IMDb, CC, MOC, FIB, all kinds of billboard that connect the item title with douban ID.
- See Assigned Snippets in CSV file 00_snippets.csv
- Un Assigned:
- Douban TOP 250 : 99_douban_top250.csv
-
upload rank data file like
01_IMDbtop250.csv
indata
dir.- The file name MUST obey the regexr rule
\d{2}_.+?\.csv
and is a csv file. - The csv file MUST have headers and the first row must be
rank
orspine
, and with a row nameddbid
.
- The file name MUST obey the regexr rule
-
Assign rank list data in
data\00_snippets.csv
with those information.
Name | Descr | Sample |
---|---|---|
file | The name of csv file | 03_CClist.csv |
title | The title of rank list | The Criterion Collection 标准收藏 |
short_title | The short title of rank list | CC标准收藏编号 |
href | The homepage url of rank list | https://www.criterion.com/shop/browse/list?sort=spine_number |
other | The other key you want to assign in json file,the format like {key1}:{value1}|{key2}:{value2} |
prefix:# |
-
If the rank list want to update automatically (By GitHub Action), you MUST write update function in
update_snippets.py
-
Run those Python3 script.
python update_snippets.py
python build_data.py