Copyright for manpages source package in Stretch

#!/usr/bin/python3
#
# Copyright © 2016 Dr. Tobias Quathamer 
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see .

import re
import sys
import textwrap
from pathlib import Path

# The standard short names in Debian are defined here:
# https://www.debian.org/doc/packaging-manuals/copyright-format/1.0/

license_information = [
	{
		'shortname': 'BSD-3-clause',
		'filename':  'BSD-3-clause-UCB',
		'upstream_names': ['BSD_3_CLAUSE_UCB'],
	},
	{
		'shortname': 'BSD-4-clause',
		'filename':  'BSD-4-clause-UCB',
		'upstream_names': ['BSD_4_CLAUSE_UCB', 'BSD_ONELINE_CDROM'],
	},
	{
		'shortname': 'Expat',
		'filename':  'Expat',
		'upstream_names': ['PERMISSIVE_MISC', 'MIT'],
	},
	{
		# This shortname is not defined by the standard.
		'shortname': 'freely-redistributable',
		'filename':  'freely-redistributable',
		'upstream_names': ['FREELY_REDISTRIBUTABLE'],
	},
	{
		'shortname': 'GPL-2',
		'filename':  'GPL-2',
		'upstream_names': ['GPLv2_MISC', 'GPLv2_ONELINE'],
	},
	{
		'shortname': 'GPL-2+',
		'filename':  'GPL-2+',
		'upstream_names': ['GPL_NOVERSION_ONELINE', 'GPLv2+',
		                   'GPLv2+_DOC_FULL', 'GPLv2+_DOC_MISC',
		                   'GPLv2+_DOC_ONEPARA', 'GPLv2+_SW_3_PARA',
		                   'GPLv2+_SW_ONEPARA'],
	},
	{
		# This shortname is not defined by the standard.
		'shortname': 'henry-spencer-regex',
		'filename':  'henry-spencer-regex',
		'upstream_names': ['MISC'],
	},
	{
		# This shortname is not defined by the standard.
		'shortname': 'LDPv1',
		'filename':  'LDPv1',
		'upstream_names': ['LDPv1'],
	},
	{
		'shortname': 'public-domain',
		'filename':  'public-domain',
		'upstream_names': ['PUBLIC_DOMAIN'],
	},
	{
		# This shortname is not defined by the standard.
		'shortname': 'verbatim',
		'filename':  'verbatim',
		'upstream_names': ['VERBATIM', 'VERBATIM_ONE_PARA',
		                   'VERBATIM_TWO_PARA', 'VERBATIM_PROF'],
	},
]

licenses_with_manpages = {}
symlinks = {}

def get_license_shortname(name):
	"""Gets the Debian shortname for the name supplied by upstream.
	
	Includes a check that upstream's names are only assigned
	to one shortname.
	"""
	shortname = ""
	already_found = False
	for info in license_information:
		if name in info['upstream_names']:
			if not already_found:
				shortname = info['shortname']
				already_found = True
			else:
				sys.exit("Fatal error: Upstream license name defined multiple times: " + name)
	return shortname

def add_manpage_to_shortname(manpage, copyright_holders, licenses):
	# Ensure a string for the filename
	filename = str(manpage)
	# Strip the leading "../"
	filename = filename[3:len(filename)]
	# Common case: only one license for the manpage, so
	# the shortname is just e.g. "GPL-2+"
	shortname = " and ".join(sorted(licenses))
	if shortname not in licenses_with_manpages:
		licenses_with_manpages[shortname] = {'files': [filename], 'copyright': copyright_holders}
	else:
		licenses_with_manpages[shortname]['files'].append(filename)
		# Do not add same lines twice
		existing_copyright_holders = licenses_with_manpages[shortname]['copyright']
		joined_copyright_holders = list(set(existing_copyright_holders + copyright_holders))
		licenses_with_manpages[shortname]['copyright'] = joined_copyright_holders

