[
Lists Home |
Date Index |
Thread Index
]
Tons o' comments from out there, much of it in the category of "how
could I have stupidly missed that" then some more stuff that's been
boiled down in xml-dev. Whereas this is unlikely to be a second SAX, I
suspect it has a chance of being another success with xml-dev stamped
on the side. This note summarizes the changes I plan that I don't
think are controversial, and requests-for-enhancement that am just
*not* gonna do. I'll follow up with some detailed posts in issues that
I think could use more argument.
1. Change comments from // to /* */. Sigh.
2. Need a "dispose" method, and a callback to set the free() when you
set the allocator
3. the genxPI call needs separate arguments for the target and the rest
of the text
4. add get* calls analogous to the set* calls, on general principles
and to allow installation of local error handlers and so on.
5. Lose all the genx*W calls (the ones with codePoint * args). Keep
the genxCharacter(genxWriter w, int character) call.
6. Lose the abort()-on-error; after the first error all genx-calls are
no-ops that keep returning error
7. A couple more errorcodes, the number of possible error conditions is
in single digits even with really granular distinctions, might as well
report 'em accurately
Requests for enhancement that I am just *not* going to do unless we get
some really new information:
1. Any output encoding other than UTF-8
2. Optional escaping of illegal characters
3. Prettyprinting support
4. Various kinds of error workarounds, and turning errorchecking off
5. Writing CDATA sections
6. Writing XML declaration (good idea, but I want the output to be
Canonical XML)
7. Default namespaces
8. Control over attribute quote characters
9. Control over <foo/> vs. <foo></foo>
10. Functions to write an internal DTD subset
11. Access to internal namespace declarations
|
