Fix for #573: Change serialization format for rendering in IntelliJ IDEA #574
Conversation
core/generated-sources/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/json.kt
Outdated
Show resolved
Hide resolved
core/generated-sources/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/json.kt
Outdated
Show resolved
Hide resolved
|
Can i ask you to move internal and private functions from this file to new writeJson.kt / readJson.kt files under impl.io package? This file became too big to navigate :c |
core/generated-sources/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/json.kt
Outdated
Show resolved
Hide resolved
|
You changed handling of value and array columns. Can you tell me more about changes? This part of serialization was not tested by unit tests thoroughly before. I worry if things might break |
Jolanrensen
changed the title
core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/jupyter/JupyterHtmlRenderer.kt
Outdated
Show resolved
Hide resolved
ermolenkodev
force-pushed
the
KTNB-272
branch
from
aa038c8
to
f12853c
Compare
…lugin * Added the necessary metadata to the serialization format of the dataframe for it to be rendered in the Kotlin notebooks plugin. * Introduced the method `toJsonWithMetadata` which includes the metadata during the serialization process of the dataframe.
…entation and tests.
This commit includes a new document that explains the JSON serialization format used for rendering dataframes in the Kotlin notebooks plugin for IntelliJ IDEA. The format does not adhere to any formal schema such as Json Schema, but is intended for illustrative purposes.
This update introduces specific rendering flows based on the version of the IDE. A new method has been added to retrieve the IDE build number. This information is then used to conditionally generate a JSON-encoded DataFrame, adapting to the specific capabilities of the IDE version in use. Additionally, relevant tests have been updated to accommodate these changes.
Changed the condition check in the fromString function. Now, it accepts build numbers with three or more parts.
* Renamed encodePrimitiveData to encodeValue * Changed return type of toJsonWithMetadata to not expose JsonObject in publec API * Moved internal and private functions from json.kt to new writeJson.kt / readJson.kt files under impl.io package
done |
I have extracted the code that filters autogenerated values and array columns into functions to be reused in new methods. I have double-checked the logic, and it seems it hasn't been altered. However, I can't guarantee this without running tests. If this is critical, we should add tests and merge them before proceeding with this pull request. |
core/generated-sources/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/io/writeJson.kt
Outdated
Show resolved
Hide resolved
core/generated-sources/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/io/writeJson.kt
Outdated
Show resolved
Hide resolved
Replaced the usage of 'rows().take().toDataFrame()' with the more concise 'take()' method in multiple files.
Fixes #573