|
Chris,
Why is it that when I modify an article, sometimes (perhaps, usually) the code snippets will lose their VS-like colorings? Eventually they get the colors back, but I was wondering if there is any way that you could make it always "re-colorize" the snippets when the article is changed.
I know it's a minor point, but I like my code snippets to have that VS look.
BTW - any plans for XAML snippet colorization? Please? 
Thanks!
:josh:
My WPF Blog[ ^]
We are what we repeatedly do. Excellence then, is not an act, but a habit. - Aristotle
|
|
|
|
|
The coloriser is broken on our IIS6 servers but works fine on the IIS5 servers. I've torn my hair out trying to find out why and have just had to move on with other stuff. It's all C++, too, which is doubly weird. web2.codeproject.com is working fine so I tend to use that one.
The old coloriser has some minor issues and doesn't handle ASP.NET well. We have a new one that's fully .NET that is awesome. I'm tossing up between running a background task to automatically recolorize all new articles after they are posted or trying to push forward the submission wizard part of the rewrite to get this part fixed.
The only problem with the background process is that if it kills the text (ie we screw up with the regular expressions we use) then there's no instant feedback to the author. I will put in place checks to ensure that at the worst it merely doesn't colorise instead of, say, ripping the guts out of your article.
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Thanks for the insight.
Chris Maunder wrote: I will put in place checks to ensure that at the worst it merely doesn't colorise instead of, say, ripping the guts out of your article.
Sounds good to me. 
:josh:
My WPF Blog[ ^]
We are what we repeatedly do. Excellence then, is not an act, but a habit. - Aristotle
|
|
|
|
|
Josh Smith wrote: Chris Maunder wrote:
I will put in place checks to ensure that at the worst it merely doesn't colorise instead of, say, ripping the guts out of your article.
Sounds good to me.
Interesting use of the word "good". I would have thought "good" would be reserved for a statement like, "I will put in place a coloriser that works properly all the time."
|
|
|
|
|
The coloriser hasn't been fully tested. It's beta. I can hold off installing it until I've had a chance to fully test it, or I can Get The Job Done and catch any mistakes it makes along the way.
So how about this: Why don't you write me one using your compiler. It should be able to take malformed HTML, find all the PRE and CODE tags, look up the lang attribute and colorise according to the language specified. It also needs to be fully extendable so we can add other languages, and it must be able to understand code such as you would find in ASP.NET pages where the language switches between HTML, Javascript, C#/VB.NET and back again.
Do that and I will give you a month's Showcase space for an article on your PEC.
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Chris Maunder wrote: I can hold off installing it until I've had a chance to fully test it, or I can Get The Job Done and catch any mistakes it makes along the way.
The "job" of a programmer includes testing and debugging. Yes, by all means, "hold off installing it until [you've] had a chance to fully test it." Or release it to a subset of the members who are willing to help you test it. But don't put it on the production site until you're sure it works. Releasing it, to the general public, when it has not been thoroughly tested is not "Getting the Job Done". And please don't say "this is a community site so different rules apply". Nonsense. This is commercial site - certainly from your perspective.
Chris Maunder wrote: and catch any mistakes it makes along the way
You're just making more work for yourself here. And you've got enough already. All that code you have to add to make sure articles don't get mangled is unnecessary in a test environment.
Chris Maunder wrote: So how about this: Why don't you write me one using your compiler. It should be able to take malformed HTML, find all the PRE and CODE tags, look up the lang attribute and colorise according to the language specified. It also needs to be fully extendable so we can add other languages, and it must be able to understand code such as you would find in ASP.NET pages where the language switches between HTML, Javascript, C#/VB.NET and back again.
I'm quite sure you're missing my point. When one is unable to do a "good" job at something, it's either because of incompetence or because the thing simply can't be done with the resources available - that is, one has "bitten off more than one can chew". Clearly, you're not incompetent, so I suggest (as I have many times before) that the whole thing has just gotten out of hand; it's all too, too complicated. There is no way you and your staff will be able to keep up with writing colorisers for every new syntax that the Microsoft dreams up. You may be fool enough to attempt such a thing; I'm not. Our Plain English coloriser works all the time, because (1) it's simple and (2) because we're not at the mercy of Microsoft regarding syntax - as you are. The best thing you can do is take a simpler approach, and try to move the burden of colorising elsewhere.
