Biblio
Description
Biblio is a full static site which can convert your repository on github into an online book shelf (I call it liblio) and reader.
I writed it for myself. I'm not familiar with page designing so the page customization will be provided only by _header.seg.html, _banner.seg.html and _footer.seg.html.
Anything is fetched from github repository (or github pages) by javascript, through the great restful github api v3.
The site contain an small sample liblio as default books for testing.
My personal liblio, which contains some Chinese and Worldwide Classical Literature:
Inebriatable Libliotheca - 可醉楼藏书
Techical
The site is writen by pure html5/css3/javascript.
Configuration
The site is based on two repositories: biblio (host site) and liblio (shelf site).
So Configuration is splitted into two. Configuration file should be placed under the /_biblio folder of repositories. If the liblio configuration is not found, some configuration item will be same as the biblio (which means you host your liblio in the same site of biblio).
_biblio/liblio.json: main configuration of the liblio site.
Configurations can also be overrided by url parameters (query string), so that you can directly access your liblio repository by this site url with just customizing query string (refer to the Security section for tracking).
The starter page of biblio site will be the README of your liblio repository.
Shelf
File types
File type is determinated by the filename, as the last extension name. Now supported file types (including planning) are:
- plain text: .txt, files should have a UTF-8 encoding and UNIX(LF) line breaking.
- markdown: .md
- html: .html/.htm,
not recommeded. - image: .jpg/.jprg/...
to be implemented - pdf: .pdf
to be implemented - epub: .pdf
to be implemented
The supporting of pdf and epub will need file format online viewer implementation of javascript. Any advice is welcomed and thankful.
Title
The title of the page in browser after content loaded will be parsed:
- For
textfiles, the filename (without extension name) will be read as title of the book. - For
markdownfiles, the first highest level header (for example, the first#line) will be read as title of the book.
Treeview
Folders name will be read as tree directory and tags of every book.
Folder with name starting as '~' or '_' will be ignore as internal folder.
Tags
- folders: children count as tag
- files:
- each parent folder name
- file size (bytes of bass64 encode)
- extension of filename
Page Layout
(To be writen...)
Cache
Since the accessing of github api will be forbidden on frequent reading, I use localStorage of browser as cache of responses from api. You can clean them to refresh the site.
Auth
(To be writen...)
Security
Any content of files will be transferred by BASE64 and HTTPs (HTTPs based on your github site settings). So theoretically they're safety. But the filenames and filder names will be plain text, so it can be monitor on network transferring.
I put a source tracking support by clustrmaps.com in code of the only html page. If you don't wanna be tracked and counted by it, fork the code and remove them by yourself. Of course you can block it by Adblock or Firefox Track Protection.
Campabilities
Tested browser include (All on latest version):
- Desktop:
- Firefox (recommended and fully tested, I don't like Chrome)
- Chrome
- Edge
- Mobile:
- Android: Firefox, Chrome
- iOS: Safari, Firefox
The site is friendly on phone/tablet layout.
Dependencies
Thanks for development of the projects helping me:
-
Pure CSS Loaders and its Site
-
Visual Studio Code
-
GitKraken