app.py

import pandas as pd
import numpy as np

import os
import io

from datetime import datetime
from datetime import timedelta

import holoviews as hvs
import panel as pn
import hvplot as hv
import hvplot.pandas

import yfinance as yf

from dotenv import load_dotenv

# from datasets import load_dataset
from utils import load_hf_dataset  # ,upload_to_hf_dataset, download_from_hf_dataset

pn.extension("bokeh", template="bootstrap")


def _extract_raw_data(ticker):
    df = yf.Ticker(ticker)
    df = df.history(period="1y", interval="1d").reset_index()
    df['SMA50'] = df['Close'].rolling(window=50).mean()
    df['SMA200'] = df['Close'].rolling(window=200).mean()
    return df


def _transform_data(raw_data: pd.DataFrame):
    data = (
        raw_data[["Date", "Open", "High", "Low", "Close", "Volume" ,"SMA50", "SMA200"]]
        .copy(deep=True)
        .rename(
            columns={
                "Date": "time",
                "Open": "open",
                "High": "high",
                "Low": "low",
                "Close": "close",
                "Volume": "volume",
            }
        )
    )
    t_delta = timedelta(hours=1)
    data["time_start"] = data.time - 9 * t_delta  # rectangles start
    data["time_end"] = data.time + 9 * t_delta  # rectangles end
    data["positive"] = ((data.close - data.open) > 0).astype(int)
    return data


def make_candle_stick(ticker):
    raw_data = _extract_raw_data(ticker=ticker)
    data = _transform_data(raw_data=raw_data)
    _delta = np.median(np.diff(data.time))
    candlestick = hvs.Segments(
        data, kdims=["time", "low", "time", "high"]
    ) * hvs.Rectangles(
        data, kdims=["time_start", "open", "time_end", "close"], vdims=["positive"]
    ) * hvs.Curve(
        data, kdims=["time"], vdims=[ "SMA50"]
    )
    candlestick = candlestick.redim.label(Low="Values")
    candlechart = pn.Column(
        candlestick.opts(
            hvs.opts.Rectangles(
                color="positive", cmap=["red", "green"], responsive=True
            ),
            hvs.opts.Segments(
                color="black", height=400, responsive=True, show_grid=True
            ),
        ),
        data.hvplot(
            x="time", y="volume", kind="line", responsive=True, height=200
        ).opts(show_grid=True),
    )
    #  data.hvplot(y="volume", kind="bar", responsive=True, height=200) )
    return candlechart


# Function to convert DataFrame to CSV
def get_csv(df):
    sio = io.StringIO()
    df.to_csv(sio, index=False)
    sio.seek(0)
    return sio


# Function to convert the 'Ticker' column to a comma-separated string in a text file
def get_text(df):
    tickers = df["Ticker"].tolist()
    tickers_str = ",".join(tickers)
    sio = io.StringIO()
    sio.write(tickers_str)
    sio.seek(0)
    return sio


# Load environment variables from .env file
load_dotenv()

# Get the Hugging Face API token from the environment; either set in .env file or in the environment directly in GitHub
HF_TOKEN = os.getenv("HF_TOKEN")

# Get the name of the GuruFocus dataset for TradingView, Finviz, MarketBeat and TipRanks to read from
dataset_name_BeatingMarket_input = os.getenv("dataset_name_BeatingMarket_input")
dataset_name_TradingView_input = os.getenv("dataset_name_TradingView_input")
dataset_name_GuruFocus_input = os.getenv("dataset_name_GuruFocus_input")
dataset_name_FinViz_input = os.getenv("dataset_name_FinViz_input")
dataset_name_TipRanks_input = os.getenv("dataset_name_TipRanks_input")


@pn.cache()
def get_tradingview(current_datetime):
    # Load lastest TradingView DataSet from HuggingFace Dataset which is always america.csv
    # download_from_hf_dataset("america.csv", dataset_name_TradingView_input, HF_TOKEN)
    return load_hf_dataset("america.csv", HF_TOKEN, dataset_name_TradingView_input)


@pn.cache()
def get_marketbeat(current_datetime):
    return load_hf_dataset(
        "beatingmarket.csv", HF_TOKEN, dataset_name_BeatingMarket_input
    )


@pn.cache()
def get_gurufocus(current_datetime):
    # Load lastest GuruFocus DataSet from HuggingFace Dataset which is always GuruFocus_merged.csv
    # daily_gurufocus_DF = load_dataset(dataset_name_GuruFocus_output , data_files="GuruFocus_merged.csv", split="train" , token=HF_TOKEN ).to_pandas()
    return load_hf_dataset(
        "GuruFocus_merged.csv", HF_TOKEN, dataset_name_GuruFocus_input
    )


@pn.cache()
def get_finviz(current_datetime):
    # Load lastest FinViz DataSet from HuggingFace Dataset which is always FinViz.csv
    return load_hf_dataset("FinViz.csv", HF_TOKEN, dataset_name_FinViz_input)


@pn.cache()
def get_tipranks(current_datetime):
    # Load lastest TipRanks DataSet from HuggingFace Dataset which is always tipranks.csv
    return load_hf_dataset("tipranks.csv", HF_TOKEN, dataset_name_TipRanks_input)


# Reading gurufocus,finviz from crawling pipelines with GitHub Action
current_datetime = datetime.now().strftime("%Y-%m-%d")
daily_gurufocus_DF = get_gurufocus(current_datetime)
daily_finviz_DF = get_finviz(current_datetime)
daily_marketbeat = get_marketbeat(current_datetime)
monthly_tiprank_DF = get_tipranks(current_datetime)

# TipRanks
monthly_tiprank_DF = monthly_tiprank_DF[["Ticker", "SmartScore"]]
monthly_tiprank_DF["SmartScore"] = monthly_tiprank_DF["SmartScore"].fillna(-1).astype(float)

# FinViz
daily_finviz_DF["FinVizPrice"] = (
    pd.to_numeric(daily_finviz_DF["Price"], errors="coerce").fillna(0).astype(float)
)
daily_finviz_DF["FinVizTarget"] = (
    pd.to_numeric(daily_finviz_DF["Target Price"], errors="coerce")
    .fillna(0)
    .astype(float)
)
daily_finviz_DF["FinVizTargetpercent"] = (
    100
    * (daily_finviz_DF["FinVizTarget"] - daily_finviz_DF["FinVizPrice"])
    / daily_finviz_DF["FinVizPrice"]
).round(2)
daily_finviz_DF_filtered = daily_finviz_DF[
    ["Ticker", "Price", "FinVizTarget", "FinVizTargetpercent"]
]

# Merging tipranks with Gurufocus
DFgurufocus = daily_gurufocus_DF[
    ["Ticker", "GFValue", "Market Capitalization", "Sector", "Industry"]
]  # , 'GFValuediff']]
DFmerge_tipranks_gurufocus = DFgurufocus.merge(monthly_tiprank_DF)

# Merging Finviz with Merged last one
DFmerge_tipranks_gurufocus = DFmerge_tipranks_gurufocus.merge(daily_finviz_DF_filtered)

# ['Ticker', 'AverageTarget', 'MaxTarget', 'LowTarget', 'CurrentPrice']
daily_marketbeat["MarketBeat"] = (
    (daily_marketbeat["MaxTarget"] > daily_marketbeat["CurrentPrice"])
    & (daily_marketbeat["LowTarget"] > daily_marketbeat["CurrentPrice"])
    & (daily_marketbeat["AverageTarget"] > daily_marketbeat["CurrentPrice"])
).astype(int)

# Merging daily_marketbeat with Merged last one
print(80 * "&")
print(DFmerge_tipranks_gurufocus.columns)
print(daily_marketbeat.columns)
# DFmerge_tipranks_gurufocus = DFmerge_tipranks_gurufocus.merge(daily_marketbeat)
DFmerge_tipranks_gurufocus = DFmerge_tipranks_gurufocus.merge(
    daily_marketbeat, how="left", on="Ticker"
)

