console needs to more clearly indicate that no input is possible when code is running

[jjallaire]

To reproduce the code snippets below, start by downloading the data set from this Kaggle competition: https://oreil.ly/B9wfd. Then, make sure you are working within the directory where the dataset was extracted.

Consider the following Python code:

import pandas as pd
import numpy as np
from pathlib import Path
from fastai.tabular.all import add_datepart, cont_cat_split, Categorify, FillMissing, TabularPandas
from fastbook import draw_tree
from sklearn.tree import DecisionTreeRegressor

df = pd.read_csv("TrainAndValid.csv", low_memory=False)
sizes = ['Large', 'Large / Medium', 'Medium', 'Small', 'Mini', 'Compact']
df['ProductSize'] = df['ProductSize'].astype('category')
df['ProductSize'].cat.set_categories(sizes, ordered = True, inplace = True)
dep_var = 'SalePrice'
df[dep_var] = np.log(df[dep_var])
df = add_datepart(df, 'saledate')
df_test = pd.read_csv('Test.csv', low_memory=False)
df_test = add_datepart(df_test, 'saledate')

procs = [Categorify, FillMissing]
cond = (df.saleYear<2011) | (df.saleMonth<10)
train_idx = np.where(cond)[0]
valid_idx = np.where(~cond)[0]
splits = (list(train_idx),list(valid_idx))
cont, cat = cont_cat_split(df, 1, dep_var=dep_var)
to = TabularPandas(df, procs, cat, cont, y_names=dep_var, splits=splits)

xs,y = to.train.xs,to.train.y
valid_xs,valid_y = to.valid.xs,to.valid.y
m = DecisionTreeRegressor(max_leaf_nodes=4)
m.fit(xs, y)
draw_tree(m, xs, size=7, leaves_parallel=True, precision=2)

If you do a Cmd+A to Select All and then Cmd+Enter to run, you'll notice that all of the code is correctly submitted to the Console however the Console prompt is also printed (implying that it is "ready for input").

[seeM]

I personally find this very annoying and would like to consider bumping it to private alpha.

EDIT: To clarify, in data science contexts we are very often executing long-running code, and I've become used to seeing the cursor on the next line without a prompt indicator (>>>) as the main indication that something is running. Moreso than any stop button or other indicator. Although, I only realised this after seeing the current Positron behaviour.

EDIT2: While the green bar indicator on the left of the input is really nice, in data science contexts we also often have lots of info printed out (e.g. when training a model or processing data), which pushes that indicator out of visibility.

[jjallaire]

I agree (will do that now)

[softwarenerd]

This appears to have been fixed by work that was done recently. @jjallaire, could you try this again using a recent build (on / after August 24th)?

[jjallaire]

I've verified that the console doesn't show the prompt but in doing so discovered a new issue. If you follow the instructions above and then edit the code block to remove the import and invocation of the draw_tree() function:

import pandas as pd
import numpy as np
from pathlib import Path
from fastai.tabular.all import add_datepart, cont_cat_split, Categorify, FillMissing, TabularPandas
from sklearn.tree import DecisionTreeRegressor

df = pd.read_csv("bluebook/TrainAndValid.csv", low_memory=False)
sizes = ['Large', 'Large / Medium', 'Medium', 'Small', 'Mini', 'Compact']
df['ProductSize'] = df['ProductSize'].astype('category')
df['ProductSize'].cat.set_categories(sizes, ordered = True, inplace = True)
dep_var = 'SalePrice'
df[dep_var] = np.log(df[dep_var])
df = add_datepart(df, 'saledate')
df_test = pd.read_csv('bluebook/Test.csv', low_memory=False)
df_test = add_datepart(df_test, 'saledate')

procs = [Categorify, FillMissing]
cond = (df.saleYear<2011) | (df.saleMonth<10)
train_idx = np.where(cond)[0]
valid_idx = np.where(~cond)[0]
splits = (list(train_idx),list(valid_idx))
cont, cat = cont_cat_split(df, 1, dep_var=dep_var)
to = TabularPandas(df, procs, cat, cont, y_names=dep_var, splits=splits)

xs,y = to.train.xs,to.train.y
valid_xs,valid_y = to.valid.xs,to.valid.y
m = DecisionTreeRegressor(max_leaf_nodes=4)
m.fit(xs, y)

The call to m.fit() completely blows up the console (it actually disappears entirely, note the JS errors at right):

@jmcphers This line of code might be trying to show HTML, so I suspect the recent HTML output work we did?

[petetronic]

Confirming I see this issue in the Dev Tools Console still, I've split this out to its own issue https://github.com/posit-dev/positron/issues/1195 and close out this original issue as fixed.