def export_to_html(fig: Figure, div_id: str) -> str:
    """
    Export plotly graph to html format

    Args:
        fig (Figure): Plotly figure to export
        div_id (str): String to use as the plot's div id.

    Returns:
        str: HTML string with plot
    """
    config = {"modeBarButtonsToRemove": ["zoom", "select", "lasso2d"]}

    return io.to_html(
        fig,
        full_html=False,
        include_plotlyjs=False,
        include_mathjax=False,
        div_id=div_id,
        config=config,
    )

def generate_taxo_plots(tax_data: Dict, contrast_df=None, output_path=None) -> Dict:
    """
    Get all taxonomy plots

    Master function to generate all plots to be used in the final report.

    Args:
        tax_data (dict): Dict resulting from
            microview.parse_taxonomy.get_tax_data
        contrast_df (pd.DataFrame): Dataframe with sample names and
            contrasts, if available.

    Returns:
        dict: Dict containing all plots, one for each key.
    """
    assigned = bar(
        tax_data["sample n reads"],
        x="index",
        y="value",
        color="variable",
        labels={
            "value": "% of reads",
            "index": "Sample name",
            "variable": "Category",
        },
        template="plotly_white",
    )
    write_table(tax_data["sample n reads"], output_path, "classified_reads.tsv")

    assigned.update_layout(
        xaxis={"categoryorder": "category ascending"},
    )

    assigned_html = export_to_html(assigned, "assigned-plot")

    # Beta diversity plots
    plot_beta_div = False if tax_data["beta div"] is None else True

    if plot_beta_div:
        pcoa_embed = (
            tax_data["beta div"]
            .samples[["PC1", "PC2"]]
            .rename_axis("sample")
            .reset_index()
        )
        var_explained = (
            tax_data["beta div"]
            .proportion_explained[:9]
            .to_frame(name="Variance Explained")
            .reset_index()
            .rename(columns={"index": "PC"})
        )

        write_table(var_explained, output_path, "pcoa_variance_explained.tsv")

        pcoa_var = line(
            var_explained,
            x="PC",
            y="Variance Explained",
            text="PC",
            template="plotly_white",
        )
        pcoa_var.update_traces(textposition="bottom right")

    # TODO: Improve this check
    if contrast_df is not None and "group" in contrast_df.columns:
        contrast_df["sample"] = [
            str(Path(s).name) for s in contrast_df["sample"].to_list()
        ]
        merged_taxas_df = merge_with_contrasts(tax_data["common taxas"], contrast_df)

        common_taxas = plot_common_taxas(
            merged_taxas_df, output_path, facet_col="group"
        )
        common_taxas.update_xaxes(matches=None)

        abund_div = plot_abund_div(
            merge_with_contrasts(tax_data["abund and div"], contrast_df),
            output_path,
            color="group",
        )
        if plot_beta_div:
            betadiv_pcoa = plot_beta_pcoa(
                merge_with_contrasts(pcoa_embed, contrast_df, left_colname="sample"),
                output_path,
                color="group",
            )

    else:
        common_taxas = plot_common_taxas(tax_data["common taxas"], output_path)

        abund_div = plot_abund_div(tax_data["abund and div"], output_path)
        if plot_beta_div:
            betadiv_pcoa = plot_beta_pcoa(pcoa_embed, output_path)

    common_taxas.update_traces(showlegend=False)
    common_taxas.update_layout(
        xaxis={"categoryorder": "category ascending"},
    )

    return_dict = {
        "assigned_plot": assigned_html,
        "common_taxas_plot": export_to_html(common_taxas, "taxas-plot"),
        "abund_div_plot": export_to_html(abund_div, "abund-div-plot"),
    }

    if plot_beta_div:
        return_dict["pcoa_var_plot"] = export_to_html(
            pcoa_var, "pcoa-explained-variance"
        )
        return_dict["beta_div_pcoa"] = export_to_html(betadiv_pcoa, "betadiv_pcoa")

    return return_dict

def merge_with_contrasts(df, contrast_df, left_colname: Optional[str] = "index"):
    """
    Merges a dataframe with the dataframe containing contrasts (or groups)
    """
    merged_df = df.merge(
        contrast_df, left_on=left_colname, right_on="sample", how="left"
    )

    return merged_df

def plot_abund_div(abund_div_df, output_path, **kwargs):
    """
    Generate scatter plot of Pielou's Evenness and Shannon's Diversity (alpha)
    """
    write_table(abund_div_df, output_path, "abund_diversity.tsv")

    return scatter(
        abund_div_df,
        x="Pielou Evenness",
        y="Shannon Diversity",
        size="N Taxas",
        hover_data=["index"],
        labels={"N Taxas": "# taxas"},
        template="plotly_white",
        color_discrete_sequence=colors.qualitative.Safe[3:],
        **kwargs,
    )

def plot_beta_pcoa(beta_pcoa, output_path, **kwargs):
    """
    Generate scatter plot of two first coordinates of Beta Diversity PCoA
    """
    write_table(beta_pcoa, output_path, "beta_pcoa.tsv")

    fig = scatter(
        beta_pcoa,
        x="PC1",
        y="PC2",
        hover_data=["sample"],
        labels={"sample": "Sample name"},
        template="plotly_white",
        **kwargs,
    )
    fig.update_traces(marker_size=10)
    return fig

def plot_common_taxas(common_taxas_df, output_path, **kwargs):
    """
    Generate bar plot with most common taxas
    """
    write_table(common_taxas_df, output_path, "common_taxas.tsv")

    return bar(
        common_taxas_df.sort_values(by=["value", "variable"], ascending=[False, True]),
        x="index",
        y="value",
        color="variable",
        labels={
            "index": "Sample name",
            "value": "% of reads",
            "variable": "Taxon name",
        },
        template="plotly_white",
        color_discrete_sequence=colors.qualitative.Alphabet,
        **kwargs,
    )

def write_table(df, output_path, path):
    """
    Write a dataframe to a file
    """
    dirpath = Path(output_path).parent.resolve() / Path("microview_tables")
    Path(dirpath).mkdir(exist_ok=True)
    path = dirpath / path

    df.to_csv(path, sep="\t", index=False)