Re: lynx-dev Re: lynx timeline

[Leonid Pauzner]

5-Jun-2003 13:01 I wrote:
> 4-Jun-2003 13:43 Ilya Zakharevich wrote:
>> On Tue, Jun 03, 2003 at 03:42:06PM -0700, Ilya Zakharevich wrote:
>>>   d) no-cache should be ignorable per lynx.cfg;
>>>
>>>   e) expiration date should be ignorable (best in two steps: I should
>>>      be able to request "no expiration in less than 6 min", AND I
>>>      should be able to request "ask-with-default-no for expiration
>>>      dates below 1.5 hour").

>> d) HONOR_NO_CACHE=yes/no

> "no_cache" is effectively the same as "expire <= now"
> it will acts differently according to your suggestions d) and e).

We currently have the following behaviour: (1) always reload documents with
"no_cache" or "expired <= now" or equivalent. (2) all other documents
(including ones expired in the future) will never be reloaded if they are in
HText cache. Do you intend to reload more often when the expire time comes,
e.g. be closer to HTTP standard?

Do you intend ignoring "no_cache" to shame a bad practice,
but reload documents if the Expired field is set properly in the future?
There wouldn't be a fallback to the current behaviour.


>> e) NO_EXPIRATION_BEFORE=seconds1
>>    ASKNO_FOR_EXPIRATION_BEFORE=seconds2
>>    ASKYES_FOR_EXPIRATION_BEFORE=seconds3

>>    gives 4 intervals for age of expired documents:

>>       0..seconds1        -    silently ignore expiration
>>       seconds1..seconds2 -    ask user about reload with default=no
>>       seconds2..seconds3 -    ask user about reload with default=yes
>>       seconds3..Inf      -    silently reload



; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden