a static string text in your class so it is to be expected that all instances share that. If you need to store per-instance data you need to use non-static class members.

Presumably, you've used static class members so that you can put your window procedure inside the class' implementation (which needs to be static). To have a static window procedure access per-instance data has been asked and answered before (like here, here, here, or here).

There are several problems in your setup:

• You don't add the package for Kafka support as described in docs. It's either needs to be added when starting pyspark, or when initializing session, something like this (change 3.0.1 to version that is used in your jupyter container):

Answer to question could be found here! https://solace.community/discussion/646/solace-integration-with-kafka-over-tcps-failing

👍

Thanks all.

I was using an older Kafka version.

I upgraded Kafka in the cluster from kafka_2.12-1.1.0 to the latest stable version kafka_2.12-2.7.0. I also upgraded zookeeper from zookeeper-3.4.6 to apache-zookeeper-3.6.2-bin version.

In addition in the run file I added the following: