header.py

import re
import linetypes

class Header:
    def __init__(self):
        self.description = None
        self.name = None
        self.author = None
        self.username = None
        self.filetype = None
        self.qualifiers = None
        self.license = None
        self.help = []
        self.bfc = None
        self.category = None
        self.keywords = []
        self.cmdline = None
        self.history = []

class BadHeader:
    def __init__(self, index, reason):
        self.index = index
        self.reason = reason
    def __repr__(self):
        return str.format(
            'header.BadHeader(index = {index!r}, reason = {reason!r})',
            index = self.index,
            reason = self.reason,
        )

geometrical_types = [
    linetypes.LineSegment,
    linetypes.Triangle,
    linetypes.Quadrilateral,
    linetypes.ConditionalLine,
]

def is_suitable_header_object(entry):
    return not any(
        isinstance(entry, linetype)
        for linetype in [
            *geometrical_types,
            linetypes.Comment,
            linetypes.Error,
        ]
    )

class HeaderError(Exception):
    def __init__(self, index, reason):
        self.index, self.reason = index, reason
    def __repr__(self):
        return str.format(
            'HeaderError({index!r}, {reason!r})',
            index = self.index,
            reason = self.reason,
        )
    def __str__(self):
        return reason

class HeaderParser:
    def __init__(self):
        self.model_body = None
        self.cursor = 0
        self.problems = []
    def parse(self, model_body):
        result = Header()
        self.order = []
        self.cursor = -1
        self.model_body = model_body
        self.skip_to_next()
        result.description = self.current()
        self.skip_to_next()
        result.name = self.parse_pattern('^Name: (.+)$', 'name')[0]
        self.skip_to_next()
        result.author = self.parse_pattern('^Author: (.+)$', 'author')[0]
        for header_entry in self.get_more_header_stuff():
            if self.try_to_match(
                '^!LDRAW_ORG ' + 
                '(' \
                    '(?:Unofficial_)?' \
                    'Part|' \
                    'Subpart|' \
                    'Primitive|' \
                    '8_Primitive|' \
                    '48_Primitive|' \
                    'Shortcut' \
                ')\s?' \
                '(ORIGINAL|UPDATE \d\d\d\d-\d\d)?$',
                'part type'):
                result.filetype, result.qualifiers = self.groups
            elif self.try_to_match(
                '^!LICENSE (.+)$',
                'license'):
                result.license = self.groups()
            else:
                self.parse_error("couldn't handle header metacommand: " + repr(header_entry.text))
        return {
            'header': result,
            'end-index': self.cursor + 1,
        }
    def parse_error(self, message):
        raise HeaderError(index = self.cursor, reason = message)
    def get_more_header_stuff(self):
        while True:
            self.cursor += 1
            if self.cursor >= len(self.model_body):
                break
            entry = self.model_body[self.cursor]
            if not is_suitable_header_object(entry):
                break
            if isinstance(entry, linetypes.MetaCommand):
                yield entry
    def skip_to_next(self, *, spaces_expected = 0):
        while True:
            if self.cursor + 1 >= len(self.model_body):
                self.parse_error('stub ldraw file')
            self.cursor += 1
            entry = self.model_body[self.cursor]
            if not is_suitable_header_object(entry):
                self.parse_error('header is incomplete')
            if isinstance(entry, linetypes.MetaCommand):
                return
    def try_to_match(self, pattern, patterntype):
        try:
            self.groups = self.parse_pattern(pattern, patterntype)
        except:
            return False
    def current(self):
        entry = self.model_body[self.cursor]
        assert isinstance(entry, linetypes.MetaCommand)
        return entry.text
    def parse_pattern(self, pattern, description):
        match = re.search(pattern, self.current())
        if match:
            self.order.append(description)
            return match.groups()
        else:
            self.parse_error(str.format("couldn't parse {}", description))