#Changelog for KindleEar
##1.20.16
- 增加一个选项,可选择使用网页标题还是feed标题做为文章标题。
##1.20.16
- A option added for choose title come from webpage or feed.
##1.20.15
- bugfix: weixinbase部分解码失败问题。
##1.20.15
- bugfix: weixinbase failed when decode some article.
##1.20.14
- bugfix: 喷嚏图啩原RSS地址已经失效,更换为另一个地址。
##1.20.14
- bugfix: change address of rss of book 'dapenti.py'.
##1.20.13
- bugfix: 修改有部分RSS将一个图片做为一篇文章(没有html包装)导致推送失败的问题。
##1.20.13
- bugfix: a picture as an article (without html wraper) will crash KindleEar.
##1.20.12
- 导出opml时将url转义。
##1.20.12
- quote url when export subscriptions to opml.
##1.20.11
- "导入订阅列表"功能支持OPML元素Outline嵌套。
##1.20.11
- "Import Feeds" supports nested outlines in OPML file.
##1.20.10
- bugfix: 修正有部分文章无法提取正文导致推送失败的问题。
##1.20.10
- bugfix: some complex articles cause failure of delivery.
##1.20.9
- 增加从opml文件导入订阅列表的功能。
- 网友seff增加特性:可选部分书籍单独推送。
##1.20.9
- New feature: import list of custom rss from a opml file.
- New feature added by seff: add a switch 'Separate' to books. book is pushed to kindle separately if it's checked.
##1.20.8
- 网友mcfloundinho增加《共识网》
##1.20.8
- a book added by mcfloundinho.
##1.20.7
- bugfix:修正trigger@appid.appspotmail.com触发投递失败的问题。
##1.20.7
- bugfix:trigger@appid.appspotmail.com trigger task deliver failed.
##1.20.6
- bugfix:继续修改urlopener处理cookie的一个bug.
##1.20.6
- bugfix: refix a bug in process cookie of module urlopener.
##1.20.5
- 增加网友mcfloundinho提供的《南方周末》。
- bugfix:修改urlopener处理cookie的一个bug.
##1.20.5
- Add book nfzm written by mcfloundinho.
- bugfix: fix a bug in process cookie of module urlopener.
##1.20.4
- 更新土耳其语翻译。
##1.20.4
- Update Turkish translation.
##1.20.3
- 修改内置书籍TED渤海湾以适应其网站改版。
##1.20.3
- fix book 'TEDxBohaiBay'.
##1.20.2
- 针对使用图片延迟加载技术的网页特殊处理,可以获取部分此类网页的图片。
##1.20.2
- Supports some webpage which images take a 'data-src' attribute to load asynchronous content.
##1.20.1
- 新特性,在合并推送时将各书籍的封面拼贴起来。默认已经开启,如果你使用以前的config.py,请设置DEFAULT_COVER_BV=None,如果不喜欢此效果,可以设置DEFAULT_COVER_BV='cv_bound.jpg'
- bugfix: 修正保存到evernote不成功的问题(1.13引入)
##1.20.1
- Paste all covers into one when merge books into one. DEFAULT_COVER_BV=None (default value) to enable the feature.
- bugfix: send mail to evernote failed. (bug from version 1.13)
##1.20
- 增加一个简单的正文提取模块,在readability失败后启用。
- 增强的网页解码器,综合考虑http响应头/html文件头/chardet检测结果,效率更高,乱码更少。
- 支持需要登陆才能查看文章的网站,请参照FAQ如何使用。
- 针对一天推送多次的需求,书籍属性‘oldest_article’大于365则使用秒为单位。
- 增强的密码安全,加salt然后md5,无法通过密码词典破解,在可接受的代价范围内无法暴力破解。 (仅新账号启用,如果需要可以删掉admin然后重新登陆就会新建admin账号)
- 整理文件夹结构,将相关库都放到lib目录下。
- 其他一些小的可用性增强。
升级注意:书籍的fetcharticle()增加了一个参数,如果你定制的书籍使用到此接口,需要修改。
##1.20
- a new simple algorithm to extract content of webpage when module readability failed.
- a new enhanced decoder for webpage which detection algorithm includes more parameters: http response header, meta of html, result of chardet.
- support site that need subscription, refers to FAQ for more detail.
- use second as unit when value of property 'oldest_article' of book more than 365.
- Enhanced password encryption with salt, more safe when face a brute force attack. (for new account only, you can delete account admin and login again for enjoy it.)
- neaten folder structure, put all libs into folder 'lib'.
- some minor improves for usability.
Note:interface fetcharticle() in base.py modified (a new parameter added), if your book implemented it, please modify it to work.