nerd-fonts/bin/scripts/name_parser/name_parser_test1

#!/usr/bin/env python3
# coding=utf8

import sys
import re
import os.path
import fontforge
from FontnameParser import FontnameParser

###### Some helpers

def get_sfnt_dict(font):
    """Extract SFNT table as nice dict"""
    d = []
    for i, el in enumerate(font.sfnt_names):
        d += [(el[1], el[2])]
    return dict(d)

def format_names(header, *stuff):
    """Unify outputs (with header)"""
    f = '{:1.1}|{:50.50} |{:1.1}| {:50.50} |{:1.1}| {:30.30} |{:1.1}| {:30.30} |{:1.1}| {:30.30} |{:1.1}| {:.30}'
    if header:
        d = '------------------------------------------------------------'
        return f.format(*stuff) + '\n' + f.format('', d, d, d, d, d, d, d, d, d, d, d)
    return f.format(*stuff).rstrip()

def lenient_cmp(s1, s2):
    """Compare two font name (parts) but be a bit lenient ;->"""
    # We do not care about:
    # - Case
    # - "Display" vs "Disp" (in Noto)
    # Allow for "IBM 3278" name
    s = [ s1, s2 ]
    for i in range(2):
        # Usually given transform from 'their' to 'our' style
        s[i] = s[i].lower()
        s[i] = re.sub(r'\bdisp\b', 'display', s[i])                     # Noto
        s[i] = s[i].replace('ibm 3270', '3270')                         # 3270
        s[i] = s[i].replace('3270-', '3270 ')                           # 3270
        s[i] = s[i].replace('lekton-', 'lekton ')                       # Lekton
        s[i] = s[i].replace('semi-narrow', 'seminarrow')                # 3270
        s[i] = s[i].replace('bolditalic', 'bold italic')
        s[i] = re.sub(r'\bfor\b', '', s[i])                             # Meslo, Monofur
        s[i] = re.sub(r'\bpowerline\b', '', s[i])                       # Meslo, Monofur
        s[i] = s[i].replace('fira mono', 'fura mono')                   # Obviously someone forgot to rename the fonts in Fira/
        s[i] = s[i].replace('aurulentsansmono-', 'aurulent sans mono ') # Aurulent fullname oddity
        s[i] = s[i].replace('mononoki-', 'mononoki ')                   # Mononoki has somtimes a dash
        s[i] = re.sub(r'\br\b', 'regular', s[i])                        # Nonstandard style in Agave
        s[i] = re.sub(r'(bitstream vera sans mono.*) oblique', r'\1 italic', s[i]) # They call it Oblique but the filename says Italic
        s[i] = re.sub(r'gohufont (uni-)?(11|14)', 'gohufont', s[i])     # They put the 'name' into the subfamily/weight
        s[i] = s[i].replace('xltobl', 'extralight oblique')             # Iosevka goes inventing names
        s[i] = re.sub(r'proggyclean(?!TT)( ?)', 'proggycleantt\1', s[i]) # ProggyClean has no TT in filename

        s[i] = re.sub(r' +', ' ', s[i]).strip()
    return s[0] == s[1]

TEST_TABLE = [
    ( '(m)plus',   '\\1+'),
    ( 'IAWriter',  'iA Writer'),
    ( 'IBMPlex',   'IBM Plex'),
    ( 'Vera',      'Bitstream Vera Sans'),
]

###### Let's go!

if len(sys.argv) < 2:
    print('Usage: {} font_name [font_name ...]\n'.format(sys.argv[0]))
    sys.exit(1)

try:
    with open(sys.argv[0] + '.known_issues', 'r') as f:
        known_issues = f.read().splitlines()
        # known_issues = [line.rstrip() for line in known_issues]
        print('Found {:.0f} known issues'.format(len(known_issues) / 3)) # approx ;)
except OSError:
    print('Can not open known_issues file')
    known_issues = []
new_issues = open(sys.argv[0] + '.known_issues.new', 'w')

