Re: Review Konrad\\'s message re xml:base by next call

Hi,

See intermixed....

Konrad Lanz escribió:
> should be treated by: ' or ".." ' --> append "/.."
>> But there are only two ways of obtaining the output:
>> "then append ".." or "/.." for the latter case respectively to the 
>> output buffer".
>> By reading this I do not know what I have to add to the output buffer 
>> in case 1. (out buffer is empty).
> append ".." (because ' "/.." for the latter case respectively ' should 
> be applied for the last case only)
>
I still think that the text is at least subject to missinterpretations, 
as the existence of only two cases, empty buffer or buffer equals to 
"../" being one and buffer equals to ".." being the other, is not 
completely clear from the text... I would suggest to reword it to reduce 
risks of missinterpretations.
>
> Please note Action-49 
> http://lists.w3.org/Archives/Public/public-xmlsec-maintwg/2007Jun/0028.html: 
>
>> //s/remove the last segment and its preceding "/" (if any) from the 
>> output buffer/remove the last segment and its preceding "/" (if any) 
>> from the output buffer and if hereby the first character in the 
>> outputbuffer was removed and it was not the root slash then delete a 
>> leading slash from the input buffer///
>
I see. Konrad, I plan to check all the examples that you provide in the 
annex.Then I take the annex that you circulated, make this text 
substitution, apply the algorithm to the examples and let you know...

Regards

Juan Carlos.
                                         

> I tried to stay as close as possible to the original text, but maybe a 
> complete rewording might simplify the processing.
>
> Eg. : Loosely speaking something like:
> Complete ".." path segments eat the previous segment, iff it is not a 
> ".." segment by itself.  Complete ".." segments accumulate to the left 
> if the URI reference is relative and no path segments to be eaten 
> appear, if however the URI is absolute complete ".." segments 
> disappear when they hit the root slash ...
>
> and so on ... treating "//", ".", "/./" and so forth
>
> regards
> Konrad
>

Received on Tuesday, 12 June 2007 16:29:24 UTC