Links for 31-Jul
Coding without comments -- yes, it is probably a good thing.
.Net4/C#4 and dynamic language features -- JavaScript envy? Oh, yeah!
Diary, commentary, reviews, snippets to preserve on-line
Coding without comments -- yes, it is probably a good thing.
.Net4/C#4 and dynamic language features -- JavaScript envy? Oh, yeah!
Posted at 09:32 No comments :
Labels: .Net , C# , dynamic languages , PowerShell , software practice
Humidity gathered all weekend. Today it was very sticky indoors, but quite pleasant when cycling (getting some sort of air movement). Went for a long spin, trying not to get caught up in the stream of London-to-Cambridge riders.
Later this afternoon, I spent a while staining the bits a cheap and cheerful pine book-case from Homebase, dripping sweat copiously in the still air in the garden. Putting the bits in the greenhouse so they could bake dry was like stepping into a sauna.
Meanwhile the cats were just lying around half-melted on the patio, watching me in a bemused sort of way.
Fell exhausted onto bed a bit before 18:00, in time to catch a weather forecast for heavy showers. Within minutes, the heavens opened, absolutely hammering down, putting about an inch of water in a tray I'd left on the patio. It eased off to merely heavy rain after about 15 minutes, but kept on going until dark.
Early anniversary dinner at the Willow Tree in Bourn -- a lovely meal, despite the couple of glitches (one of the dishes running out and confusion that engendered), and good conversation, at one point turning to an anecdote about an old woman whose eyes and teeth were the only bits not being ravaged by time. So of course I remarked that I was having pretty much the opposite, but thankfully the teeth seemed to have all been repaired now, and yes, I would have the ice-cream selection for dessert.
And indeed the teeth were not sensitive to cold. Just that a chunk fell off one of them.
Monday I had a temporary crown fitted with permanent cement. Tuesday I got a slot to get an impression taken, so the mildly uncomfortable temp had to be drilled off. And after some inspection, the space was deemed to be very tight, so it's going to be a case of more bling for the grin in a couple of weeks' time. At least the new temporary is so far entirely happy rather than twinging when it gently contacted the tooth below.
linking everywhere -- PoC for HTML5 style href="<url>" attributes on everything.
Posted at 16:53 No comments :
Labels: dynamic languages , F# , Javascript , Python , Ruby , web design
using + lambda/delegate = problem.
Egoless programming, intent programming.
Erlang is gaining momentum for massively concurrent applications.
JRuby 1.1.3 -- with enough enhancements I'd've called it 1.2!
jQuery resources and widgets - toast, progress-bar,
Posted at 11:01 No comments :
Labels: C# , concurrency , functional programming , Javascript , Ruby , software practice , web design
Functional C# -- Part 5 and last, the Match operator.
IronPython 1.1.2RC1 and Jython 2.5 first alpha release.
F# -- semi-formal spec
oMeta (and oMeta#) -- Object-Oriented Language for Pattern Matching i.e. a lexer/parser/... kit
Posted at 15:06 No comments :
Labels: .Net , C# , dynamic languages , F# , functional programming , IronPython , jython
Nester -- a C# version of the Jester code mutator.“Why just think your tests are good when you can know for sure?”
LinFu -- interesting .Net meta-programming library -- including support for declarative Design by Contract.
CLR security team site on Codeplex.
Zermatt -- MSFT claims based identity system in public beta.
Aspects of Functional Programming in C# Presentation and Code.
Practical Concurrency -- Read/Write cache.
Avoiding having to replicate .snk files across projects.
Good use of anonymous delegates.
Cyclone -- safe 'C'.
British Standards in Web Accessibility is being updated.
Strong-naming legacy assemblies -- ilasm hackery.
Google Protocol Buffers -- now open sourced.
Posted at 12:41 No comments :
Labels: .Net , accessibility , C# , Computer Security , concurrency , F# , functional programming , Identity Management , Javascript
Spartan Programming (Is this madness? No, it's Sparta!)
Replacing [ExpectedException(typeof(T))] with MyAssert.Throws<T> -- Checking the right line threw is less of an issue when you combine test runs with code-coverage, but even there this technique still helps as it moves all the post-Exception uncovered coded points into one central location for you.
Powershell Script blog; also PowerShell repository and code signing.
PowerTab for v2 CTP2 -- intellisense-like tab completion.
Bash vs PoSh -- good words from the Linux world.
PoSh + IronRuby -- custom PSHost example
1.To(3) -- Ruby style iterators in C# 3.0
Stream as Iterator -- useful C# idiom.
WCF Gotcha -- abuse of Dispose().
nUnit Attach -- Dev Studio macro.
Windows SDK tools -- exposing the under-documented
Depender -- coupling analysis for testability.;
About a year ago, I revamped my main static site, clearing stuff that duplicated blogging type content into blogs, and changing the navigation to use accordion folds. That was using Prototype for a basis and Rico for the accordion (I looked at Dojo, but I didn't care for their use of custom attributes which baked a lot of behaviour into the HTML as well as making it fail validation).
After Google announced that they were hosting the main Ajax libraries, I had been meaning to move from having local copies of Prototype. And then I saw two articles in quick succession, only on A List Apart, about binding vs closures, which made me waver towards jQuery on purism grounds, and then via Christopher Steen's link blog, an introduction to jQuery UI - which showed that jQuery accordions require a lot less structural markup and hand-rolled class assignments even than Rico's. So I've made the change, and was delighted to find that, not only was the deadweight on pages was reduced, but I could just add mouseover unfolding as well.
There is a jQuery rounded corners plug-in out there, but I've not managed to get it to work in conjunction with a definition-list based accordion, so for the moment, I'm cheating with -moz-border-radius and leaving it at that.
One word of warning for people serving XHTML as application/xhtml+xml, though. In Google's shortcut
<script src="..."></script> <script> // Load jQuery google.load("jquery", "1"); …
where the src is http://www.google.com/jsapi, the loading is done by injecting a script element via document.write()
, so fails utterly -- I had to fall back to directly loading from a fixed URL, in this case, http://ajax.googleapis.com/ajax/libs/jquery/1.2.6/jquery.min.js for the current jQuery release.
IIS7 PowerShell provider is at TP2.
PowerShell - a gentle introduction in two parts with some scripts to simplify the code-signing business.
Massively multi-core processors -- threat or promise?
JavaScript binding -- an overview.
Slideshow Gem offers S5 support.
Ruby Shoes nifty deployment model -- one-click deployment for your Ruby apps.
.Net concurrency -- learning from Erlang.
psake -- PowerShell does make.
Posted at 09:27 No comments :
Labels: .Net , concurrency , functional programming , MSFT utilities , PowerShell
Previous instalments at Life before Blogging.
Forum Administrator : EvaGeeks.org — An Evangelion Fan Community
Copyright © Steve Gilham, 2003-2013
This work is licensed under a Creative Commons License.
Code snippets released under the WTFPL