def get_copyright_stanza(shortname, file_info):
	stanza = ""
	# Collect files and symlinks into a common list
	all_files = file_info['files']
	for file in file_info['files']:
		if file in symlinks:
			all_files = all_files + symlinks[file]
	# Join the files into a whitespace separated list,
	# at most 76 characters long
	files = " ".join(sorted(all_files))
	# The wrap is 69 + 7 (length of "Files: ") = 76
	files = textwrap.wrap(files, width=69, break_long_words=False, break_on_hyphens=False)
	files = "\n       ".join(files)
	# Now format the copyright holders
	copyright = "\n           ".join(sorted(file_info['copyright']))
	# An empty field is an error, so ensure a value
	if len(copyright) == 0:
		copyright = "(could not be detected automatically)"
	# Finally, create the stanza
	stanza += "Files: " + files
	stanza += "\nCopyright: " + copyright
	stanza += "\nLicense: " + shortname + "\n\n"
	return stanza

def get_license_text(shortname):
	"""Gets the text for the Debian license shortname."""
	text = ""
	for info in license_information:
		if info['shortname'] == shortname:
			text += "License: " + shortname + "\n"
			with open("licenses/" + info['filename']) as licensefile:
				for line in licensefile:
					text += " " + line
			return text

p = Path("..")
for manpage in p.glob("man?/*"):
	with manpage.open() as file:
		licenses = []
		copyright_holders = []
		manpage_is_symlink = False
		for line in file:
			# Do not create copyright stanzas for symlink files
			# but add them to a symlink list
			symlink = re.search(r"^\.so (.*)", line)
			if symlink:
				manpage_is_symlink = True
				# Ensure a string for the filename
				linkname = str(manpage)
				# Strip the leading "../"
				linkname = linkname[3:len(linkname)]
				filename = symlink.group(1)
				if filename in symlinks:
					symlinks[filename].append(linkname)
				else:
					symlinks[filename] = [linkname]
				break
			# Only parse the header, so stop after seeing ".TH"
			if re.search(r"^\.TH", line):
				break
			# Extract all copyright holders
			copyright = re.search(r"^\.\\\".*?Copyright (.*)", line)
			if copyright:
				copyright_holders.append(copyright.group(1))
			# Match the beginning of the license
			license_start = re.search(r"^\.\\\" %%%LICENSE_START\((.*)\)", line)
			if license_start:
				license_name = license_start.group(1)
				license_short_name = get_license_shortname(license_name)
				if not license_short_name:
					sys.exit("Fatal error: Upstream license name not known: " + license_name)
				licenses.append(license_short_name)
		if not manpage_is_symlink:
			add_manpage_to_shortname(manpage, copyright_holders, licenses)

# Flatten the symlinks by detecting and removing
# a symlink which points to another symlink.
for link_to_test in symlinks:
	# Now cycle through all symlink entries
	for link in symlinks:
		if link_to_test in symlinks[link]:
			symlinks[link] = symlinks[link] + symlinks[link_to_test]

# Make sorting of licenses deterministic
stanzas = ""
license_texts = []
for shortname in sorted(licenses_with_manpages):
	stanzas += get_copyright_stanza(shortname, licenses_with_manpages[shortname])
	text = get_license_text(shortname)
	if text:
		license_texts.append(text)

# Read in the first lines of copyright, without
# the automatically generated parts. Stop after
# the third occurence of "License".
manual_lines = ""
license_line_count = 0
with open("copyright") as copyright_file:
	for line in copyright_file:
		manual_lines += line
		if re.search(r"^License:", line):
			license_line_count += 1
			if license_line_count == 3:
				# Add a final newline for separation
				manual_lines += "\n"
				break

# Open the file for output
with open("copyright", "w") as copyright_file:
	copyright_file.write(manual_lines)
	copyright_file.write(stanzas)
	copyright_file.write("\n".join(license_texts))