From setok at fishpool.com Tue Feb 5 16:10:06 2008 From: setok at fishpool.com (Kristoffer Lawson) Date: Tue, 5 Feb 2008 17:10:06 +0200 Subject: [Xotcl] Pre-release of a new XOTcl-based template system Message-ID: <76F11D6F-70BD-4E27-B29E-E38BB25CB365@fishpool.com> Having become somewhat fed up with the Template Toolkit, built on Perl, I decided to create my own HTML/XML template system with the help of XOTcl. It takes a rather different approach to how to create templates, totally separating the actual HTML from the code that defines the contents. It's simple and easy, and there are things missing from it, but the model is something I'm pretty keen to explore further. Not an actual release, but you can fetch it to play around with it. Some more information plus a rant against Template Toolkit (and similar HTML template systems) at: http://blog.fishpool.fi/?p=17 / http://www.scred.com/ / http://www.fishpool.com/~setok/ From jade at volunteersolutions.org Wed Feb 6 16:18:08 2008 From: jade at volunteersolutions.org (Jade Rubick) Date: Wed, 6 Feb 2008 07:18:08 -0800 Subject: [Xotcl] Xotcl Digest, Vol 39, Issue 1 In-Reply-To: References: Message-ID: Kristoffer: Can you post some more information about how the API looks and works? Your blog post doesn't have much information on how it works or looks. Jade Jade Rubick Acting Chief Technology Officer United eWay jade.rubick at uwa.unitedway.org w: 503.285.4963 f: 707.671.1333 http://www.UNITEDeWAY.org On Feb 6, 2008, at 3:00 AM, xotcl-request at alice.wu-wien.ac.at wrote: From setok at fishpool.com Wed Feb 6 17:14:13 2008 From: setok at fishpool.com (Kristoffer Lawson) Date: Wed, 6 Feb 2008 18:14:13 +0200 Subject: [Xotcl] Illusion templates In-Reply-To: References:

Message-ID: <94357017-5BF8-4258-90E0-52B44E5E71E6@fishpool.com> On 6 Feb 2008, at 17:18, Jade Rubick wrote: > Kristoffer: > > Can you post some more information about how the API looks and works? > Your blog post doesn't have much information on how it works or looks. Sure. Well, for a start the package itself contains a couple of really simple examples, but I'll copy the simplest one here. Here is what the template looks like: Hello world. Let's try some magic:

%%colour%%

End of magic This template is for creating a list of colours. example.tcl, as specified on the first line, contains the logic for the content of objects and their amount. It is evaluated along with the template. This is what it looks like: Class ColourList -parameter { colour } foreach colour {red green blue} { set ob [ColourList new -colour $colour] ColourList lappend obOrder $ob } So basically it creates a single ColourList object for each colour we want. It also appends them to obOrder to display them in the right order (this phase is optional). The template uses the 'colour' method to receive values, which is here implemented with -parameter, but could easily be a method with some more logic instead. To get the result we simply do: set data [illusion::processTemplate example] Here's the result: Hello world. Let's try some magic:

red
green
blue

End of magic I also have added a feature to include other template files. I've kept it simple on purpose as I don't think a template file should be complex. The complexity should be in the Tcl. I do think this could lead to quite a versatile solution, though. / http://www.scred.com/ / http://www.fishpool.com/~setok/ From neumann at wu-wien.ac.at Sun Feb 24 16:23:18 2008 From: neumann at wu-wien.ac.at (Gustaf Neumann) Date: Sun, 24 Feb 2008 16:23:18 +0100 Subject: [Xotcl] Announcement: XOTcl 1.6.0 available Message-ID: <47C18BE6.6020200@wu-wien.ac.at> Dear XOTcl Community, XOTcl 1.6.0 is available. This release provides in short: - orthogonality improvements for introspection, - more introspection methods (one can now e.g. query, into which classes a mixin class is mixed into) and - performance improvements (some methods are significantly faster. e.g. by a factor of 1000 for larger systems). In addition, the XOTcl source code follows now closer the Tcl source code guidelines (e.g. variable naming), a few potential crashes have been fixed. See below for more details. Best regards -gustaf neumann =============================================== Announcing XOTcl 1.6.0 ************************* We are pleased to announce the availability of XOTcl 1.6.0. Major changes relative to 1.5.6 are: * Provide a uniform interface to the following info subcommands info superclass ?-closure? ?pattern? info subclass ?-closure? ?pattern? info instances ?-closure? ?pattern? info instmixin ?-closure? ?pattern? The new option "-closure" returns the transitive set of the relation (e.g. .. info subclass -closure) returns the subclasses and the subclasses of the subclasses. For "info instances -closure" the instances of the subclasses are returned as well. For more details, please see the language reference manual. In cases, where a pattern is specified, and the pattern contains meta-characters, a list of results is returned matching the pattern (like "string match"). When no matching value is found, an empty list is returned. In cases, where a pattern is specified, and the pattern contains no meta-characters, a single value is returned corresponding to the specified value. The pattern is used to lookup an object or class, such it is not necessary to provide fully qualified names). if there is no match, empty is returned. Previously, "info superclass" and "info subclass" returned a boolean value and performed always a transitive search. Returning "" is more consistent and more in line with Tcl. Note that " info superclass -closure" is a replacement for " info heritage", and " info instances -closure" is a replacement for " allinstances". The old commands will be marked as deprecated in the near future. Please note, that the behavior of the match pattern has changed and is therefor not completely compatible with prior versions. * New info subcommands: info instmixinof ?-closure? ?pattern? info mixinof ?pattern? These info subcommands are used to determine, into which classes a mixin class was mixed into. These inverse functions of mixin and instmixin are used as well internally. These functions help to speed certain operations (e.g. superclass, or registering a mixin class) up by a factor of 1000 (!) or more, when a large number of objects exist. This functionality was primarily implemented by Martin Matuska. Many thanks! * Made the behavior "pattern" in the following calls identical concerning wild cards and object lookups