convert_capek.py

import os
import re
from typing import Dict

import jsonlines
from tqdm import tqdm

TARGET=".data/capek.vert"


def process_vert_format(vert_content: str) -> Dict[str, str]:
    # Pattern to match document boundaries and extract metadata
    doc_pattern = re.compile(r'<doc[^>]*>.*?</doc>', re.DOTALL)
    metadata_pattern = re.compile(
        r'<doc id="([^"]*)".+?autor="([^"]*)".+?nazev="([^"]*)".+?genre="([^"]*)"')
    sent_pattern = re.compile(r'<s[^>]*id="([^"]*)".*?>(.*?)</s>', re.DOTALL)

    # Pattern to remove whitespace before punctuation
    ws_before_punct = re.compile(r'\s+([.,!?:;])')

    # Find all documents
    documents = re.findall(doc_pattern, vert_content)
    processed_documents = {}

    for doc in tqdm(documents):
        # Extract metadata
        metadata_match = re.search(metadata_pattern, doc)
        if metadata_match:
            # r'<doc id="([^"]*)".+?rok="([^"]*)".+?misto="([^"]*)" sidlotyp="([^"]*)".+?tema="([^"]*)" pocetml="([^"]*)"')

            doc_id = metadata_match.group(1)
            author = metadata_match.group(2)
            title = metadata_match.group(3)
            genre = metadata_match.group(4)

            metadata_str = (f"Autor: {author}, "
                            f"Název: {title}, "
                            f"Žánr: {genre}")
        else:
            raise ValueError("Metadata not found in document")

        # Initialize an empty list to hold processed document text
        processed_document = [metadata_str]

        # Find all speaker turns within the document
        for sent_match in re.findall(sent_pattern, doc):
            sp_content = sent_match[1]

            # remove tags from each line, and join text
            tokens = [line.split("\t")[0].strip() for line in sp_content.split("\n") if line.strip() != ""]
            speaker_text = " ".join(tokens)

            # replace more than one space with one space
            speaker_text = re.sub(r'\s+', ' ', speaker_text).strip()

            # remove whitespace before ., !, ?
            speaker_text = re.sub(ws_before_punct, r'\1', speaker_text)

            # - sometimes lines in oral are empty? e.g. 08A009N // REMOVE THESE LINES
            if speaker_text.strip() == "":
                continue
            processed_document.append(speaker_text)

        final_text = '\n'.join(processed_document)
        processed_documents[doc_id] = final_text

    return processed_documents


# Read the content from the file
with open(TARGET, "r") as f:
    vert_content = f.read()

# Process the content
processed_documents = process_vert_format(vert_content)

# write all splits into same json file in  .data/hf_dataset/cnc_fictree/test.jsonl
OF = ".data/hf_dataset/cnc_capek/test.jsonl"
os.makedirs(os.path.dirname(OF), exist_ok=True)
with jsonlines.open(OF, "w") as writer:
    for doc_id, doc in list(processed_documents.items()):
        writer.write({"text": doc, "id": doc_id})