Assembler/Compiler testing
in [ASM]
|
From: o//annabee on 13 Mar 2006 04:27 P? Mon, 13 Mar 2006 05:33:18 +0100, skrev <Chewy509(a)austarnet.com.au>: > > \\o//annabee wrote: >> Chewy509 wrote: >> >> >> Here is a link to what I work on : < http://szmyggenpv.com/ > >> > >> > Hmm... not valid HTML. And oh my eyes, please stop the bleeding! >> > (Doesn't pass the w3c HTML validator on any HTML level)... >> >> :))) Good! The validator seems to do its work, indeed. I am much >> positivly >> surpriced with it. Good tool indeed. > > So I take it, you'll be updating your website shortly to be nothing but > valid HTML? Of cource not. I ment that this validator is great at limiting people conserned with standards from acessing the site. I think it works to perfection. They dont want to visit a site that isnt within standards, so using this validator, they can protect themselfes from this immense annoyance, and I on the other hand can avoid having them visiting. So I think this wonderful validator does double duty. Its nothing short of Brilliant if you'd ask me. :)) > PS. As a token of good will, I'm more than happy to update your website > for you, so that's HTML v4.01 Strict w/CSS and also give it a more > modern look at the same time... Once at that point, it shouldn't be too > hard for you to maintain yourself. Thank you very much for this offer. I will have to refuse. Because I am _allergic_ to standards. Sorry. > PPS. That is a serious offer! (I can't stand websites that look like > they're done by a 2 year old using frontpage). :))) Yes, I guess. But at least it aint trying to cover up for it. > > -- > Darran (aka Chewy509) brought to you by Google Groups! > --
From: o//annabee on 13 Mar 2006 04:30 P? Mon, 13 Mar 2006 07:49:51 +0100, skrev sevagK <kahlinor(a)yahoo.com>: > > Chewy509(a)austarnet.com.au wrote: > >> PS. As a token of good will, I'm more than happy to update your website >> for you, so that's HTML v4.01 Strict w/CSS and also give it a more >> modern look at the same time... Once at that point, it shouldn't be too >> hard for you to maintain yourself. >> >> PPS. That is a serious offer! (I can't stand websites that look like >> they're done by a 2 year old using frontpage). >> >> -- >> Darran (aka Chewy509) brought to you by Google Groups! > > > Seems okay to me, no distractions and straight to the point. > > > Wannabee: Your current project looks interesting. -- crashed on my > system after closing the About box. Now I can't run it at all anymore. > Freezes everytime. You might want to look into it. (WinXP). Thanks SevagK. :) > > -sevag.k > www.geocities.com/kahlinor > -- Bansai!
From: Frank Kotler on 13 Mar 2006 06:11 Evenbit wrote: .... > The only defense I can think of is to say that it is part > of the ASM culture to "purposely" employ a "circa '95 theme" in proud > defiance of modern eye-candy. :) Right on! Best, Frank
From: Dragontamer on 13 Mar 2006 08:59 o//annabee wrote: > På Mon, 13 Mar 2006 05:33:18 +0100, skrev <Chewy509(a)austarnet.com.au>: > > PS. As a token of good will, I'm more than happy to update your website > > for you, so that's HTML v4.01 Strict w/CSS and also give it a more > > modern look at the same time... Once at that point, it shouldn't be too > > hard for you to maintain yourself. > > Thank you very much for this offer. I will have to refuse. > Because I am _allergic_ to standards. Sorry. Such as the GPL standard I assume? IMO, that is why I don't like the GPL. If you aren't part of the "GPL standard", aka, have the GPL license, you lose out on benifits. There is something about this "join us" that I don't like; it is like an enforced standard of sorts. *sorry for the sudden change of topic in this off topic topic :)* --Dragontamer
From: randyhyde@earthlink.net on 13 Mar 2006 15:33
Chewy509(a)austarnet.com.au wrote: > > Hi Randy, > > I'm doing something similar, just a basic script that compiles all the > examples and compares to the previous version, as well as additional > files that check for error/fail conditions and out-of-band syntax (eg > symbols are the too large to fit into the internal buffers (more than > 256 characters), nesting too many files (32 files deep), or overly > complex preprocessor nesting (256 levels deep of #ifdef/#else/#endif), > etc). Obvisously, like yourself each fail-condition is a separate file. Sounds reasonable. :-) > > But the one thing that has helped enormously, at least in my case is > rewritting the compiler in it's own language... (As this has tested > roughly 99% of the code path, and dealing with a cross-platform product > it gets a good workout). I'm assuming your finding that experience > similar with rewritting HLA v2 in HLA? I've definitely found *some* bugs by doing this. But keep in mind that HLA v1.x existed a long time before I started doing serious work on HLA v2.0. By then, lots of different people were using it and a lot of the easy bugs had already been encountered. As for exercising the code path, HLA is such a rich language that I don't even come close with HLA v2.0's source code. For example, I don't use many object-oriented programming features in HLA v2.0. Recently, I've done some work on a new terminal emulation class for the HLA standard library and the HLA Standard Template Library, both of which use object-oriented programming facilities, and I've discovered some problems there (Sevag has also been doing some OOP stuff, and has discovered a few little problems). In any case, that demonstrates that you won't get code coverage by simply rewriting HLA in itself. > > The thing was, I was wondering if there was a better or alternate way > of testing... Well, code coverage is the *minimum* that the IEEE suggests for "professional quality code". There are lots of other testing mechanisms you can use above and beyond code coverage. A good set of functional tests are needed to complement the things that code coverage (white box test data) misses. But the main thing is to have a formal test suite with a good set of regression tests you can run whenever you do a milestone build. > > At least for my purpose, there doesn't appear to be. Thanks to everyone > for their feedback! :) > > PS. Good to see Rene is back in fine form! ;) :-( Cheers, Randy Hyde |