Bug Fixes
Issue Ref | Summary | Description
-|-|-
[9](https://github.com/deeplime-io/onecode/issues/9) | Correct `gatherUsageStats` env variable | `onecode-start` used wrong variable name to prevent sending stats.
[No ref] | `kind` is now a property of the `InputElement` and `OutputElement` | `kind` would have potentially been overwritten while extracting element through `onecode-extract`.
Enhancements
Issue Ref | Summary | Description
-|-|-
[5](https://github.com/deeplime-io/onecode/issues/5) | Improving DeepLearning example | Allow initialization of `NeuralNetInput` custom element with different layer specs.
[10](https://github.com/deeplime-io/onecode/issues/10) | Allow for custom import and init statements in InputElement and OutputElement | Streamlit `app.py` generated through `onecode-start` now has placeholders for import and init statements. See `imports()` and `init()` static methods in `InputElement` and `OutputElement`.
[No ref] | Allow meta-data for elements | Meta-data can now be attached to any Input/Output element through `**kwargs`.
[No Ref] | Namespace logging | Logger (info, warning, critical and debug) now uses a namespace `|OneCode|` rather than the root logging.
New Features
Issue Ref | Summary | Description
-|-|-
[1](https://github.com/deeplime-io/onecode/issues/1) | New input element `FolderInput` | `FolderInput` allows folder selection. `LASConverter` example showcases it.
[2](https://github.com/deeplime-io/onecode/issues/2) | New output element `PlotlyOutput` | `PlotlyOutput` allows to visualize Plotly charts. `ExperimentalVariography` example showcases it.
[No ref] | Setup Project configuration through environment variables | Allow additional way to setup Project configuration options from command line.
:warning: Breaking changes
* [10](https://github.com/deeplime-io/onecode/issues/10) introduces a potential breaking change if you have custom `InputElement` or `OutputElement` where the `streamlit()` method relies on one of these statements being present in `app.py`:
python
import tkinter as tk
from tkinter import filedialog
import numpy as np
from pyarrow import csv as pacsv
_root = tk.Tk()
If this is the case, simply inherit `import()` and/or `init()` static methods so that they return the relevant statements.
See `FileInput.imports()` and `FileInput.init()` as examples.
* Custom `InputElement` or `OutputElement` can no longer initialized with an argument named `kind`:
it now is a reserved attribute. This change was triggered by a bug fix as `kind` was potentially overwritten
when extracting element through `onecode-extract`. Upgrade your custom element by changing your argument name.