lint_source

#!/usr/bin/env python3

"""
Performs a 'linting' of the assembly, ensuring that formatting errors do not make it into
the final output.
"""

import getopt
import re
import sys

# The input file to be linted.
INPUT_FILE = None

# Whether to abort the linting process on encountering any error.
ABORT_ON_ERROR = False
# The maxmimum length of any line.
MAX_LENGTH_LINE = 80
# The maximum length of a label.
MAX_LENGTH_IDENTIFIER = 36
# The prefix for local labels.
LOCAL_LABEL_PREFIX = "."

def lint_line(line):
    """
    Processes an individual line of source code, checking for formatting errors.
    """

    # Check the length of a comment line.
    REGEX_LINE_COMMENT = r"^;.*"
    if re.match(REGEX_LINE_COMMENT, line):
        if len(line.rstrip()) > MAX_LENGTH_LINE:
            return (False, 'Comment line too long')

    # Check the total length of the line.
    if len(line.rstrip()) > MAX_LENGTH_LINE:
        return (False, 'Line too long')

    # Check the maximum length of each label.
    REGEX_LABEL = r"([.\w?]+):"
    label_match = re.match(REGEX_LABEL, line)
    if label_match:
        if len(label_match.group(1)) > MAX_LENGTH_IDENTIFIER:
            return (False, f"Label '{label_match.group(1)}' too long")

    return (True, None)


def check_for_duplicate_labels(file_text):
    """
    Search the source text for duplicate label identifiers.
    """

    # A map of duplicate labels -> The number of occurences.
    label_count = {}
    # The regex to use when searching for all labels.
    REGEX_LABEL = r"^({}?[\w?]+:)".format(LOCAL_LABEL_PREFIX)

    # Iterate over all labels, adding each to the label count map.
    all_labels = re.findall(REGEX_LABEL, file_text, flags=re.MULTILINE)
    for label_name in all_labels:
        if not label_name in label_count:
            label_count[label_name] = 1
        else:
            label_count[label_name] = label_count[label_name] + 1

    return label_count


def check_for_unnecessary_labels(file_text):
    """
    Search the source text for labels that are defined, but not used.
    """

    unnecessary_labels = []
    # The regex to use when searching for all labels.
    REGEX_LABEL = r"^({}?[\w?]+):".format(LOCAL_LABEL_PREFIX)

    # Iterate over all labels, searching for labels that are only found once.
    all_labels = re.findall(REGEX_LABEL, file_text, flags=re.MULTILINE)
    for label_name in all_labels:
        start = 0
        count = 0
        while start := file_text.find(label_name, start) + 1:
            count += 1

        if count == 1:
            unnecessary_labels.append(label_name)

    return unnecessary_labels


def lint_file():
    """
    Lints the input file.
    """

    try:
        with open(INPUT_FILE, 'r') as input_file:
            # The current line number.
            line_number = 1
            # The total number of errors encountered.
            error_count = 0

            source_lines = input_file.readlines()
            for source_line in source_lines:
                (lint_result, lint_error) = lint_line(source_line)
                if not lint_result:
                    print(f"Error: '{lint_error}' on line {line_number}", file=sys.stderr)
                    error_count = error_count + 1

                    if ABORT_ON_ERROR:
                        exit(1)

                line_number = line_number + 1

            # 'Rewind' the file stream, and check for duplicate identifiers.
            input_file.seek(0)
            source_text = input_file.read()
            label_count = check_for_duplicate_labels(source_text)

            for label_name in label_count:
                if label_count[label_name] > 1:
                    error_count = error_count + 1
                    print(
                        f"Error: Duplicate label: '{label_name}' - ",
                        f"Encountered {label_count[label_name]} times."
                    )

            unnecessary_labels = check_for_unnecessary_labels(source_text)
            for label_name in unnecessary_labels:
                error_count = error_count + 1
                print(f"Error: Unnecessary label: '{label_name}'")

            print(f"Found {error_count} errors in total.")

    except FileNotFoundError:
        print("Error: Unable to open assembler files. Exiting.", file=sys.stderr)
        exit(1)


def print_usage():
    """Prints script usage to STDOUT."""

    print("Usage: lint_source --input_file <string>")


if __name__ == "__main__":
    try:
        OPTS, ARGS = getopt.getopt(
            sys.argv[1:],
            "h",
            [
                "input_file=",
            ]
        )
    except getopt.GetoptError:
        print_usage()
        sys.exit(2)
    for opt, arg in OPTS:
        if opt == '-h':
            print_usage()
            sys.exit()
        elif opt == "--input_file":
            INPUT_FILE = arg

    if INPUT_FILE is None:
        print("No input file provided! Exiting.")
        exit(1)

    lint_file()