I met this error when generating poster for duolingo

File "/home/<user>/GitHubPoster/loader/base_loader.py", line 29, in <listcomp>
month_list = [m for m in month_list if m < pendulum.now()]  

ValueError: Posix spec does not match last transition     

I think we need to check if the timezone matches?

Traceback (most recent call last):
  File "/Users/xxxxxxxxx/opt/miniconda3/envs/py38/bin/github_poster", line 8, in <module>
    sys.exit(main())
  File "/Users/xxxxxxxxx/opt/miniconda3/envs/py38/lib/python3.8/site-packages/github_poster/cli.py", line 118, in main
    s.make_skyline()
  File "/Users/xxxxxxxxx/opt/miniconda3/envs/py38/lib/python3.8/site-packages/github_poster/skyline/skyline.py", line 115, in make_skyline
    skyline_info_card = self._make_skyline_card(text_info)
  File "/Users/xxxxxxxxx/opt/miniconda3/envs/py38/lib/python3.8/site-packages/github_poster/skyline/skyline.py", line 76, in _make_skyline_card
    w, h = measure_text(FONT, TEXT)
  File "/Users/xxxxxxxxx/opt/miniconda3/envs/py38/lib/python3.8/site-packages/sdf/text.py", line 19, in measure_text
    font = ImageFont.truetype(name, 96)
  File "/Users/xxxxxxxxx/opt/miniconda3/envs/py38/lib/python3.8/site-packages/PIL/ImageFont.py", line 855, in truetype
    return freetype(font)
  File "/Users/xxxxxxxxx/opt/miniconda3/envs/py38/lib/python3.8/site-packages/PIL/ImageFont.py", line 852, in freetype
    return FreeTypeFont(font, size, index, encoding, layout_engine)
  File "/Users/xxxxxxxxx/opt/miniconda3/envs/py38/lib/python3.8/site-packages/PIL/ImageFont.py", line 211, in __init__
    self.font = core.getfont(
OSError: cannot open resource

Hi, there.

I just add the WakaTime feature for this repo.

Here is a example:

I'm not so familiar with this project honestly speaking.

So i just use the second to calculate the counts, maybe your can change it later.

Here a typical json item is json result:

{
      "categories": [],
      "dependencies": [],
      "editors": [],
      "grand_total": {
        "digital": "0:00",
        "hours": 0,
        "minutes": 0,
        "text": "0 secs",
        "total_seconds": 0
      },
      "languages": [],
      "machines": [],
      "operating_systems": [],
      "projects": [],
      "range": {
        "date": "2020-01-01",
        "end": "2020-01-01T15:59:59Z",
        "start": "2019-12-31T16:00:00Z",
        "text": "Wed Jan 1st 2020",
        "timezone": "Asia/Shanghai"
      }
}

I use item["range"]["date"] for date, and item["grand_total"]["total_seconds"] for value.

随着支持的 loader 越来越多，每个 loader 可能会有自己的依赖，导致整个项目的依赖越来越多，但是很多用户实际上并不会用到所有的 loader，导致了一些不必要的依赖安装。

这个 PR 在不做大的重构的条件下，

1. 将各个 loader 的依赖放到了独立的 extras_require 中，比如想使用 twitter loader，就需要 pip install github_poster[twitter]。默认情况下也提供了 github_poster[all] 来简化一些小白用户的使用。
2. 将 loader 的依赖 import 延迟到了真正要使用的时候。为需要额外加载依赖的 loader 提供了 try_import_deps 方法，在使用真正的功能前尝试加载它的依赖，如果 import 失败，抛出异常，提示用户需要安装额外的依赖。

Option argument count_type, you can specify statistics type:

• record: post num (default)
• like: post be liked num
• share: post be share num
• comment: post be comment num
• repost: post be repost num

Greatest poster generator ever!

WakaTime is a widget that can be embedded into lots of IDEs to track coding time.

Here is a report page demo:

And the corresponding API Doc:

• WakaTime Doc

So i'm wondering if there is any chances generating the WakaTime's poster?

Thanks a lot!

Traceback (most recent call last): File "F:\github\GitHubPoster\cli.py", line 11, in from loader import ( File "F:\github\GitHubPoster\loader_init_.py", line 5, in from .github_loader import GitHubLoader File "F:\github\GitHubPoster\loader\github_loader.py", line 1, in from parser import GitHubParser ImportError: cannot import name 'GitHubParser' from 'parser' (unknown location)

Currently GitHubPoster only supports scraping Twitter data directly from Twitter.com. This process is time-consuming and error-prone due to, for instance, network limits.

If one has a lot of tweets that span a long range of time, it makes more sense to download a Twitter Archive and use that as the data source.

This PR adds a simple script that converts the Twitter Archive into a simple JSON file that can be fed into GitHubPoster's JSON loader.

usage

build

docker build -t github_poster .

run

docker run --rm -v $(pwd)/OUT_FOLDER:/app/OUT_FOLDER github_poster duolingo --duolingo_user_name xxxxx --duolingo_password yyyyy --me whoami

The arguments after image_name (here is github_poster) will pass to python scripts directly.

Add Notion Loader.

Use your Notion as a CMS, get data from Notion database to generate poster.

• notion_token: The Notion internal integration token, see here for more detail.
• database_id: The Notion database id, you can learn from Where can I find my database's ID?
• prop_name: The Notion database property name which stored the date and time, an ISO 8601 format date, with optional time.

NOTE

The prop_name property should be set as Date property type, or the loader won't work properly.

A Notion template as example:

• 使用子命令对命令行参数进行分组、隔离
• 由于第一条改动，命令行参数去除type名称前缀，不会有歧义
• 同样由于第一条改动，现在在适当的命令行参数上加上required=True限制
• 将loader相关参数收拢到loader类中，替代type字典
• --is-cn 更名为 --cn
• 合并youtube的两个参数为一个
• 更新README中说明

Close #19