Dispersion plot with Shiny for Python

app.py

from collections import Counter

import matplotlib as mpl
import matplotlib.pyplot as plt
import pandas as pd
import regex as re
from shiny import App, reactive, render, ui

mpl.rcParams.update(
    {
        "axes.grid": True,
        "axes.grid.axis": "both",
        "axes.spines.left": False,
        "axes.spines.right": False,
        "axes.spines.top": False,
        "axes.spines.bottom": False,
    }
)

app_ui = ui.page_fluid(
    ui.panel_title("Lexical dispersion plot"),
    ui.layout_sidebar(
        ui.panel_sidebar(
            ui.input_text_area("text", "Text to analyze"),
            ui.input_text("words", "Space-separated words to plot"),
            ui.input_checkbox("icase", "Ignore case"),
        ),
        ui.panel_main(
            ui.output_plot("dispersion_plot"),
            ui.output_table("freq_dist"),
        ),
    ),
)


def tokenize(text):
    return re.findall(
        r"\p{Alphabetic}+(?:\S+\p{Alphabetic}+)*|[\S&&\P{Alphabetic}]+",
        text,
        flags=re.VERSION1,
    )


def server(input, output, session):
    @reactive.Calc
    def tokenized_text():
        text = input.text().strip()
        if not text:
            ui.notification_show("Please provide an input text.", type="warning")
            return []
        if input.icase():
            text = text.lower()
        return tokenize(text)

    @reactive.Calc
    def split_words():
        words = input.words().strip()
        if not words:
            ui.notification_show("Please provide words to plot.", type="warning")
            return {}
        if input.icase():
            words = words.lower()
        return {w: i for (i, w) in enumerate(reversed(words.split()))}

    @output
    @render.plot(alt="Dispersion plot of chosen words in input text")
    def dispersion_plot():
        text = tokenized_text()
        words = split_words()
        if not (text and words):
            return

        xs = []
        ys = []
        for i, tok in enumerate(text):
            if tok in words:
                xs.append(i)
                ys.append(words[tok])

        fig, ax = plt.subplots()
        ax.plot(xs, ys, marker="|", markersize=20, linestyle="")
        ax.set_yticks(range(len(words)), labels=list(words))
        ax.tick_params(axis="both", length=0)
        ax.set_xlabel("Word offset")
        return fig

    @output
    @render.table()
    def freq_dist():
        text = tokenized_text()
        words = set(split_words())
        if not (text and words):
            return

        freq_dist = Counter(t for t in text if t in words)
        df = pd.DataFrame(freq_dist.most_common(), columns=["word", "frequency"])
        return df


app = App(app_ui, server)

requirements.txt

Jinja2