Re: CLJ newsgroup FAQ
- From: "Bart Van der Donck" <bart@xxxxxxxxxx>
- Date: 24 Jul 2006 01:22:52 -0700
Jim Ley wrote:
On 23 Jul 2006 01:48:30 -0700, "Bart Van der Donck" <bart@xxxxxxxxxx>
wrote:
(1) If the owner of an interested website is willing to give
username/password (S/FTP), then the software at my end could
automatically replace the file on a daily base. This does not require
any extra tools, as I assume all websites are accessible by FTP one way
or another. Of course I can understand that not all website owners are
willing to share their FTP login data.
I certainly wouldn't make a website available to FTP, and I see no
reason to provide FTP simply to have SFTP when there are much better
methods.
The bigger problem with automatic updating of content generated from
an unknown webapp that allows html input, is avoiding insertion
attacks, or even simply avoiding spam - web versions of FAQ's tend to
be high ranked so well worth getting links on.
[...]
Another (better?) project description could maybe consist of the
following:
- Crontab file starts up Perl program once a day
- Perl program grabs http://www.jibbering.com/faq/index.xml
- Checks on successful GET-request, valid XML, well-formed XML,
security, etc.
- Determine which FAQ-entry is to send to Usenet
- Parse XML and extract the part we need
- Format message towards Usenet and fire it off
- Keep track of which FAQ is to be sent next time (number)
- Exit
Of course, this scenario is simpler as my initial proposal. It doesn't
touch anything about the way the current FAQ (or its
maintainance/structure/contents) works. It's just an add-on that is
responsible for the sending of one-FAQ-entry-a-day, and that's it.
I investigated the structure of http://www.jibbering.com/faq/index.xml,
this should be suitable; some questions:
- What does the "hash"-argument of the <URL> tag mean ?
- As Richard said, the code examples should be adapted in the data so
that it doesn't exceed 68 characters per line.
- It should be logical that each FAQ-entry would have its number in the
XML-data as well. This would be easier for Perl to remember which
FAQ-entry is to send the next day. Now Perl needs to invent its own
counting mechanism to determine which FAQ-entry is to send out. (which
shouldn't be a problem technically though, so generally this point is
only a side-remark). But when new entries are added in the future, this
might at some point result in a jump/reverse in the counting mechanism.
I don't know about the possibilities of something like
<FAQNR>4.22</FAQNR> or <CONTENT FAQNR="4.22">.
- As a professional pedantrist, it would be better practice to start
with <?xml version="1.0" encoding="UTF-8"?> , use some DTD/schema
info and do
<CHAPTER TITLE="Quick Answers" CHAPNR="4">
in stead of
<CONTENT TITLE="Quick Answers">
and perhaps
<FAQENTRY TITLE="What is ECMAScript?" ENTRYNR="7">
in stead of
<CONTENT TITLE="What is ECMAScript?">
to get the touch of real XML-ish structurisation. If you want to, I
could adapt the XML file that way. But the current XML structure is
suitable as well (these are optimisation issues only).
--
Bart
.
- Follow-Ups:
- Re: CLJ newsgroup FAQ
- From: Richard Cornford
- Re: CLJ newsgroup FAQ
- References:
- CLJ newsgroup FAQ
- From: Dr John Stockton
- Re: CLJ newsgroup FAQ
- From: Bart Van der Donck
- Re: CLJ newsgroup FAQ
- From: Dr John Stockton
- Re: CLJ newsgroup FAQ
- From: Bart Van der Donck
- Re: CLJ newsgroup FAQ
- From: Jim Ley
- CLJ newsgroup FAQ
- Prev by Date: Re: dynamically loading additional code from an url
- Next by Date: Re: Automation of comp.lang.javascript FAQ
- Previous by thread: Re: CLJ newsgroup FAQ
- Next by thread: Re: CLJ newsgroup FAQ
- Index(es):
Relevant Pages
|