wiki-archive/twiki/pub/TWiki/KupuContrib/_kupu/doc/FAQ.txt

115 lines
5.5 KiB
Plaintext

===============================
Kupu Frequently Asked Questions
===============================
* Will this work with Safari or Opera or ...?
No, Kupu will only run on browsers with the IE engine (IE 5.5+) or those
with a Gecko engine (Mozilla 1.3.1+ and Netscape 7.1+), not on the
Konqueror one (which Safari uses) or on Opera or any other browser.
Actually Mozilla copied a feature unique to Internet Explorer, which is
not part of any standard, which Kupu uses. Because it's not part of a
standard, it is not very likely any other browsers will implement it,
although we keep hoping some day they will.
* Kupu allows users to do all kinds of things - like coloring text -
that I don't want in my system.
All of the kupu style functions can be turned on and off using CSS or in
the configuration. Additionally, the element controls can be extended to
accomodate special needs of your content.
* Is my platform supported?
It probably is: Kupu has requirements on the client side, but works with
just about any web server. As long as your server supports POST and some
server-side scripting language or CGI (to process form variables if Kupu
is part of a form) or HTTP PUT you can install Kupu on your website or
application.
* Why do I get all kinds of weird characters?
Kupu seems to work correctly when you use UTF-8 as the character set for
encoding, but gives unpredictable results when you use another one such
as latin-1 (ISO-8859-1). Make sure to always use UTF-8 as the character
set for incoming data and set the correct Content-Type header or meta-tag,
and Kupu will send back UTF-8 data.
* Saving seems to work, but my browser keeps showing old stuff
Browsers tend to store web pages a while to reduce the amount of data that
has to be transferred. Obviously this can get in the way when using Kupu,
so the browsers need to be told not to store (cache) the Kupu iframe contents
on retrieval. This can be done either with HTTP headers or with meta-tags.
For an example of which meta-tags to use, see fulldoc.html (the example
content page).
* What's this '1223' status code I see in the code?
There seems to be a bug in Internet Explorer that in some cases it says that
the status code of a save action is '1223' (a code that doesn't exist in
HTTP, it's not described in the specs and there seems to be no information
about this on the web). Since it looks like this code is only returned on a
succeeding save action, it seems that reasonable safe work-around is to add
the status code as one of the allowed codes in common/kupueditor.js (search
for '204' and you'll find a line that checks if the status code is either 200
or 204, add 1223 there). This is what we've done, however, if you find
something can go wrong with that workaround, please let us know.
* Does Kupu produce valid XHTML?
Well... Kupu produces *well-formed XML*, which out-of-the-box does not
contain any elements that aren't allowed in XHTML anymore, and are written
in a form compliant to the XHTML notation (lowercase tag and attribute
names, attributes all have values and all values are in quotes) but it is
theoretically possible to create constructions that are not valid according
to the XHTML DTD. For most cases, however, the data produced by Kupu is
perfectly usable, even after copied from another website or notorious Word.
* What about i18n?
Currently Kupu is not adapted to i18n. The plan is to write i18n support for
the JavaScript sources in one of the next versions (probably using XML
inside the HTML, a technique sometimes called 'XML data islands', to store
all the JavaScript strings) but currently it's not possible to convert the
JavaScript strings on the server without having to hack in the sources.
Current technologies make it hard to handle i18n in a clean way on the
client, so therefore Kupu will at first choose to let the server take care
of it. Since Kupu is not dependent on a specific server, however, we can not
guarantee that support for your server will be available even in the future,
but we will at least provide infrastructure and probably also dictionaries
for different languages.
* What about Bitflux, Xopus, and others?
The authors of these editors have done great work. They are superb
applications, but the major drawback is they work in only one browser (as of
this writing). Kupu was created to work in both Mozilla/Netscape and Internet
Explorer.
* What's with the xsltproc error when 'make'ing kupu?
xsltproc wants to load external entities references in the XML
files, such as DTDs. External entities that are referenced by URIs
are typically loaded when you have a network
connection. Alternatively, you can install the DTD on your harddrive
and point xsltproc to an XML catalog that resolves the URIs to local
harddrive directories.
Most users that generate the kupu macros probably don't have the
DTDs installed on their harddrive. When they invoke xsltproc, it
will load them from the net. If you don't have an internet
connection, you'll see the above error.
To stop xsltproc from getting resources from the net (because it is
time consuming and unnecessary), we pass it the --nonet switch. A
bug (?) in xsltproc makes it then complain about not being able to
load entities from the net--but that's exactly what we want to
achieve! So, this error message is not an error message for us--it
is the contrary!
Other questions
---------------
Additional questions for the FAQ are welcome. Please post them on:
kupu-dev@codespeak.net