- From: Jose Kahan <jose.kahan@w3.org>
- Date: Wed, 12 Apr 2006 16:55:11 +0200
- To: Glenn Marcy <gmarcy@us.ibm.com>
- Cc: public-xml-core-wg@w3.org
Hi Glenn, I went thru your update, specifically section 2.4. I have some minor comments.a [xml:base URL resolution algorithm] Is the algorithm that's described in section 2.4 useful for other joint heritable attributes, or just for xml-base? If this is not the case, I think it would be better to have a subsection (2.4.1?) for dealing specifically with xml:base, [ Simplifying par. 4] par. 4 of section 2.4 says: <quote> A joint heritable attribute is xml:base, which needs a "join URI function" which takes any URI (uri-1) from an ancestor and joins a relative URI of E (rel-uri-2) after the last slash of the former and then normalizes the result. This function may also be called with a null URI, i.e. no xml:base attribute exists in E (not to be confused with xml:base=""). Although there are many ways to do this, we will describe a simple method using a separate string buffer in a manner similar to that found in section 5.2 of RFC 2396. Perform the following steps in order: </quote> I think this paragraph is a bit confusing: - I'd move the following text to a line below the description of the algorithm and say explicityl what the function should return with a null URI. This function may also be called with a null URI, i.e. no xml:base attribute exists in E (not to be confused with xml:base=""). The reason is that this paragraph is defining the function and algorithm and you're already describing how it should handle special cases before having described the general ones. - I'd propose a rewording of the following paragraph to remove some colloquial language and to remove a potential confusion along the lines of (there are many ways, this is only one of them, you may use others): Although there are many ways to do this, we will describe a simple method using a separate string buffer in a manner similar to that found in section 5.2 of RFC 2396. Perform the following steps in order: I think this phrase should say something more like "we will now describe a simple method for canonicalizing xml:base joint heritable attributes. This method uses a separate string buffer in a manner similar to that found in section 5.2 of RFC 2396. Perform the following steps in order: I'm writing this quickly. Sorry about it. - I'm not sure if your method takes into account the "path/path/../../something/". Step 5 says: All occurrences of "<segment>/../", where <segment> is a complete path segment not equal to "..", so that's why I'm confused. -jose
Received on Wednesday, 12 April 2006 14:59:41 UTC