nalayiram_helper.py

import json
from dataclasses import dataclass
from collections import defaultdict
import logging

logging.basicConfig()
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)


@dataclass
class Pasuram:
    prabandham_code: str
    azhwar_name: str
    prabandham_name: str


def get_standardized_prabandham_names() -> list[Pasuram]:
    """
    Get a list of prabandham names along with the azhwars who authored them in divya_prabandham,
    sorted by the prabandham name (3rd field, index 2).
    """
    with open("./data/azhwars.json", "r", encoding="utf-8") as f:
        azhwars = json.load(f)
        header = azhwars[0]
        rows = azhwars[1:]

        # Sort by 3rd field (index 2)
        rows.sort(key=lambda row: row[2])

        final_azhwars = [Pasuram(**dict(zip(header, row))) for row in rows]

    return final_azhwars


def get_standardized_azhwar_names() -> list[str]:
    """
    Get a list of azhwar names along with the pasurams they have authored in divya_prabandham
    """
    with open("./data/azhwars.json", "r", encoding="utf-8") as f:
        azhwars = json.load(f)
        header = azhwars[0]
        rows = azhwars[1:]
        final_azhwars = [row[1] for row in rows]  ## 2nd field is the azhwar name

    return sorted(set(final_azhwars))


def get_standardized_divya_desam_names() -> list[str]:
    """
    Get a list of divya desam names in divya_prabandham
    """
    with open("./data/divya_desams.json", "r", encoding="utf-8") as f:
        divya_desams = json.load(f)  # FIXED

    selected_fields = [
        "title",
        "other_names",
        "name_ta",
        "alwars",
        "area",
        "state",
        "thirukolam",
        "direction",
        "sampradayam",
        "divya_desam",
    ]
    data = [
        {key: row[key] for key in selected_fields if key in row}
        for row in divya_desams["pageProps"]["hits"]
    ]
    return sorted(set([row["title"] for row in data]))

def reorder_taniyan(collection):
    logger.info("reorder_taniyan: started")

    # Fetch all docs with ids + metadatas
    data = collection.get(include=["metadatas"])
    ids = data.get("ids", [])
    metas = data.get("metadatas", [])

    if not ids or not metas:
        logger.warning("reorder_taniyan: no data found in collection")
        return

    # sort globally by current _global_index
    records = sorted(
        [(i, m) for i, m in enumerate(metas)],
        key=lambda x: x[1].get("_global_index", float("inf")),
    )

    # group by prabandham_code
    grouped = defaultdict(list)
    for i, meta in records:
        prabandham = meta.get("prabandham_code")
        if prabandham:
            grouped[prabandham].append((i, meta))

    updates = []
    global_counter = 1  # running _global_index across the collection

    for prabandham, items in grouped.items():
        taniyan_items = [
            (i, m) for i, m in items if m.get("section_type", "").startswith("taniyan")
        ]
        non_taniyan_items = [
            (i, m)
            for i, m in items
            if not m.get("section_type", "").startswith("taniyan")
        ]

        if not taniyan_items and not non_taniyan_items:
            continue

        # sort both groups by original _global_index
        taniyan_items.sort(key=lambda x: x[1]["_global_index"])
        non_taniyan_items.sort(key=lambda x: x[1]["_global_index"])

        # --- taniyans first (verse starts from 1) ---
        for verse_no, (i, meta) in enumerate(taniyan_items, start=1):
            updates.append(
                {
                    "id": ids[i],
                    "metadata": {
                        **meta,
                        "_global_index": global_counter,
                        "verse": verse_no,
                    },
                }
            )
            global_counter += 1

        # --- non-taniyans continue from their base verse ---
        if non_taniyan_items:
            base_verse = min(m["verse"] for _, m in non_taniyan_items)
            for offset, (i, meta) in enumerate(non_taniyan_items):
                updates.append(
                    {
                        "id": ids[i],
                        "metadata": {
                            **meta,
                            "_global_index": global_counter,
                            "verse": base_verse + offset,
                        },
                    }
                )
                global_counter += 1

    if updates:
        logger.info("reorder_taniyan: updating %d records...", len(updates))
        collection.update(
            ids=[u["id"] for u in updates],
            metadatas=[u["metadata"] for u in updates],
        )
        logger.info("reorder_taniyan: update complete.")
    else:
        logger.info("reorder_taniyan: nothing to update")

    logger.info("reorder_taniyan: finished")


def delete_taniyan(collection):
    logger.info("delete_taniyan: started")        

    # Fetch all docs (only ids + metadata needed)
    data = collection.get(include=["metadatas"])
    ids = data["ids"]
    metas = data["metadatas"]

    # Collect ids where section_type starts with "taniyan"
    taniyan_ids = [
        ids[i] for i, meta in enumerate(metas)
        if meta.get("section_type", "").startswith("taniyan")
    ]

    if taniyan_ids:
        logger.info("delete_taniyan: Deleting %d taniyan records...", len(taniyan_ids))
        collection.delete(ids=taniyan_ids)
        logger.info("delete_taniyan: Deleted %d taniyan records", len(taniyan_ids))
    else:
        logger.info("delete_taniyan: No taniyan records found")

    logger.info("delete_taniyan: finished")


if __name__ == "__main__":
    logger.info(get_standardized_azhwar_names())