Andrus Adamchik
Andrus Adamchik
A message "Source contains property not mapped in the target" was intended to warn when the extractors have unexpected attributes, but more often than not these attributes are added intentionally...
From what I can tell extractor model attribute "type" is only used by the CsvExtractor, so it is entirely optional most of the time. Yet it is required in the...
In many cases that I've seen multiple interdependent `LmTasks` are run sequentially, wrapped in a single job. Users can parallelize these sequences via Java 8 [`CompletableFuture` ](https://docs.oracle.com/javase/8/docs/api/java/util/concurrent/CompletableFuture.html), still it takes...
"Unflattenning" extractor data matrix into a normalized schema has been a pain so far. This task aims at solving a very common problem in this area. The purpose here is...
Per #59 we allowed LinkMove to include extra source columns that are not directly mapped to anything in the target. We broke our convention for target naming though. Now transformation...
Let's allow `-H` to take an optional argument that will be treated as a config path, so the command would limit printed config to just the children of that path
The current approach to inflating a factory with configuration is rather verbose: ```java @Singleton @Provides MyType provide(ConfigurationFactory cf, MyOtherType ot) { return cf.config(MyTypeFactory.class, "prefix").create(ot); } // or, after #345 @Singleton...
Auto-expand arrays and Iterables when creating DataFrames from in memory objects (currently both cases require explicit "extractors") : ```java DataFrame .byRow() .appender(2) .append(List.of(1,2)) .append(List.of(3,4)) ``` ```java DataFrame .byArrayRow() .appender(2) .append(1,...
`TableLoader.limit()` must result in generation of the `SQL LIMIT` clause. Applying limit in memory results in bad performance (at least in MySQL), as the entire `ResultSet` is retrieved in memory...
1BRC brought to light significant optmizations that can be done to `CsvLoader`. Large files [can be "memory-mapped" and loaded by segments in parallel](https://questdb.io/blog/billion-row-challenge-step-by-step/)