nagrest/src/nagparse.py

#!/usr/bin/env python3
import re
from enum import Enum, auto
from nag_object import NagObject, NagObjectType

class ParserState(Enum):
    COMMENT = auto()
    DEFINE = auto()
    DIRECTIVE = auto()
    MULTILINE = auto()


class NagParse:
    def __init__(self, config_files: list[str]):
        self.m_config_files: list[str] = config_files
        self.m_objects: list[NagObject] = list()
        for config_file in self.m_config_files: 
            for object in self.parse_config(config_file):
                self.m_objects.append(object)
    def __str__(self):
        string = str()
        for object in self.m_objects:
            string += str(object) + "\n"
        return string

    def parse_config(self,config_file: str) -> list[NagObject]:
        state = ParserState.DEFINE
        objects: list[NagObject] = list()
        with open(config_file, 'r') as file_handle:
            line_no = 0
            cur_object = NagObject(config_file=config_file)
            for rawline in file_handle.readlines():
                # Clean up the config
                line = re.sub(r'([a-z]){', r'\1 {',rawline.strip())
                #line = re.sub(r'([a-zA-Z0-9])}', r'\1\n}',line.strip())
                #line = re.sub(r'([a-zA-Z0-9]) }', r'\1\n}',line.strip())
                line_no += 1
                # This means it is a comment, we might want to save those later on
                if len(line) == 0 or re.match('^[\t ]*[#]+', line):
                    continue
                # This means we are done parsing and should reset for new object
                if re.match(r'}', line):
                    objects.append(cur_object)
                    del cur_object
                    state = ParserState.DEFINE
                    continue
                #if we get here it is not a comment and not the end of an object 
                fragments: list[str] = re.sub(' +', ' ', line.strip()).split()

                # This is a new object and we should see what type it is, and what line number it is on
                if state == ParserState.DEFINE:
                    if fragments[0] == 'define':
                        cur_object = NagObject(config_file=config_file, directives=list())
                        cur_object.m_line_no = line_no
                        cur_object.m_type = NagObjectType(fragments[1])
                        # For the next line we should be in DIRECTIVE state
                        state = ParserState.DIRECTIVE
                    else:
                        print("It's fucked up man")
                        break
                # This is the DIRECTIVE state, where we want to collect key, values and find out if we should go to multiline mode
                elif state == ParserState.DIRECTIVE:
                    if fragments[-1][-1] == '\\':
                        state = ParserState.MULTILINE
                    if re.match('^[a-z]+', fragments[0]):
                        cur_object.m_directives.append(
                            [fragments[0], " ".join(fragments[1:])])
                # This is the MULTILINE state, directives can be on many lines if the '\n' is escaped with '\'
                elif state == ParserState.MULTILINE:
                    cur_object.m_directives[-1][1] = cur_object.m_directives[-1][1].strip('\\') + ' ' + " ".join(fragments) 
                    if fragments[-1][-1] != '\\':
                        state = ParserState.DIRECTIVE
                else:
                    print("It's fucked up man")
                    break
        return objects