enag@ifi.uio.no (Erik Naggum) (11/29/90)
Can some kind soul please enlighten me as to the purpose of the "Entity end" (out-of-band) "signal" and it's use in SGML documents? I got so far as to understand how you would represent it (ASCII NUL or DEL seem to fit the description very nicely), but as to the rationale for its existence, I'm at a complete loss. If you think this is narrow interest, mail me. If I get enough replies like this, I'll summarize. Would prefer to use the group, though. -- [Erik Naggum] Snail: Naggum Software / BOX 1570 VIKA / 0118 OSLO / NORWAY Mail: <erik@naggum.uu.no>, <enag@ifi.uio.no> My opinions. Wail: +47-2-836-863 --
ath@prosys.se (Anders Thulin) (11/30/90)
In article <ENAG.90Nov29020830@hild.ifi.uio.no> enag@ifi.uio.no (Erik Naggum) writes: >Can some kind soul please enlighten me as to the purpose of the >"Entity end" (out-of-band) "signal" and it's use in SGML documents? As far as I can see, it is used in the same was as EOF is in C: it signals 'end of file' for all entities, also those that are internal. I'm not certain what you mean with 'out-of-band' though. >I got so far as to understand how you would represent it (ASCII NUL or >DEL seem to fit the description very nicely), Perhaps you are right. Given the rather bewildering possibilities of mapping characters to and fro in SGML I'd be inclined to think that no byte value in *any* code table can be used as a universal EE. But perhaps I'm bewildered beyond what's required ... :-). > but as to the rationale for its existence, I'm at a complete loss. I don't know about this. I'm inclined to suspect that it's something to do with preventing markup inside entities to propagate outside. <!ENTITY foo "<p>some random text"> and <p>This is &foo; as an example</p> The <p> in foo isn't permitted to last beyond the EE of foo. I don't have the standard handy to check this out, so I may be off track here. -- Anders Thulin ath@prosys.se {uunet,mcsun}!sunic!prosys!ath Telesoft Europe AB, Teknikringen 2B, S-583 30 Linkoping, Sweden