main.py

# -*- coding: utf-8 -*-
# author: yihong0618
# This blog is powered by Gitblog.

import argparse
import os
import re

import markdown
from feedgen.feed import FeedGenerator
from github import Github
from lxml.etree import CDATA
from marko.ext.gfm import gfm as marko

MD_HEAD = """## ChinoBing's Blog: https://chinobing.github.io 
[RSS Feed](https://raw.githubusercontent.com/{repo_name}/master/feed.xml)
"""

BACKUP_DIR = "BACKUP"
ANCHOR_NUMBER = 5
TOP_ISSUES_LABELS = ["Top"]
TODO_ISSUES_LABELS = ["TODO"]
FRIENDS_LABELS = ["Friends"]
ABOUT_LABELS = ["About"]
# DRAFT_LABEL = '草稿'
FRONT_LABELS = ['一、','二、','三、','四、','五、','六、','七、','八、','九、','十、']
IGNORE_LABELS = FRIENDS_LABELS + TOP_ISSUES_LABELS + TODO_ISSUES_LABELS + ABOUT_LABELS

FRIENDS_TABLE_HEAD = "| Name | Link | Desc | \n | ---- | ---- | ---- |\n"
FRIENDS_TABLE_TEMPLATE = "| {name} | {link} | {desc} |\n"
FRIENDS_INFO_DICT = {
    "名字": "",
    "链接": "",
    "描述": "",
}


def get_me(user):
    return user.get_user().login


def is_me(issue, me):
    return issue.user.login == me


def is_hearted_by_me(comment, me):
    reactions = list(comment.get_reactions())
    for r in reactions:
        if r.content == "heart" and r.user.login == me:
            return True
    return False


def _make_friend_table_string(s):
    info_dict = FRIENDS_INFO_DICT.copy()
    try:
        string_list = s.splitlines()
        # drop empty line
        string_list = [l for l in string_list if l and not l.isspace()]
        for l in string_list:
            string_info_list = re.split("：", l)
            if len(string_info_list) < 2:
                continue
            info_dict[string_info_list[0]] = string_info_list[1]
        return FRIENDS_TABLE_TEMPLATE.format(
            name=info_dict["名字"], link=info_dict["链接"], desc=info_dict["描述"]
        )
    except Exception as e:
        print(str(e))
        return


# help to covert xml vaild string
def _valid_xml_char_ordinal(c):
    codepoint = ord(c)
    # conditions ordered by presumed frequency
    return (
        0x20 <= codepoint <= 0xD7FF
        or codepoint in (0x9, 0xA, 0xD)
        or 0xE000 <= codepoint <= 0xFFFD
        or 0x10000 <= codepoint <= 0x10FFFF
    )


def format_time(time):
    return str(time)[:10]


def login(token):
    return Github(token)


def get_repo(user: Github, repo: str):
    return user.get_repo(repo)


def parse_TODO(issue):
    body = issue.body.splitlines()
    todo_undone = [l for l in body if l.startswith("- [ ] ")]
    todo_done = [l for l in body if l.startswith("- [x] ")]
    # just add info all done
    if not todo_undone:
        return f"[{issue.title}]({issue.html_url}) all done", []
    return (
        f"[{issue.title}]({issue.html_url})--{len(todo_undone)} jobs to do--{len(todo_done)} jobs done",
        # todo_done + todo_undone,
        todo_undone,
    )


def get_top_issues(repo):
    return repo.get_issues(labels=TOP_ISSUES_LABELS)


def get_todo_issues(repo):
    return repo.get_issues(labels=TODO_ISSUES_LABELS)


def get_repo_labels(repo):
    return [l for l in repo.get_labels()]


def get_issues_from_label(repo, label):
    return repo.get_issues(labels=(label,))


def add_issue_info(issue, md):
    time = format_time(issue.created_at)
    md.write(f"- [{issue.title}]({issue.html_url})--{time}\n")


def add_md_todo(repo, md, me):
    todo_issues = list(get_todo_issues(repo))
    if not TODO_ISSUES_LABELS or not todo_issues:
        return
    with open(md, "a+", encoding="utf-8") as md:
        md.write("## TODO\n")
        for issue in todo_issues:
            if is_me(issue, me):
                todo_title, todo_list = parse_TODO(issue)
                md.write("TODO list from " + todo_title + "\n")
                for t in todo_list:
                    md.write(t + "\n")
                # new line
                md.write("\n")


def add_md_top(repo, md, me):
    top_issues = list(get_top_issues(repo))
    if not TOP_ISSUES_LABELS or not top_issues:
        return
    with open(md, "a+", encoding="utf-8") as md:
        md.write("## 置顶文章\n")
        for issue in top_issues:
            if is_me(issue, me):
                add_issue_info(issue, md)


def add_md_firends(repo, md, me):

    s = FRIENDS_TABLE_HEAD
    friends_issues = list(repo.get_issues(labels=FRIENDS_LABELS))
    if not FRIENDS_LABELS or not friends_issues:
        return
    friends_issue_number = friends_issues[0].number
    for issue in friends_issues:
        for comment in issue.get_comments():
            if is_hearted_by_me(comment, me):
                try:
                    s += _make_friend_table_string(comment.body or "")
                except Exception as e:
                    print(str(e))
                    pass
    s = markdown.markdown(s, output_format="html", extensions=["extra"])
    with open(md, "a+", encoding="utf-8") as md:
        md.write(
            f"## [友情链接](https://github.com/{str(me)}/gitblog/issues/{friends_issue_number})\n"
        )
        md.write("<details><summary>显示</summary>\n")
        md.write(s)
        md.write("</details>\n")
        md.write("\n\n")


def add_md_recent(repo, md, me, limit=5):
    count = 0
    with open(md, "a+", encoding="utf-8") as md:
        # one the issue that only one issue and delete (pyGitHub raise an exception)
        try:
            md.write("## 最近更新\n")
            for issue in repo.get_issues():
                if is_me(issue, me):
                    add_issue_info(issue, md)
                    count += 1
                    if count >= limit:
                        break
        except Exception as e:
            print(str(e))


def add_md_header(md, repo_name):
    with open(md, "w", encoding="utf-8") as md:
        md.write(MD_HEAD.format(repo_name=repo_name))
        md.write("\n")


def add_md_label(repo, md, me):
    labels = get_repo_labels(repo)

