db/parsers/bnss/section_parser.py

from __future__ import annotations

import re
from typing import List, Dict
from db.parsers.bnss.cleaner import BNSSTextCleaner


class SectionParser:
    """
    BNS Section Parser

    Supports:

        1.
        2.
        103.
        356.

    Returns:

        section_no
        section_title
        text
        start
        end
    """

    # =====================================================
    # SECTION HEADER
    # =====================================================

    SECTION_RE = re.compile(
        r"(?m)^(\d{1,3})\.\s+(.+)$"
    )
    # =====================================================
    # VALID SECTION NUMBER
    # =====================================================

    SECTION_NO_RE = re.compile(
        r"^\d+$"
    )

    # =====================================================
    # CLEAN TITLE
    # =====================================================

    @staticmethod
    def clean_title(
        title: str
    ) -> str:

        title = title.strip()

        title = re.sub(
            r"\s+",
            " ",
            title
        )

        title = title.strip(
            ".-— "
        )

        return title

    # =====================================================
    # RECOVER TITLE
    # =====================================================

    def recover_title(
        self,
        section_text: str,
        current_title: str
    ) -> str:

        if current_title:

            return self.clean_title(
                current_title
            )

        lines = [
            x.strip()
            for x in section_text.split(
                "\n"
            )
            if x.strip()
        ]

        if len(lines) < 2:
            return ""

        return self.clean_title(
            lines[1]
        )
    
    @staticmethod
    def remove_statement_of_objects(text):

        marker = (
            "The Commission suggested various enactments"
        )

        idx = text.find(marker)

        if idx != -1:
            text = text[:idx]

        return text

    # =====================================================
    # EXTRACT SECTIONS
    # =====================================================

    def extract_sections(
        self,
        text: str
    ) -> List[Dict]:
        

        text = self.remove_statement_of_objects(text)

        matches = list(
            self.SECTION_RE.finditer(
                text
            )
        )




        if not matches:
            return []

        sections = []

        for i, match in enumerate(
            matches
        ):

            start = match.start()

            end = (
                matches[i + 1].start()
                if i + 1 < len(matches)
                else len(text)
            )

            section_no = (
                match.group(1)
                .strip()
            )

            section_title = (
                match.group(2)
                .strip()
            )

            section_text = (
                text[start:end]
                .strip()
            )

            section_title = (
                self.recover_title(
                    section_text,
                    section_title
                )
            )

            sections.append(
                {
                    "section_no":
                        section_no,

                    "section_title":
                        section_title,

                    "text":
                        section_text,

                    "start":
                        start,

                    "end":
                        end
                }
            )

        return sections

    # =====================================================
    # VALIDATION
    # =====================================================

    def validate_sections(
        self,
        sections: List[Dict]
    ) -> List[str]:

        errors = []

        seen = set()

        for section in sections:

            section_no = (
                section["section_no"]
            )

            if section_no in seen:

                errors.append(
                    f"Duplicate Section "
                    f"{section_no}"
                )

            seen.add(
                section_no
            )

            if not self.SECTION_NO_RE.match(
                section_no
            ):

                errors.append(
                    f"Invalid Section "
                    f"{section_no}"
                )

            if not section[
                "section_title"
            ]:

                errors.append(
                    f"Missing title "
                    f"for Section "
                    f"{section_no}"
                )

        return errors

    # =====================================================
    # FIND SECTION
    # =====================================================

    def get_section(
        self,
        sections: List[Dict],
        section_no: str
    ):

        for section in sections:

            if (
                section["section_no"]
                == section_no
            ):
                return section

        return None


# =========================================================
# EXAMPLE
# =========================================================

if __name__ == "__main__":

    with open(
        "../../pdfs/bns.txt",
        "r",
        encoding="utf8"
    ) as f:

        text = f.read()

    cleaner=BNSSTextCleaner()
    text=cleaner.clean(text)

 

    


    parser = SectionParser()

    sections = (
        parser.extract_sections(
            text
        )
    )


    




    print(
        f"Sections Found: "
        f"{len(sections)}"
    )

    print()

    for section in sections[:20]:

        print(
            f"{section['section_no']} "
            f"-> "
            f"{section['section_title']}"
        )

    print()

    errors = (
        parser.validate_sections(
            sections
        )
    )

    print(
        f"Errors: "
        f"{len(errors)}"
    )

    for e in errors:
        print(e)