if (
    "Price" in DFmerge_tipranks_gurufocus.columns
    and "GFValue" in DFmerge_tipranks_gurufocus.columns
):
    DFmerge_tipranks_gurufocus["GFValuepercent"] = (
        100
        * (DFmerge_tipranks_gurufocus["GFValue"] - DFmerge_tipranks_gurufocus["Price"])
        / DFmerge_tipranks_gurufocus["Price"]
    )
    DFmerge_tipranks_gurufocus["GFValuepercent"] = DFmerge_tipranks_gurufocus[
        "GFValuepercent"
    ].round(2)
    DFmerge_tipranks_gurufocus["Market Capitalization"] = (
        DFmerge_tipranks_gurufocus["Market Capitalization"] / 1e9
    )
    DFmerge_tipranks_gurufocus["MarketCap"] = DFmerge_tipranks_gurufocus[
        "Market Capitalization"
    ].round(1)
    DFmerge_tipranks_gurufocus = DFmerge_tipranks_gurufocus.drop(
        columns=["Market Capitalization"]
    )
    DFmerge_tipranks_gurufocus = DFmerge_tipranks_gurufocus[
        [
            "Ticker",
            "Sector",
            "Industry",
            "MarketCap",
            "SmartScore",
            "Price",
            "GFValue",
            "GFValuepercent",
            "FinVizTarget",
            "FinVizTargetpercent",
            "MarketBeat",
        ]
    ]

# widget

tickername = pn.state.location.query_params.get("ticker", "No ticker provided")
if tickername == "No ticker provided":
    ticker = pn.widgets.AutocompleteInput(
        name="Ticker",
        options=list(DFmerge_tipranks_gurufocus.Ticker),
        placeholder="Write Ticker here همین جا",
        value="ALL",
        restrict=False,
    )
else:
    ticker = pn.widgets.AutocompleteInput(
        name="Ticker",
        options=list(DFmerge_tipranks_gurufocus.Ticker),
        placeholder="Write Ticker here همین جا",
        value=f"{tickername}",
        restrict=False,
    )

SmartScore = pn.widgets.EditableRangeSlider(
    name="Smart Score", start=-1, end=10, value=(-1, 10), step=1
)
Industry = pn.widgets.CheckBoxGroup(
    name="Select Industry",
    value=list(set(DFmerge_tipranks_gurufocus.Industry)),
    options=list(set(DFmerge_tipranks_gurufocus.Industry)),
    inline=True,
)
Sector = pn.widgets.MultiSelect(
    name="Select Sector",
    value=list(set(DFmerge_tipranks_gurufocus.Sector)),
    options=list(set(DFmerge_tipranks_gurufocus.Sector)),
)  # , inline=False)
GFValuepercent = pn.widgets.FloatSlider(
    name="GF Value %", start=-100, end=100, step=1, value=12.0
)
FinVizTargetpercent = pn.widgets.FloatSlider(
    name="FinViz Target %", start=-100, end=100, step=1, value=12.0
)
MarketCap = pn.widgets.FloatSlider(
    name="Market Capital (B$)", start=0, end=1000, step=1, value=10
)
MarketBeat_radio_group = pn.widgets.RadioButtonGroup(
    name="MarketBeat",
    options=["Default Targets", "Above all Targets"],
    button_type="success",
    value="Above all Targets",
)

# Create a file download link
download_button_GuruFocus = pn.widgets.FileDownload(
    filename=f"GuruFocus_merged_{current_datetime}.csv",
    callback=pn.bind(get_csv, daily_gurufocus_DF),
    button_type="primary",
    label="Download GuruFocus",
    width=300,
)

download_button_FinViz = pn.widgets.FileDownload(
    filename=f"FinViz_{current_datetime}.csv",
    callback=pn.bind(get_csv, daily_finviz_DF),
    button_type="primary",
    label="Download FinViz",
    width=300,
)

download_button_MarketBeat = pn.widgets.FileDownload(
    filename=f"MarketBeat_{current_datetime}.csv",
    callback=pn.bind(get_csv, daily_marketbeat),
    button_type="warning",
    label="Download MarketBeat",
    width=300,
)

download_button_TipRanks = pn.widgets.FileDownload(
    filename=f"TipRanks_{current_datetime}.csv",
    callback=pn.bind(get_csv, monthly_tiprank_DF),
    button_type="warning",
    label="Download TipRanks",
    width=300,
)


def get_DF_filter(
    DF, ticker, SmartScore, GFValuepercent, FinVizTargetpercent, Sector, MarketCap
):
    return DF.query(
        "SmartScore>=@SmartScore[0] & SmartScore <= @SmartScore[1] & GFValuepercent>=@GFValuepercent & FinVizTargetpercent>@FinVizTargetpercent & Sector in @Sector & MarketCap>@MarketCap"
    )


download_button_Filter = pn.widgets.FileDownload(
    filename=f"Filter_{current_datetime}.csv",
    callback=pn.bind(
        get_csv,
        pn.bind(
            get_DF_filter,
            DF=DFmerge_tipranks_gurufocus,
            ticker=ticker,
            SmartScore=SmartScore,
            GFValuepercent=GFValuepercent,
            FinVizTargetpercent=FinVizTargetpercent,
            Sector=Sector,
            MarketCap=MarketCap,
        ),
    ),
    button_type="danger",
    label="Download your Filter",
    width=300,
)

download_button_watchlist = pn.widgets.FileDownload(
    filename=f"watchlist_{current_datetime}.txt",
    callback=pn.bind(
        get_text,
        pn.bind(
            get_DF_filter,
            DF=DFmerge_tipranks_gurufocus,
            ticker=ticker,
            SmartScore=SmartScore,
            GFValuepercent=GFValuepercent,
            FinVizTargetpercent=FinVizTargetpercent,
            Sector=Sector,
            MarketCap=MarketCap,
        ),
    ),
    button_type="success",
    label="Download Watchlist",
    width=300,
)

png_pane = pn.pane.GIF(
    "https://cdn.dribbble.com/userupload/23265870/file/original-619180a5b218d92d799a2d60617a09fb.gif",
    width=300,
)  # , height=150)
png_pane2 = pn.pane.GIF(
    "https://www.mywallstreetadvisor.com/_next/image?url=%2FMyWallStreetAdvisorLOGO.png&w=64&q=75",
    width=50,
)  # , height=150)

pn.extension('echarts')

def get_gauge(text, value, low, high):
    if value:
        return pn.indicators.Gauge(
            name=f'{text}', value=value, bounds=(low, high), format='{value}',
            colors=[(0.2, 'red'), (0.8, 'gold'), (1, 'green')], width=300, height=300)


def get_DF(
    DF,
    ticker,
    SmartScore,
    GFValuepercent,
    FinVizTargetpercent,
    Sector,
    MarketCap,
    MarketBeat,
):
    if ticker and ticker != "ALL":
        table1 = pn.widgets.Tabulator(
            DF.query("Ticker == @ticker"), height=200, widths=200, show_index=False
        )
        table1.disabled = True
        chart1 = make_candle_stick(ticker)

        url_button_finviz = pn.widgets.Button(
            name=f"{ticker} on FinViz", button_type="primary", width=250
        )
        url_button_finviz.js_on_click(
            code=f"window.open('https://finviz.com/quote.ashx?t={ticker}&p=d#', '_blank')"
        )

        url_button_benzinga = pn.widgets.Button(
            name=f"{ticker} on Benzinga", button_type="primary", width=250
        )
        url_button_benzinga.js_on_click(
            code=f"window.open('https://www.benzinga.com/quote/{ticker}', '_blank')"
        )

        url_button_tipranks = pn.widgets.Button(
            name=f"{ticker} on TipRanks", button_type="warning", width=250
        )
        url_button_tipranks.js_on_click(
            code=f"window.open('https://www.tipranks.com/stocks/{ticker}/forecast', '_blank')"
        )

        url_button_gurufocus = pn.widgets.Button(
            name=f"{ticker} on GuruFocus", button_type="warning", width=250
        )
        url_button_gurufocus.js_on_click(
            code=f"window.open('https://gurufocus.com/stock/{ticker}', '_blank')"
        )

        url_button_barchart = pn.widgets.Button(
            name=f"{ticker} on BarChart", button_type="success", width=250
        )
        url_button_barchart.js_on_click(
            code=f"window.open('https://www.barchart.com/stocks/quotes/{ticker}/expected-move', '_blank')"
        )

        url_button_optionchart = pn.widgets.Button(
            name=f"{ticker} on optionchart", button_type="danger", width=250
        )
        url_button_optionchart.js_on_click(
            code=f"window.open('https://optioncharts.io/options/{ticker}', '_blank')"
        )

        url_button_tradingview = pn.widgets.Button(
            name=f"{ticker} on tradingview", button_type="light", width=250
        )
        url_button_tradingview.js_on_click(
            code=f"window.open('https://www.tradingview.com/symbols/NASDAQ-{ticker}/technicals/', '_blank')"
        )

