import datetime
import frontmatter
import os
import re
import unicodedata


input_dir = "content"


def slugify(string):
    return re.sub(
        r"\W+",
        "",
        unicodedata.normalize("NFKD", string)
        .encode("ascii", "ignore")
        .decode("ascii")
        .lower(),
    )


def parse_entry_data(entry_data):
    entry_data["date"] = datetime.datetime.fromisoformat(entry_data["date"] + "+00:00")
    if "description" in entry_data:
        entry_data["description"] = entry_data["description"].strip()
    if "links" in entry_data:
        for link in entry_data["links"]:
            link["title"] = link["title"].strip()
            if "description" in link:
                if link["description"].strip() != "":
                    link["description"] = link["description"].strip()
                else:
                    del link["description"]
            clean_link = dict((k, link[k]) for k in link)
    if "modified" in entry_data:
        entry_data["modified"] = datetime.datetime.fromisoformat(
            entry_data["modified"] + "+00:00"
        )
    if "tags" in entry_data:
        entry_data["tags"] = [tag.strip() for tag in entry_data["tags"].split(",")]
    if "series" in entry_data:
        entry_data["series"] = entry_data["series"].strip()
    entry_data["title"] = entry_data["title"].strip()
    if os.path.isdir(f"{input_dir}/{entry_data['id']}"):
        entry_data["dir"] = f"{input_dir}/{entry_data['id']}"


def _entries():
    entries = {}

    for filename in os.listdir(input_dir):
        if filename.endswith(".md") or filename.endswith(".gmi"):
            entry_data = frontmatter.load(f"{input_dir}/{filename}")
            entry_data["id"], _ = map(slugify, os.path.splitext(filename))
            if "date" in entry_data:
                parse_entry_data(entry_data)
                entries[entry_data["id"]] = entry_data
    entries = {
        id: entries[id]
        for id in sorted(entries.keys(), key=lambda entry: (entries[entry]["date"]))
    }
    return entries


entries = _entries()


def _series():
    series = []
    for entry_data in reversed(entries.values()):
        if "series" in entry_data and slugify(entry_data["series"]) not in map(
            slugify, series
        ):
            series += [entry_data["series"].strip()]
    return series


series = _series()


def _tags():
    tags = []
    for entry_data in entries.values():
        if "tags" in entry_data:
            tags += [
                tag.strip()
                for tag in entry_data["tags"]
                if slugify(tag) not in map(slugify, tags)
            ]
    return sorted(tags)


tags = _tags()

years = sorted(
    set([datetime.datetime(entry["date"].year, 1, 1) for entry in entries.values()])
)

months = sorted(
    set(
        [
            datetime.datetime(entry["date"].year, entry["date"].month, 1)
            for entry in entries.values()
        ]
    )
)

days = sorted(
    set(
        [
            datetime.datetime(
                entry["date"].year, entry["date"].month, entry["date"].day
            )
            for entry in entries.values()
        ]
    )
)

updated = sorted(
    [entry["date"] for entry in entries.values()]
    + [entry["modified"] for entry in entries.values() if "modified" in entry]
)[-1]