Transform MacJournal individual text entries to Zim

macjournal2zim.py

from __future__ import print_function
import locale
import datetime
import re
import os
from os.path import isfile, join
import unicodedata
from unidecode import unidecode


# Configuration
loc = locale.setlocale(locale.LC_TIME, ("es_ES.utf8", "es_ES.utf8"))


def parse_date(input_line):
    """
    Parse the date in the MacJournal format
    """
    date_numbers = list(re.findall("(\d{1,2}) de (.*?) de (\d{2,4}), (\d{1,2}):(\d{1,2})", input_line)[0])
    date_numbers[0] = "{:02d}".format(int(date_numbers[0])) # Day
    date_numbers[3] = "{:02d}".format(int(date_numbers[3])) # Hour
    date_re = "{} {} {} {}:{}".format(*date_numbers)
    return datetime.datetime.strptime(date_re, "%d %B %Y %H:%M")

def parse_file(input_file):
    """
    Parse a MacJournal txt single entry file
    """
    with open(input_file, "r") as f:
        line1 = f.readline().split("\t")
        assert line1[1] == "Fecha:", "Wrong format of line 1"
        date_entry = parse_date(line1[2])
        
        line2 = f.readline()
        if "Tema:" in line2:
            title_entry = line2.split("\t")[2].strip()
        else:
            title_entry = input_file.split("/")[-1].strip().replace(".txt", "")

        content = [line for line in f]
    return title_entry, content, date_entry

def format_zim(title, content, date, creation_date=False):
    """
    Generate a Zim-like list of lines
    """
    out = []
    out.append("Content-Type: text/x-zim-wiki\n")
    out.append("Wiki-Format: zim 0.4\n")
    out.append("Creation-Date: "+date.isoformat()+"+00:00\n\n")
    out.append("====== {} ======\n".format(title))
    if creation_date:
        out.append("Creado "+date.isoformat(" ")+"\n")
    out.extend(content)
    return out


def transform_file(input_file, output_file, creation_date=True):
    """
    Enter a MacJournal file and output a Zim file
    """
    title_entry, content, date_entry = parse_file(input_file)
    out = format_zim(title_entry, content, date_entry, creation_date=creation_date)
    with open(output_file, "w") as f:
        for line in out:
            f.write(line)


if __name__ == "__main__":
    output_dir = "MacJournal"
    dirs = ["Blog", "LOFAR", "Personal"]
    for d in dirs:
        path = join("text_1", d)
        entries = [f for f in os.listdir(path) if isfile(join(path, f)) and f.endswith("txt")]
        if not os.path.exists(join(output_dir, d)):
            os.mkdir(join(output_dir, d))
        for entry in entries:
            #name = unicodedata.normalize('NFKC', entry.replace(" ", "_"))
            name = unidecode(entry.replace(" ", "_"))
            print(entry, name)
            
            input_file = join("text_1", d, entry)
            output_file = join(output_dir, d, name)
            
            transform_file(input_file, output_file)