FontForge "Community guidelines" authoring script

manage_guidelines.py

#!/usr/bin/python3
# This program outputs the "Link to this section" links, and also a table of
# contents for the "Community guidelines" wiki FontForge page.
#
# Warning: pretty hacky...
#
# Original author: Fredrick R. Brennan
# License: FontForge original BSD license
#
# Run as e.g.:
# ./manage_guidelines.py < cgs.md > /tmp/cgs.md
#
# "Never spend 6 minutes doing something by hand when you can spend 6 hours failing to automate it." ~ Zhuowei Zhang (σ╝╡σìôσüë)

import re
import sys

cg = sys.stdin.read()
lines = cg.splitlines()

# Find our generated TOC and delete it if applicable.
TOCSTART = '# Table of Contents <a name="toc"/><!-- TOCSTART -->'
TOCEND = '<!-- TOCEND -->\n'
TOCi = None
ntoc = 0
for i, line in enumerate(lines):
    if not TOCSTART in line:
        continue
    else:
        TOCi = i

    j = 0
    while lines[i+j+1].lstrip().startswith('*'):
        ntoc+=1
        j   +=1

    if TOCEND[:-1] in lines[i+j+1]:
        ntoc+=2

    if ntoc!=0: 
        ntoc+=1

    break

if TOCi is not None:
    for i in range(TOCi, TOCi+ntoc):
        del lines[TOCi]

# HTML parser to find anchors and remember the last found.
from html.parser import HTMLParser

class AnchorParser(HTMLParser):
    @staticmethod
    def attrs_to_dict(attrs):
        ret = dict()

        for k, v in attrs:
            if k in ret:
                raise ValueError("Double attribute in tag")
            ret.setdefault(k, v)

        return ret

    curanchor = None
    parsed = 0
    def handle_starttag(self, tag, attrs):
        attrs = AnchorParser.attrs_to_dict(attrs)
        if tag == "a" and "name" in attrs:
            self.parsed+=1
            self.curanchor = attrs["name"]

parser = AnchorParser()

# Write out the "quick links" and populate the TOC "tree" (which is really not a tree data structure per se)
anchors = list()
outlines = list() # We don't want to output in the loop as it might error.
tree = list()
firstheading = None
QUICKLINK = '<!-- QUICKLINK -->'
for i, line in enumerate(lines):
    if not QUICKLINK in line: outlines.append(line)

    if not re.match(r'^#', line) or TOCSTART in line:
        continue

    parsedb4 = parser.parsed
    parser.feed(line)

    if parser.parsed == parsedb4 or parser.curanchor == None:
        raise ValueError("No anchor in heading `{}`, add one.".format(line[line.index(' ')+1:]))

    if parser.curanchor in anchors:
        raise ValueError("Document has two anchors with name `{}`, cannot continue. Rename one of them.".format(parser.curanchor))

    outlines.append("<sup>[Quick link to section](#{})</sup>".format(parser.curanchor)+QUICKLINK)
    # Add a blank line after link if one isn't already there, and doing so is "appropriate"
    if (i+1 <= len(lines)-1) and ((len(lines[i+1].strip()) > 0 and lines[i+1][0] == '<') ^ \
                                  (len(lines[i+1].strip()) > 0)):
        outlines.append("")
    
    tree.append((line.count('#', 0, line.index(' ')), parser.curanchor, line[line.index(' ')+1:line.index('<')]))

    if firstheading is None: firstheading = i
    anchors.append(parser.curanchor)

# Write our TOC
toc = list()
toc.append(TOCSTART)
for level, anchor, name in tree:
    toc.append("""{} <a href="#{}">{}</a>""".format(' '*2*(level-1)+'*', anchor, name))
toc.append(TOCEND)

if firstheading is None:
    raise ValueError("No headings, add one.")
outlines = outlines[:firstheading] + toc + outlines[firstheading:]

for line in outlines:
    print(line)
    pass