IDE sample of "unsupported sources"->DataFrame #1231
Conversation
Jolanrensen
added
examples
Jolanrensen
force-pushed
the
unsupported-data-sources-examples
branch
from
2b58d6c to
738cd46
Compare
Jolanrensen
force-pushed
the
unsupported-data-sources-examples
branch
from
738cd46 to
0e16817
Compare
Jolanrensen
force-pushed
the
unsupported-data-sources-examples
branch
from
80a82ae to
46128e8
Compare
Jolanrensen
force-pushed
the
unsupported-data-sources-examples
branch
from
408b0a5 to
05fd49e
Compare
…e examples clearer and more reproducible
# Conflicts: # settings.gradle.kts
| * This can be useful for storing matrices for easier access later or to simply organize data read from other files. | ||
| * For example, MRI data is often stored as 3D arrays and sometimes even 4D arrays. | ||
| */ | ||
| fun main() { |
There was a problem hiding this comment.
Seems like we could claim it as a NumPy array source reader via Multik
There was a problem hiding this comment.
Within limits, yes. Multik cannot read Fortran-contiguous numpy arrays for instance. Something I came across when looking for MRI data.
| // created by Customers.toDataFrameSchema() | ||
| // The same can be done for the other tables | ||
| @DataSchema | ||
| data class CustomersDf( |
There was a problem hiding this comment.
Maybe CustomersDFSchema?
There was a problem hiding this comment.
that would be a bit verbose wouldn't it? ;P "Simply cast it to the CustomersDfSchema DataSchema and you're good to go"
There was a problem hiding this comment.
Sometimes I want to name schema interface/data class as "...Schema" too. Let's discuss it and make some conventions 😄!
There was a problem hiding this comment.
Noo please don't. If you have a DataFrame<Person> it's clear you use Person as the schema for the dataframe, right? Adding Schema everywhere would just make it more verbose IMO. Remember that each data schema is already annotated with @DataSchema, plus the compiler plugin makes them all extend DataRowSchema as well. So you would get a "schema"-overload:
@DataSchema
data class MySchema(val a: Int): DataRowSchema
val df: DataFrame<MySchema>not to forget auto-generated data schemas based on JDBC/openAPI etc. using their own names. Would those break the convention?
...ported-data-sources/src/main/kotlin/org/jetbrains/kotlinx/dataframe/examples/exposed/main.kt
Show resolved
Hide resolved
| * @see toDataFrameSchemaWithNameNormalizer | ||
| */ | ||
| @Suppress("UNCHECKED_CAST") | ||
| fun Table.toDataFrameSchema(columnNameToAccessor: MutableMap<String, String> = mutableMapOf()): DataFrameSchema { |
There was a problem hiding this comment.
Very good proposal for KDF-Exposed integration module
Fixes #1215
Adds examples/guides in idea projects for using DataFrame +:
Updates README and documentation: