Exporting console as PNG (BytesIO)

[411A]

Hi, first of all, I want to say thanks for your awesome work.
It's been about 4 hours that I'm trying to get the beautiful syntax-highlighted SVG as PNG using libraries such as wand, svglib & cairosvg... But they do not generate a high-quality PNG, and many times they're struggling with the font and sometimes things like alpha channels.
If you add such an ability to get PNG out of our console (As BytesIO so we decide to send it and delete it from the memory or as .png file), that would be great. It would benefit many developers.
Thanks and Keep up the great work.

[github-actions[bot]]

Thank you for your issue. Give us a little time to review it.

PS. You might want to check the FAQ if you haven't done so already.

This is an automated reply, generated by FAQtory

[willmcgugan]

Exporting a PNG would require an additional dependencies that I wouldn't want to add I'm afraid.

[411A]

I understand & appreciate your work, but is this possible to add an option to the save_svg so we can change the font-family?
For now, I will write it down as a source for other users who want to achieve the same thing:
I used PyMuPDF & Ubuntu's inkscape to get the desired output as a list of BytesIO PNGs, although it won't display some Unicode characters and emojis (because of the nature of PDFs!?), though I tried SVG to PNG using inkscape and wand to, they don't display the colored emojis either (wand will messup line characters). I think I should go through some workarounds to convert HTML to image but most of them are opening the whole chromium browser just to do one conversion which is not promising...

Source Code

# pip install pandas
import pandas as pd

# pip install rich
from rich.table import Table
from rich.console import Console
from rich.terminal_theme import MONOKAI

# To name files temporarily
import time

# sudo apt-get install inkscape (~360 MB)
import subprocess as sp

# pip install PyMuPDF
# PyMuPDF - Convert PDF to PNG
import fitz
# BytesIO - Save images in memory
from io import BytesIO
# pahtlib.Path - Delete unnecessary files
from pathlib import Path


def rich_svg_pdf_png():

    # Stores BytesIO PNGs
    ListOfPNGsBytesIO = list()

    # Create a DataFrame
    df = pd.DataFrame([('User1', '🐧', 180, 40),
                    ('User2', '└', 200, 50),
                    ('User3', '𓁳', 210, 30)],
            columns=('Name', 'Character', 'Height', 'Score'))

    # Select the fields to display in the table
    table_data = df[
        [
            "Name", "Character", "Height",
            "Score",
        ]
    ]

    # Create a table
    #table = Table(title="Data Table", title_justify="center")
    table = Table()
    table.add_column("N", justify="center")
    table.add_column("Name", justify="center")
    table.add_column("Character", justify="center")
    table.add_column("Height", justify="center")
    table.add_column("Score", justify="center")

    # Add rows to the table
    for index, row in enumerate(table_data.itertuples()):
        table.add_row(
            str(index), str(row.Name), str(row.Character),
            str(row.Height), str(row.Score)
        )

    # Set the width of the table
    table.width = 115

    # Create a console
    console = Console(record=True)

    # Print the table
    console.print(table)

    # Set a name for SVG file, then delete it using pathlib.Path.unlink
    # tempnowtime, remove . from floating time to have a integrated integer number
    tempnowtime = str(time.time()).replace(".", "")
    SVGFileName = f"{tempnowtime}.svg"
    # Save as SVG file
    console.save_svg(
        path=SVGFileName, title=f"Users Table",
        theme=MONOKAI
    )

    PDFFileName = f"{tempnowtime}.pdf"

    # Convert SVG to PDF
    sp.run(["inkscape", SVGFileName, "-o", PDFFileName], stdout=sp.DEVNULL, stderr=sp.DEVNULL)

    # Open the PDF file
    pdf_doc = fitz.open(PDFFileName)

    # Iterate through the pages of the PDF file
    for page_num in range(pdf_doc.page_count):
        # Select the page
        pdf_page = pdf_doc.load_page(page_num)

        """ # Do something with the page, for example, save it as an image
        pix = pdf_page.get_pixmap()
        pix.save(f'page_{page_num + 1}.png') """

        # Render the page as a pixmap with a resolution of 300 DPI
        pixmap = pdf_page.get_pixmap(dpi=300)

        # Convert the pixmap to PNG bytes
        png_bytes = BytesIO(pixmap.tobytes("png"))

        # Append the png_bytes to the list
        ListOfPNGsBytesIO.append(png_bytes)

    # Close the PDF file
    pdf_doc.close()

    # Delete the unnecessary files
    resolve_svg = Path(SVGFileName).resolve()
    resolve_pdf = Path(PDFFileName).resolve()
    Path.unlink(resolve_svg)
    Path.unlink(resolve_pdf)

    return ListOfPNGsBytesIO

# Get BytesIO pics
get_pics = rich_svg_pdf_png()

# Iterate thorugh pics and write them as PNG file
for number, pic in enumerate(get_pics):
    with open(f"pic{number}.png", "wb") as f:
        f.write(pic.getbuffer())

[lolbat]

Changing the font won't work. I actually tested that as I am running into the same issue.

I suspect that it is related to the glyphs being used in the svg output. @willmcgugan is there a way to disable the glyphs in the svg output so I can test that hypothesis? Or is it possible to just do a quick find and replace? Do you know offhand which glyphs you are using?

There is a similar Inkscape issue being discussed that involves glyphs in Japanese generating the same error. Since Fira is being used I suspect that it is to use the code glyphs?

[FredHappyface]

https://github.com/FHPythonUtils/AnsiToImg might fit your use case? This stands on the shoulders of Textualize/rich (originally, I used custom implementations for these kind of exports). Hope this helps :)