unpack | Create Web apps without a bundler | Build Tool library

Your problem isn't with how you call gather. It's with how you define main. The clue is with the error message

In Java 11, you can find out the flags that have been passed to the JVM and the "ergonomic" ones that have been set by the JVM by adding -XX:+PrintCommandLineFlags to the JVM options.

That should tell you if the container you are using is overriding the flags you have given.

Having said that, its is (IMO) unlikely that the container is what is overriding the parameters.

It is not unusual for a JVM to use more memory that the -Xmx option says. The explanation is that that option only controls the size of the Java heap. A JVM consumes a lot of memory that is not part of the Java heap; e.g. the executable and native libraries, the native heap, metaspace, off-heap memory allocations, stack frames, mapped files, and so on. Depending on your application, this could easily exceed 300MB.

Secondly, OOMEs are not necessarily caused by running out of heap space. Check what the "reason" string says.

Finally, this could be a difference in your app's memory utilization in a containerized environment versus when you run it locally.

Looking at the three instances of apiString and rewriting them to use the more succinct F-strings, they all appear to be of the form:

If you want to loop over df.columns two values at a time, you can do so like this:

The cryptic error message is of little help, but the stack trace shows that the error occurs in the line:

1. you should apply sum to each column separately:

search_columns = [db.func.sum(getattr(Model, i)) for i in col_headers]

1. you can use Query.add_columns to add columns to the query results:

query = query.add_columns(*search_columns)

or you can pass them directly:

Model.query.with_entities(Model.m_t, *search_columns)

Putting it all together:

The move semantics creates new objects. It is not a copy or assignment, it is a move-constructor that is invoked. Thus, if you call move(innerVector) you actually create a new vector object as entry in containerOfVectors. This automatically contains the content/data of the original innerVector at the moment you make the call. Afterwards, the innerVector is empty, and whatever you do with it will not affect the new object inside containerOfVectors. There is no link between those two distinct objects. The memory address of an object created with move-construction is certainly always different from its original object.

This fully explains the output of your example. With any other data type inside the vector, also structs, this will be just the same.

The problem is data[3:6] is only 3 bytes long, but by default the format "BH" is requires four bytes because a default alignment with padding is assumed. You can prove this with a:

After seeing whole log line entry