print('Examining {} font files'.format(len(sys.argv) - 1))
all_files = 0
issue_files = 0
known_files = 0

print(format_names(True, '', 'Filename', '', 'Fullname', '', 'Family', '', 'Subfamily', '', 'Typogr. Family', '', 'Typogr. Subfamily'))

for filename in sys.argv[1:]:
    fullfile = os.path.basename(filename)
    fname = os.path.splitext(fullfile)[0]
    if fname == 'NotoColorEmoji':
        continue # font is not patchable
    n = FontnameParser(fname).enable_short_families(False, 'Noto').add_name_substitution_table(TEST_TABLE)
    # Example for name injection:
    # n.inject_suffix("Nerd Font Complete Mono", "Nerd Font Complete M", "Nerd Font Mono")

    font = fontforge.open(filename, 1)
    sfnt = get_sfnt_dict(font)
    font.close()
    all_files += 1

    sfnt_full =    sfnt['Fullname']
    sfnt_fam =     sfnt['Family']
    sfnt_subfam =  sfnt['SubFamily']
    sfnt_pfam =    sfnt['Preferred Family'] if 'Preferred Family' in sfnt else ''
    sfnt_psubfam = sfnt['Preferred Styles'] if 'Preferred Styles' in sfnt else ''

    t1 = not lenient_cmp(sfnt_full,    n.fullname())
    t2 = not lenient_cmp(sfnt_fam,     n.family())
    t3 = not lenient_cmp(sfnt_subfam,  n.subfamily())
    t4 = not lenient_cmp(sfnt_pfam,    n.preferred_family())
    t5 = not lenient_cmp(sfnt_psubfam, n.preferred_styles())

    # Lenience: Allow for dropping unneeded prefered stuff:
    # New (sub)family is same as old preferred sub(family)
    if t4 and n.preferred_family() == '' and sfnt_pfam.lower() == n.family().lower():
        t4 = False
    if t5 and n.preferred_styles() == '' and sfnt_psubfam.lower() == n.subfamily().lower():
        t5 = False

    if t1 or t2 or t3 or t4 or t5:
        m1 = '+'; m2 = '-'
    else:
        m1 = ''; m2 = ''
    if not n.parse_ok:
        m1 = '!'
    t1_ = 'X' if t1 else ''
    t2_ = 'X' if t2 else ''
    t3_ = 'X' if t3 else ''
    t4_ = 'X' if t4 else ''
    t5_ = 'X' if t5 else ''

    o1 = format_names(False, m1, fullfile, t1_, n.fullname(), t2_, n.family(), t3_, n.subfamily(), t4_, n.preferred_family(), t5_, n.preferred_styles())
    o2 = format_names(False, m2, fullfile, '', sfnt_full, '', sfnt_fam, '', sfnt_subfam, '', sfnt_pfam, '', sfnt_psubfam)

    if len(m1):
        issue_files += 1
        if not o1 in known_issues or not o2 in known_issues:
            new_issues.writelines(['#### AUTOGENERATED\n', o1 + '\n', o2 + '\n'])
        else:
            known_files += 1
            idx = known_issues.index(o1) - 1 # should be the index of the explanation  line
            if known_issues[idx][0] != '#':
                sys.exit('Problem with known issues file, line', known_issues.index(o1))
            new_issues.writelines([known_issues[idx] + '\n', o1 + '\n', o2 + '\n'])
            # remove known_issue triplet
            known_issues.pop(idx)
            known_issues.pop(idx)
            known_issues.pop(idx)

    print(o1, o2, sep='\n')

print('Fonts with different name rendering: {}/{} ({}/{} are in known_issues)'.format(issue_files, all_files, known_files, issue_files))

if len(known_issues) > 0:
    print('There are {} lines not needed in known_issues, appending commented out in new known_issues'.format(len(known_issues)))
    new_issues.write('\n#### The following lines are not needed anymore\n\n')
    for l in known_issues:
        new_issues.writelines([' ', l, '\n'])

new_issues.close()