In this particular case, I suggest that you save yourself a great deal of time and energy by simply having contributors submit their articles as PDFs. That way, they can author using any tool of their choice, their work will appear exactly as they intended it, every article will be printable in a wysiwyg format, and all you will have to provide is posting and search capability which are within reach of your available resources. The burden of colorising would then be forever removed from your shoulders. Not to mention the fact that this approach would eliminate the temptation to include Intellitxt!
|
|
|
|
|
The Grand Negus wrote: In this particular case, I suggest that you save yourself a great deal of time and energy by simply having contributors submit their articles as PDFs.
I love PDFs just as much as the next guy - for eBooks! If we start moving everything to PDFs rather than hypertext for the Web we may as well go the whole nine yards and dump HTTP in favor of Gopher with its obviously superior format.
|
|
|
|
|
Jeremy Falcon wrote: I love PDFs just as much as the next guy - for eBooks! If we start moving everything to PDFs rather than hypertext for the Web we may as well go the whole nine yards and dump HTTP in favor of Gopher with its obviously superior format.
Perhaps something like that is called for.
|
|
|
|
|
The Grand Negus wrote: Perhaps something like that is called for.
We already have it in most major cities around the world. It's called the library. Because moving to that would be no different except one is on paper and the other has a splash screen by Adobe.
|
|
|
|
|
|
dan neely wrote: For people with visual impairment or really small screens it's even worse.
You can enlarge the view in most readers.
dan neely wrote: PDF's rigid control over formating makes it the jackbooted thug of document formats.
Well, to each his own. The only thing I don't particularly care for with PDFs is Adobe's reader. I think the format is a-ok.
Anyway, I do like eBooks which is a lot like a manual. I wouldn't want to have them replace the Web at all though.
|
|
|
|
|
Jeremy Falcon wrote: dan neely wrote:
For people with visual impairment or really small screens it's even worse.
You can enlarge the view in most readers.
and have to use horizontal scrolling to read every single line. 
--
Rules of thumb should not be taken for the whole hand.
|
|
|
|
|
You've gone from mildy amusing and arrogant... to absolutely rude and arrogant.
You have done nothing but complain about this site for the last week or two. What, to everyone else is a minor bugbear or annoyance, to you is some major showstopper that we "just shouldnt tolerate from a premier website!!!". Balls. We all manage to use this site every day and it works just fine 99.9% of the time.
Let me spell this out as your intelligence level seems to have droped sub-Dan and possibly sub-Chuckles.
You. Are. The. Only. One. Complaining. Sensible. Readers. Make. Suggestions. To. Improve. The. Site. They. Never. Complain. That. The. Site. Is. Unusable.
Gerry, you cuss this site all day and insult its readership. Just go away. Begone. Don't come back this time. Life will be sweeet for you (no more CP bugs to upset you) and for all of us (no more osmosi-clan to upset us).
|
|
|
|
|
J4amieC wrote: The. Site. Is. Unusable.
But. Not. Working. Properly.
|
|
|
|
|
The Grand Negus wrote: Interesting use of the word "good". I would have thought "good" would be reserved for a statement like, "I will put in place a coloriser that works properly all the time."
Do you pay a membership fee to CodeProject? No.
Are they contractually obligated to provide code snippet colorization? No.
Should beggars be choosers? No.
:josh:
My WPF Blog[ ^]
We are what we repeatedly do. Excellence then, is not an act, but a habit. - Aristotle
|
|
|
|
|
Josh Smith wrote: Do you pay a membership fee to CodeProject? No.
Are they contractually obligated to provide code snippet colorization? No.
Should beggars be choosers? No.
Should programs work? Not work for me, but work simply as a matter of principle?
|
|
|
|
|
The Grand Negus wrote: Should programs work? Not work for me, but work simply as a matter of principle?
So every last program you've ever written never had a bug in it, eh? If that's the case, you may wish to speak to God and let him know where he went wrong in making the rest of the world.
Really, even if you had semi-valid points, you totally do yourself more harm than good by coming off as a sour apple. None of us here have really seen you do anything useful - just complain. In my book, programmers solve problems and not and sit and whine and complain about them. That's what users do.
And lastly, this ain't the soapbox man. At least try to keep the discussions on topic rather than infest yet another message board with your drivel. Bugs or no bugs, CP is a great resource and you know it; otherwise, you would've left already. Why don't you do something weird and start talking about its strengths for a change.
|
|
|
|
|
Jeremy Falcon wrote: So every last program you've ever written never had a bug in it, eh?