        url_button_marketwatch = pn.widgets.Button(
            name=f"{ticker} on marketwatch", button_type="danger", width=250
        )
        url_button_marketwatch.js_on_click(
            code=f"window.open('https://www.marketwatch.com/investing/stock/{ticker}', '_blank')"
        )

        url_button_chartmill_TA = pn.widgets.Button(
            name=f"{ticker} on ChartMill TA", button_type="danger", width=250
        )
        url_button_chartmill_TA.js_on_click(
            code=f"window.open('https://www.chartmill.com/stock/quote/{ticker}/technical-analysis', '_blank')"
        )

        url_button_chartmill_FA = pn.widgets.Button(
            name=f"{ticker} on ChartMill FA", button_type="danger", width=250
        )
        url_button_chartmill_FA.js_on_click(
            code=f"window.open('https://www.chartmill.com/stock/quote/{ticker}/fundamental-analysis', '_blank')"
        )

        url_button_seekingalpha = pn.widgets.Button(
            name=f"{ticker} on seekingalpha", button_type="light", width=250
        )
        url_button_seekingalpha.js_on_click(
            code=f"window.open('https://seekingalpha.com/symbol/{ticker}', '_blank')"
        )

        return pn.Column(
            table1,
            chart1,
            pn.Column(
                pn.Row(
                url_button_finviz,
                url_button_benzinga,
                url_button_gurufocus,
                url_button_tipranks,
                url_button_barchart,
                url_button_optionchart,
                url_button_tradingview,
                url_button_marketwatch,
            ),
                pn.Row(
                   url_button_chartmill_TA,
                   url_button_chartmill_FA,
                   url_button_seekingalpha

                ),
                pn.Row(
                    get_gauge('SmartScore', float(DF.query('Ticker == @ticker').SmartScore.iloc[0]), 0, 10),
                    get_gauge('GuruFocus %', float(DF.query('Ticker == @ticker').GFValuepercent.iloc[0]), -100, 100),
                    get_gauge('FinViz %', float(DF.query('Ticker == @ticker').FinVizTargetpercent.iloc[0]), -100, 100),
                     align="center") )
            )
        
    else:
        if MarketBeat == "Above all Targets":
            tabulator = pn.widgets.Tabulator(
                DF.query(
                    "SmartScore>=@SmartScore[0] & SmartScore <= @SmartScore[1] & GFValuepercent>=@GFValuepercent & FinVizTargetpercent>@FinVizTargetpercent & Sector in @Sector & MarketCap>@MarketCap & MarketBeat==1"
                ),
                height=800,
                widths=200,
                show_index=False,
                theme="modern",
            )
        else:
            tabulator = pn.widgets.Tabulator(
                DF.query(
                    "SmartScore>=@SmartScore[0] & SmartScore <= @SmartScore[1] & GFValuepercent>=@GFValuepercent & FinVizTargetpercent>@FinVizTargetpercent & Sector in @Sector & MarketCap>@MarketCap"
                ),
                height=800,
                widths=200,
                show_index=False,
                theme="modern",
            )
        tabulator.disabled = True
        return tabulator


pn.extension("tabulator")
bound_plot = pn.bind(
    get_DF,
    DF=DFmerge_tipranks_gurufocus,
    ticker=ticker,
    SmartScore=SmartScore,
    GFValuepercent=GFValuepercent,
    FinVizTargetpercent=FinVizTargetpercent,
    Sector=Sector,
    MarketCap=MarketCap,
    MarketBeat=MarketBeat_radio_group,
)

pn.Column(
    pn.Row(
        pn.Column(
            pn.Row(png_pane2, align="center"),
            ticker,
            SmartScore,
            GFValuepercent,
            FinVizTargetpercent,
            MarketCap,
            pn.Row(MarketBeat_radio_group, align="center"),
            png_pane,
            download_button_GuruFocus,
            download_button_FinViz,
            download_button_MarketBeat,
            download_button_TipRanks,
            download_button_Filter,
            download_button_watchlist,
        ),
        pn.Row(bound_plot, align="center"),
    )
).servable(
    title="Fair Value Ranking - Merged Gurufocus & Tipranks & FinViz & MorningStars & MarketBeat"
)