    # sort lables by description info if it exists, otherwise sort by name,
    # for example, we can let the description start with a number (1#Java, 2#Docker, 3#K8s, etc.)
    labels = sorted(
        labels,
        key=lambda x: (
            x.description is None,
            x.description == "",
            x.description,
            x.name,
        ),
    )

    with open(md, "a+", encoding="utf-8") as md:
        # skip_issues = get_issues_from_label(repo, DRAFT_LABEL)
        j = 0 #inital index for FRONT_LABELS
        for label in labels:
            # we don't need add top label again
            if label.name in IGNORE_LABELS:
                continue
            issues = get_issues_from_label(repo, label)
            if issues.totalCount:
                md.write("## " + f'{FRONT_LABELS[j]}' + label.name + "\n")
                issues = sorted(issues, key=lambda x: x.created_at, reverse=True)
                j += 1
            i = 0
            for issue in issues:
                if not issue:
                    continue
                # if (issue in skip_issues) and (label.name != DRAFT_LABEL):
                #     continue
                if is_me(issue, me):
                    if i == ANCHOR_NUMBER:
                        md.write("<details><summary>显示更多</summary>\n")
                        md.write("\n")
                    add_issue_info(issue, md)
                    i += 1
            if i > ANCHOR_NUMBER:
                md.write("</details>\n")
                md.write("\n")


def get_to_generate_issues(repo, dir_name, issue_number=None):
    md_files = os.listdir(dir_name)
    generated_issues_numbers = [
        int(i.split("_")[0]) for i in md_files if i.split("_")[0].isdigit()
    ]
    to_generate_issues = [
        i
        for i in list(repo.get_issues())
        if int(i.number) not in generated_issues_numbers
    ]
    if issue_number:
        to_generate_issues.append(repo.get_issue(int(issue_number)))
    return to_generate_issues


def generate_rss_feed(repo, filename, me):
    generator = FeedGenerator()
    generator.id(repo.html_url)
    generator.title(f"RSS feed of {repo.owner.login}'s {repo.name}")
    generator.author(
        {"name": os.getenv("GITHUB_NAME"), "email": os.getenv("GITHUB_EMAIL")}
    )
    generator.link(href=repo.html_url)
    generator.link(
        href=f"https://raw.githubusercontent.com/{repo.full_name}/master/{filename}",
        rel="self",
    )
    for issue in repo.get_issues():
        if not issue.body or not is_me(issue, me) or issue.pull_request:
            continue
        item = generator.add_entry(order="append")
        item.id(issue.html_url)
        item.link(href=issue.html_url)
        item.title(issue.title)
        item.published(issue.created_at.strftime("%Y-%m-%dT%H:%M:%SZ"))
        for label in issue.labels:
            item.category({"term": label.name})
        body = "".join(c for c in issue.body if _valid_xml_char_ordinal(c))
        item.content(CDATA(marko.convert(body)), type="html")
    generator.atom_file(filename)


def main(token, repo_name, issue_number=None, dir_name=BACKUP_DIR):
    user = login(token)
    me = get_me(user)
    repo = get_repo(user, repo_name)
    # workspace_path = os.environ.get('GITHUB_WORKSPACE')
    # md_file = workspace_path+"/README.md"
    md_file = "README.md"
    # add to readme one by one, change order here
    add_md_header(md_file, repo_name)
    for func in [add_md_firends, add_md_top, add_md_recent,add_md_label, add_md_todo]:
        func(repo, md_file, me)

    generate_rss_feed(repo, "feed.xml", me)
    to_generate_issues = get_to_generate_issues(repo, dir_name, issue_number)

    # save md files to backup folder
    for issue in to_generate_issues:
        save_issue(issue, me, dir_name)


def save_issue(issue, me, dir_name=BACKUP_DIR):
    md_name = os.path.join(
        dir_name, f"{issue.number}_{issue.title.replace('/', '-').replace(' ', '.')}.md"
    )
    meta_info = """---
title: {title}
categories:
{categories}
date: {date}
{_build}
---

"""
    title = issue.title
    date = issue.created_at.strftime("%Y-%m-%d")
    categories = ""
    _build = ""
    for label in issue.labels:
        categories += f"- {label.name}\n"
        if label.name in IGNORE_LABELS:
            _build = """_build:
   list: never
"""
    with open(md_name, "w") as f:
        f.write(meta_info.format(title=title, categories=categories, date=date, _build=_build))
        f.write(f"# [{issue.title}]({issue.html_url})\n\n")
        f.write(issue.body or "")
        if issue.comments:
            for c in issue.get_comments():
                if is_me(c, me):
                    f.write("\n\n---\n\n")
                    f.write(c.body or "")


if __name__ == "__main__":
    if not os.path.exists(BACKUP_DIR):
        os.mkdir(BACKUP_DIR)
    parser = argparse.ArgumentParser()
    parser.add_argument("github_token", help="github_token")
    parser.add_argument("repo_name", help="repo_name")
    parser.add_argument(
        "--issue_number", help="issue_number", default=None, required=False
    )
    options = parser.parse_args()
    main(options.github_token, options.repo_name, options.issue_number)