"""
* Version: 0.1.0.0.0

"""

import sys, pathlib, re
import pingcap.pe.flash.util as util
import pingcap.pe.flash.ai_editor as editor
import pingcap.pe.flash.util_srt as util_srt


if __name__ == "__main__":

    input_srt_file_path = sys.argv[1]

    # Stage 1
    parsed_srt_file_path = sys.argv[1].split(".")[0] + "_parsed_srt_to_paragraph.md"
    print(util.speak_title("Parsing SRT content to a Markdown file ..."))

    fr = open(input_srt_file_path, "r")
    lines = fr.read()
    fr.close()
    f = open(parsed_srt_file_path, "w")
    cache = util_srt.process_srt_to_paragraph(lines)
    f.write(cache["paragraph"])
    f.close()

    # Refine the original SRT
    refined_srt_file_path = parsed_srt_file_path.replace(
        "_parsed_srt_to_paragraph.md", "_refined.srt"
    )
    wr = open(refined_srt_file_path, "w")
    append = False
    first = True
    for line in lines.split("\n"):
        if re.match("^[0-9]*$", line):
            append = False
            if first:
                wr.write(line + "\n")
                first = False
            else:
                wr.write("\n" + line + "\n")
        elif re.match("^[0-9][0-9]:[0-9][0-9]:[0-9][0-9],[0-9][0-9][0-9]", line):
            wr.write(line + "\n")
        elif re.match("^$", line):
            wr.write("\n\n")
        else:
            if not append:
                wr.write(line.replace("\xa0", ""))
                append = True
            else:
                wr.write(" " + line.replace("\xa0", ""))
    wr.close()
    print(
        util.speak_result("parsed_srt_to_paragraph_file_path: " + parsed_srt_file_path)
    )

    # Stage 2
    sentences_file_path = sys.argv[1].split(".")[0] + "_parsed_srt_to_sentences.md"
    print(util.speak_title("Parsing paragraph to sentences to a Markdown file ..."))

    f = open(sentences_file_path, "w")
    cache = util_srt.split_en_paragraph_to_sentences(cache)
    sentences = []
    for k in cache.keys():
        if k.startswith("S-"):
            sentences += [cache[k]]
    f.write("\n\n".join(sentences))
    f.close()

    # Stage 3
    timestamp_sentences_file_path = (
        sys.argv[1].split(".")[0] + "_parsed_srt_to_timestamp_sentences.md"
    )
    print(
        util.speak_title(
            "Parsing paragraph to timestamp sentences to a Markdown file ..."
        )
    )

    f = open(timestamp_sentences_file_path, "w")
    timestamp_sentences = []
    for k in cache.keys():
        if k.startswith("TS-"):
            timestamp_sentences += [cache[k]]
    f.write("\n\n".join(timestamp_sentences))
    f.close()

    # Stage 4
    ja_sentences_file_path = (
        sys.argv[1].split(".")[0] + "_translated_srt_en2ja_sentences.md"
    )
    print(util.speak_title("Translating sentences to a Markdown file ..."))

    cache = util_srt.translate_sentence_en2ja(cache)

    f = open(ja_sentences_file_path, "w")
    ja_sentences = []
    for k in cache.keys():
        if k.startswith("JS-"):
            ja_sentences += [cache[k]]
    f.write("\n\n".join(ja_sentences))
    f.close()

    # Stage 5
    enja_sentences_file_path = (
        sys.argv[1].split(".")[0] + "_workbench_srt_en2ja_sentences.md"
    )
    print(util.speak_title("Generating workbench file to a Markdown file ..."))

    f = open(enja_sentences_file_path, "w")
    enja_sentences = []
    for k in cache.keys():
        if k.startswith("EJS-"):
            enja_sentences += [cache[k]]
    f.write("\n\n".join(enja_sentences))
    f.close()

    print(util.speak_title("sentences"))
    for k in cache.keys():
        if k.startswith("S-"):
            print(cache[k])
    print(util.speak_title("ja sentences"))
    for k in cache.keys():
        if k.startswith("JS-"):
            print(cache[k])
    print(util.speak_title("time-sentences"))
    for k in cache.keys():
        if k.startswith("TS-"):
            print(cache[k])
    print(util.speak_title("enja sentences"))
    for k in cache.keys():
        if k.startswith("EJS-"):
            print(cache[k])

    print(util.speak_result("workbench_file_path: " + enja_sentences_file_path))