Don't be silly; nobody's perfect. But we don't release programs with known bugs. And we do consider testing and debugging part of "Getting the Job Done".
Jeremy Falcon wrote: At least try to keep the discussions on topic
The topic, brought up by the poster I initially replied to, was "workarounds". My thought was twofold. First, that "workarounds" are generally not good because they take the pressure off actual fixes. And secondly, that we shouldn't need "workarounds" for the most basic of operations - like add, change, and delete of profile information. After seven years.
|
|
|
|
|
The Grand Negus wrote: Don't be silly; nobody's perfect. But we don't release programs with known bugs. And we do consider testing and debugging part of "Getting the Job Done".
So you assume your testing is perfect then? I've known testers (for a larger company) that were awesome in finding my bugs. They were better at it than I, and yet still a few bugs would slip through the cracks. That's what makes them bugs.
The Grand Negus wrote: And secondly, that we shouldn't need "workarounds" for the most basic of operations - like add, change, and delete of profile information. After seven years.
Which is my point. You offer nothing much but criticism. Oh sure, there's a time and place for that, but if that's all you do than it's a problem that's no better or worse than refusing to admit the truth.
Really man, to me it seems like you're wasting your talents on being bitter.
|
|
|
|
|
Jeremy Falcon wrote: So you assume your testing is perfect then?
No. Just as good as we know how to do.
Jeremy Falcon wrote: I've known testers (for a larger company) that were awesome in finding my bugs. They were better at it than I, and yet still a few bugs would slip through the cracks. That's what makes them bugs.
But the bugs that "slip through the cracks" are not known bugs, which is the topic here.
Jeremy Falcon wrote: You offer nothing much but criticism.
Not so. I offer constructive criticism which is significantly different. When Chris approached me directly regarding the colorising problem (above) I didn't just complain, I offered a reasonable analysis of the problem and suggested a workable solution. One that would make the thing better for all and free up a great deal of his time in the process.
|
|
|
|
|
The Grand Negus wrote: But the bugs that "slip through the cracks" are not known bugs, which is the topic here.
Ok, but none of us really know Chris' schedule or priorities. Say what you will about that, but until we're in his position all we can do is judge and second-guess. Regardless, his priorities don't necessarily affect (directly) the quality of the software in itself.
In the real world things do cost money. That includes time and programmers. Maybe he's making a trade-off depending on the circumstances. I don't know. Oh jeeze, now I sound like I'm defensive. Whatever, my point is we can't always assume we know.
The Grand Negus wrote: Not so. I offer constructive criticism which is significantly different.
But it is criticism nonetheless. Constructive criticism in itself isn't bad, but that's all you do. It's like all day long you look to speak of CP's flaws while avoiding increasing the merits in your own work. And yet you're surprised to be called a sour apple? Too much of anything is bad, and I'm sure we can both agree on that.
The Grand Negus wrote: suggested a workable solution
What you suggested was akin to undermining the entire structure of the Web rather than actually addressing his issue. Which is what I'm getting at. Your idea of solutions revolve around sweeping generalizations that don't really help, all while you criticize and offer no real work/contribution of your own.
For that matter, I think an extensible colorizer for future syntax is way more doable than writing a PDF writer (you'd have to let authors write for free or else most wouldn't) for every platform under the sun. Either way, I'd rather have to colorize my HTML manually before I was forced to make everything online a bunch of PDFs.
So, in essence what you suggested doing was not only more complicated it's extremely impractical.
|
|
|
|
|
The Grand Negus wrote: But we don't release programs with known bugs. And we do consider testing and debugging part of "Getting the Job Done".
Hint: plain english Claude Monet demo + no internet connection = crash + burn
Not a known bug? Not a very good tester!
The Grand Negus wrote: most basic of operations
Like a valid internet connection in a program that requires an internet connection? Like error reporting to the user when something goes wrong?
Your arguments are about as watertight as a pair of fishnet tights.
|
|
|
|
|
J4amieC wrote: Hint: plain english Claude Monet demo + no internet connection = crash + burn
It doesn't crash and burn; it displays an appropriate error message.
|
|
|
|
|
Aaaah but it didnt when you released it did it mr nag us.
|
|
|
|
|
J4amieC wrote: Aaaah but it didnt when you released it did it mr nag us.
Simply not true. The current version is the same as the original. The alleged error was mis-reported.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
