Difference between revisions of "XmlWiki"
From Organic Design wiki
Line 22: | Line 22: | ||
# INIT | # INIT | ||
# - do admin and exit if running directly | # - do admin and exit if running directly | ||
− | # - get wiki | + | # - include geshi |
− | # - get article | + | # - get wiki & request environment |
− | # - | + | # - get article (as an object if possible) |
− | # - build env (if action=raw, env is empty | + | # - get user-info |
+ | # - get object-specific info (perms, doc-root, doc-type) | ||
+ | # - build env (if action=raw, env is empty) | ||
# INPUT | # INPUT | ||
− | # - | + | # - deny access if !writable |
− | # - | + | # - currently not doing anything else |
# OUTPUT | # OUTPUT | ||
# - trap MediaWiki output | # - trap MediaWiki output | ||
− | # - extract | + | # - extract links (not used yet) |
− | # - | + | # - apply security |
− | # | + | # - transform for raw, view, preview |
− | + | # - if saving, do publishing | |
− | # - if saving, | + | # - render output & messages |
− | |||
− | # | ||
# TRANSFORM | # TRANSFORM | ||
− | # - | + | # - if article is an object, |
− | # - | + | # - process transforms if doctype is one of ours, |
+ | # - validate and apply XSLT if not | ||
+ | # - apply geshi on article | ||
# TRANSFORMS | # TRANSFORMS | ||
# - document transforms | # - document transforms | ||
Line 46: | Line 48: | ||
# - layout transforms | # - layout transforms | ||
# - database table transforms | # - database table transforms | ||
+ | # - transform object to php-nav-tree | ||
# FUNCTIONS | # FUNCTIONS | ||
# - get article | # - get article |
Revision as of 09:47, 2 June 2005
File:Xmlwiki old.jpg Old logo: File:Xmlwiki wht.jpg
A Wiki of Objects
xmlWiki is a modification of the index.php page offering new features which are all maintained from editing normal wiki articles.
- Objects can be objects stored in XML
- A permissions structure offering basic read/write control over articles with users and groups.
- PHP Functions can be created and maintained by developers and can be dynamically applied to articles before rendering.
- Articles can made to automatically publish remotely over HTTP or FTP when edits are saved.
If the browser is fully XML-capable, then requested articles can be returned as XML for the client to transform with an XSLT and render. If not, xmlWiki can transform the document before returning it to the client.
The structure of xmlWiki
The xmlWiki layer is composed of a single PHP script called xmlwiki.php which, when called directly hooks or unhooks itself from three places in the MediaWiki index.php script.
The eneral structure of the xmlwiki.php script is as follows:
Current script here <php>
- INIT
- - do admin and exit if running directly
- - include geshi
- - get wiki & request environment
- - get article (as an object if possible)
- - get user-info
- - get object-specific info (perms, doc-root, doc-type)
- - build env (if action=raw, env is empty)
- INPUT
- - deny access if !writable
- - currently not doing anything else
- OUTPUT
- - trap MediaWiki output
- - extract links (not used yet)
- - apply security
- - transform for raw, view, preview
- - if saving, do publishing
- - render output & messages
- TRANSFORM
- - if article is an object,
- - process transforms if doctype is one of ours,
- - validate and apply XSLT if not
- - apply geshi on article
- TRANSFORMS
- - document transforms
- - image transforms
- - layout transforms
- - database table transforms
- - transform object to php-nav-tree
- FUNCTIONS
- - get article
- - etxract hrefs from text
- - message/log
- - geshi syntax highlighting
- ADMIN
- - enable/disable patch
</php>