import gradio as gr
import pandas as pd
from collections import defaultdict

df = pd.read_csv("./stackv2_languages_freq.csv")

langs = sorted(list(df["language"].unique()))
exts = list(df["extension"].unique())


lang_to_ext = defaultdict(set)

for lang, ext in zip(df["language"], df["extension"]):
    if isinstance(ext, str):
        lang_to_ext[lang].update([ext])

def compute(lang):
    
    report = f"## Summary:\n\n The `{lang}` language has {len(lang_to_ext[lang])} extensions: {', '.join(sorted(['`'+ext+'`' for ext in lang_to_ext[lang]]))}.\n\n"
    df_lang = df[df["language"]==lang]

    for i, (ext, example) in enumerate(zip(df_lang["extension"], df_lang["content"])):
        example_string = f"**Example {i+1} (extension=`{ext}`):**\n\n ```\n{example}\n```\n\n"
        report += example_string

    return report.strip()

with gr.Blocks() as demo:
    gr.Markdown(f"# Programming Languages for The Stack v2\n\nIn total there are **{len(langs)} languages** and **{len(exts)} extensions.**")

    lang_select = gr.Dropdown(choices=langs, value="Python")
    md = gr.Markdown("")

    lang_select.change(fn=compute, inputs=[lang_select], outputs=[md])
    demo.load(fn=compute, inputs=[lang_select], outputs=[md])
demo.launch()