- From: Daniel Barclay <Daniel.Barclay@digitalfocus.com>
- Date: Tue, 28 Mar 2000 12:31:38 -0500
- To: www-xml-infoset-comments@w3.org
There are a number of errors in the document currently at http://www.w3.org/TR/xml-infoset: * Section 2.3.2, para. 1, sentence 1 says "following property"; it should say "... properties" (multiple properties follow). * Section 2.5, para. 1 says: A validating XML processor will never generate reference to skipped entity information items for a valid XML document. - "reference" should be "references" * Section 2.6, para. 2 says "a LF ... character." That should probably be "an LF ... character." (I just realized that the author must have been saying "linefeed"; however, since the actual text uses the acronym (which is pronounced with an initial vowel, "ell eff") instead of the word "linefeed", it should probably say "an LF ...") (By the way, are those CR/LF processing rules stated correctly? They seem to be ambiguous. The second paragraph says: Note, however, that a CR (#xD) character that is followed by a LF (#xA) character is not represented by any information item. Furthermore, a CR character that is not followed by a LF character is treated as a LF character. ... In what order must the rules be applied to characters, and when a CR character is treated as an LF character does that include in evaluation of those rules? That is, it the following a valid application of the rules? Consider the four-character sequence "x<CR><CR>x". - The second CR character is not followed by an LF character, so it is treated as an LF character (and is represented). - The first CR character is now followed by an LF character (treating the second CR character as an LF character), so is it not represented. Then, is this also a valid application?: Consider the same sequence above. - The first CR character is not followed by an LF character, so it is treated as an LF character (and is represented). - The second CR character is not followed by an LF character, so it is treated as an LF character (and is represented). Note the different results: on represented LF character vs. two of them. Does "treated" means everywhere _except_ in application of those rules? If so, it would be good to clarify that. Also, what about "<CR><CR><LF>"? (The rules aren't ambiguous, but I wanted to check the intent.) - The first CR is not followed by an LF character, so it is treated as an LF character (and represented, as an LF). - The second CR is followed by an LF, so it is not represented. - The LF character is represented. Note how this represents two line breaks. Is that intentional? (On a teletypewriter, that sequence (any number of CRs in a row followed by an LF) would accomplish _one_ line break. On the other hand, extra CR characters before a CR/LF sequence aren't so innocuous in many places in modern operating systems and networks.) ) * In Section 2.13, para. 1, sentence 1 says "an peripheral." * I can't find it right now, but it appears that one right bracket (after an attribute name) was missing space between it and the following word. Daniel -- Daniel Barclay Digital Focus Daniel.Barclay@digitalfocus.com
Received on Tuesday, 28 March 2000 12:29:25 UTC