Inconsistent permissions on CPAN modules

栏目: IT技术 · 发布时间: 5年前

In the past, if multiple people released the same distribution, and each of them added new modules to the distribution, then you'd end up with different people having ownership of modules making up a single distribution. That's a pain if you want to give co-maint to someone. PAUSE has been fixed now, to ensure consistent permissions, but there are historical inconsistencies. I'll describe the problem, and how we're fixing it.

This article assumes you know something about PAUSE Indexing Permissions. If you don't, then you should first read the relevant section of the PAUSE Operating Model .

If you are the first person to release a module to CPAN, you are given the first-come indexing permission on that module. At some later date you give co-maint to someone, we'll call her Mary, and they start doing regular releases. In time they gain confidence, and decide to refactor your code, adding some new modules.

In the past Mary would have been given the first-come indexing permission on the new modules that she released, resulting in a situation where you own some of the modules, and Mary owns the rest. This became a pain if you wanted to give co-maint to a third person, as you'd have to do it for some modules, and Mary for the others.

So now PAUSE handles this scenario differently. When new modules are added to a distribution, PAUSE gives the first-come indexing permission to whoever has first-come on the lead module of the distribution, and gives co-maint to everyone who has co-maint on that same module. This means that distributions should keep a single owner, unless you intentionally break that.

We're left with the problem of the historical cases where a co-maint added one or more modules to a distribution and were granted first-come on them. There are about 550 such distributions, and I plan to gradually work through them. This will take a while, and I'll start off working through the easy cases, but if you'd like me to sort out one or more of your distributions, feel free to email me (neil at bowers dot com).

There are some hairy cases, and for those I'll get in touch with the involved parties, to ensure the right/best person ends up with first-come. And then there are core modules; I'll get to those last :-)


以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持 码农网

查看所有标签

本站部分资源来源于网络,本站转载出于传递更多信息之目的,版权归原作者或者来源机构所有,如转载稿涉及版权问题,请联系我们

轻松学算法

轻松学算法

赵烨 / 电子工业出版社 / 2016-7 / 69.00元

《轻松学算法——互联网算法面试宝典》共分为12 个章节,首先介绍了一些基础的数据结构,以及常用的排序算法和查找算法;其次介绍了两个稍微复杂一些的数据结构——树和图,还介绍了每种数据结构和算法的适用场景,之后是一些在工作与面试中的实际应用,以字符串、数组、查找等为例介绍了一些常见的互联网面试题及分析思路,便于读者了解这些思路,顺利地通过互联网公司的面试;最后介绍了一些常见的算法思想,便于读者对今后遇......一起来看看 《轻松学算法》 这本书的介绍吧!

JS 压缩/解压工具
JS 压缩/解压工具

在线压缩/解压 JS 代码

JSON 在线解析
JSON 在线解析

在线 JSON 格式化工具

HEX HSV 转换工具
HEX HSV 转换工具

HEX HSV 互换工具