Re: Re: [PEPr] Comment on HTTP::HTTP_Cache

Hi, Ian Eure wrote:

I disagree. There is nothing wrong with using ETag in conjunction with Last-Modified, and I don't see how this would cause a BC break. ETag support is a new feature, not a change to an existing one. Even if there were problems, ETag support could be made optional.

To support all features of HTTP_Cache, the constructor needed to be changed. The constructor currently accepts a timespan. I would like to keep the output buffering feature, I have in HTTP_Cache and it's perfect to enable this feature in the constructor. That's why I think, there needed to be some changes to the class, which could lead to a BC break. Maybe it would be better to start from scratch as HTTP_Header_Cache looks like it was designed only with Last-Modified in mind. Furthermore I guess this functionality should be in its own package, I couldn't find this functionality in PEAR as I did not look in the description of HTTP_Header... I guess a lot of people could use the functionality hidden inside the HTTP_Header package. Stephan -- http://www.php-tools.net http://www.schst.net http://pear.php.net