Join docs #1437
Join docs #1437
Conversation
|
While this looks much better already, I wonder if there's something we can do with colors, like we have for joinWith: https://kotlin.github.io/dataframe/joinwith.html I think these could help a lot with the visualization of which values end up where. For instance, imagine a yellow dataframe and a light blue one being joined, and the resulting dataframe consisting of the yellow columns, the blue ones and green for the join column(s). We could probably even use the |
|
|
||
| <!---IMPORT org.jetbrains.kotlinx.dataframe.samples.api.multiple.JoinSamples--> | ||
|
|
||
| Joins two [`DataFrame`](DataFrame.md) object by join columns. |
There was a problem hiding this comment.
ah you just copied this part from the original file
| } | ||
|
|
||
| private fun nameAndCityToColor(name: String, city: String?): RgbColor = when(name to city) { | ||
| "Alice" to "London" ->RgbColor(242, 210, 189) |
|
|
||
| <inline-frame src="./resources/notebook_test_join_6.html" width="100%" height="500px"></inline-frame> | ||
|
|
||
| If mapped columns have the same name, just select join columns from the left [`DataFrame`](DataFrame.md): |
There was a problem hiding this comment.
Hmm it's hard to see where the previous example ends and the new one begins. Maybe you could give them a small title?
There was a problem hiding this comment.
you do exactly that below :) nice
|
The colors help a lot! Thanks :) |
|
|
||
| ### Join with implicit keys (all columns with the same name) {collapsible="true"} | ||
|
|
||
| If `joinColumns` is not specified, columns with the same name from both [`DataFrame`](DataFrame.md) |
There was a problem hiding this comment.
You say joinColumns is not specified, yet in the example, you show { name and city }
| ```kotlin | ||
| // LEFT JOIN: | ||
| // Keep ALL left rows. If (name, city) matches, attach right columns; | ||
| // if not, right columns are null (e.g., Alice–London has no right match). |
There was a problem hiding this comment.
*columns from the right dataframe
Also, Alice-London does have a match, Bob-Dubai does not, so isBusy == null
| ```kotlin | ||
| // INNER JOIN: | ||
| // Keep only rows where (name, city) match on both sides. | ||
| // In this dataset both Charlies match twice (Moscow, Milan) → 2 merged rows. |
There was a problem hiding this comment.
what do you mean "both charlies"?, the result shows Alice and Charlie
| ```kotlin | ||
| // FULL JOIN: | ||
| // Keep ALL rows from both sides. Where there's no match on (name, city), | ||
| // the other side is filled with nulls. |
There was a problem hiding this comment.
could you make all nulls bold? so it's clearer these are new, added to fill in the gaps
There was a problem hiding this comment.
.format().with { if (it == null) bold else null } will do
I still find the examples hard to follow, I'm wondering how we can make it as clear as possible :)
Closes #583