How can I reliably retrieve the weight classification of a font (Condensed, Compressed, Expanded, Wide, Etc.) using FontTools / TTX / FontForge CLI? #3157
Replies: 1 comment 2 replies
-
|
What you are trying to do, If I well understood, is sorting fonts in folders. Since you have already installed ftCLI, you can just use https://github.com/ftCLI#ftcli-utils-font-organizer If the fonts are not grouped as a superfamily (width style is in Family Name, not in Subfamily Name), it will do exactly what asked. Otherwise, the question is a bit more complex. |
Beta Was this translation helpful? Give feedback.
-
|
I am indeed looking to sort fonts in folders, but in a very specific way. I only want to create sub-folders for font width. And I want to keep the width naming consistent. I don't see an option with font-organizer to sort fonts by width unfortunately. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, that works only if fonts with different widthClass have different family name. Running If you really want to sort fonts to in folders named after width style name, you should create a dict and map usWidthClass values to style names: import os
import click
from fontTools.misc.cliTools import makeOutputFileName
from ftCLI.Lib.Font import Font
from ftCLI.Lib.utils.cli_tools import check_input_path
from ftCLI.Lib.utils.click_tools import add_file_or_path_argument, generic_info_message, generic_error_message
WIDTHS = {
1: "Compressed",
2: "ExtraCondensed",
3: "Condensed",
4: "Narrow",
5: "Normal",
6: "Wide",
7: "Extended",
8: "ExtraExtended",
9: "Expanded"
}
@click.command()
@add_file_or_path_argument()
def cli(input_path):
files = check_input_path(input_path)
for file in files:
try:
font = Font(file)
width_class = font.os_2_table.usWidthClass
width_style_name = WIDTHS.get(width_class)
file_name = os.path.basename(file)
output_dir = os.path.join(os.path.dirname(file), width_style_name)
output_file = makeOutputFileName(file_name, outputDir=output_dir, overWrite=False)
os.renames(file, output_file)
generic_info_message(f"{os.path.basename(file)} moved")
except Exception as e:
generic_error_message(e)
if __name__ == "__main__":
cli()
This is a bit tricky: if you don't trust usWidthClass and want to recalculate it from a string, you can give a look here and adapt the function to your needs: https://github.com/ftCLI/ftCLI/blob/98e9a489e95f4486159081f971d5807fda0ba8b3/ftCLI/Lib/assistant/fonts_data.py#L31 |
Beta Was this translation helpful? Give feedback.
-
I am writing a script that aims to reliably sort a font family into it's individual weights if present. For instance:
After the script runs, it should create the following directories:
Condensed,ExtraCondensed,SemiCondensed,Wide- and then move the appropriate fonts to these folders.The small dataset above is just for illustration, obviously I could just search for these keywords in the filenames but this isn't at all reliable as many fonts are named more cryptically or follow different naming conventions.
I've got a Python VENV set up with
FontTools,TTX, andftcliset up in an attempt to extract metadata so I can make more informed decisions in my script.I have a script that uses
ttLibto extract theOS/2usWidthClass:And a script to dump metadata using
FontMeta:But I've found the usWidthClass to be hit or miss. A lot of typographers just set this value as
5(Normal) no matter what width the font is. And the data extracted from FontMeta doesn't really supply any useful information for this problem.I could really use some guidance on how to move forward and scrape the information I need. Right now I'm just doing regex gymnastics on the supplied filenames and suspect there is a much more reliable way to go about this.
Any help greatly appreciated.
Beta Was this translation helpful? Give feedback.
All reactions