|
JimmyRopes wrote: javascript errors
They aren't errors. But just 'Warning' messages. It can be safely ignored.
|
|
|
|
|
Yes they are warning messages from the browser but they are caused by a scripting error.
There seems to be some misunderstanding here.
There is a popular belief that they are style element warnings and not script errors.
When they appear in a style sheet they are ignored just like any other style the browser doesn't understand. That is the CSS paradigm.
The only time these warnings appear is when they are in a script, and when they do they cause the script to function improperly. If the rendered page is OK without the desired scripting action then there is no problem.
In my case I needed the scripting action to operate properly to render the page as desired. Once it was pointed out to me where my scripting error was (thanks to Guffa) then not only didn't I get any more warnings but my script also operated properly.
In the course of diagnosing my scripting error I noticed that CP was getting similar warnings. I just brought it to the attention of the staff to be helpful -- no good deed goes unpunished.
In my case I fixed a malfunctioning script and the warnings along with the problem went away. I call that a scripting error, or, more precisely, warnings as the result of a scripting error.
|
|
|
|
|
I see these warnings all the time but I still don't consider them a scripting error since it's not the javascript engine that's throwing a wobbly. It's malformed (to Firefox) CSS. It's interesting that these malformed and unrecognised CSS attributes are only showing up when there is javascript present, but javascript certainly doesn't need to be present for Firefox to still not like the attributes.
A small subtlety here is that there are, kind of, script errors in the CSS. I'm using the IE-only eval method to get around some weirdness and this is, in fact script. But only in IE. But in IE this script in the CSS is well-formed so it's not a script error. Hence there are no script errors in the CSS. (tada!)
I know I should do some funky browser specific fiddly things to try and hide all the bits that each browser doesn't like but it doesn't actually affect the operation of either browser (except to actually make the requisite browser do what I'm telling it to do. Sometimes).
But: I definitely appreciate the heads up.
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
In all the forums, I see people asking questions and surrounding their code snippets with the <code> tags, thinking it will preserve their formatting. Unfortunately, that code button is misleading; it doesn't preserve whitespace and formatting. Without formatting, code snippets are more difficult to decipher, causing some potential answer-givers to move on, resulting in fewer answered questions.
The code button should actually insert <pre> tags instead, which do preserve formatting.
<br />
<br />
|
|
|
|
|
I concur something needs improving; I'm not sure "var" would help tho.
CODE tag is fine for "quoting" one or a few technical terms on a single line
PRE tag is to be used for all code snippets that span two or more lines
it is a pitty one has to manipulate code inside a PRE tag:
- empty lines must get a space added, or they are thrown out
- less-than signs often need to be protected from the hungry HTML beast
For those unfamiliar to HTML the name of the PRE tag and the anomalies listed above
are surprising, and if possible should be dealt with automatically (i.e.
always accept CODE, forget PRE tags on input; handle them as if they were PRE
if multi-line, and CODE if single-line)
And (a plug is coming up now) if Internet Explorer makes it hard yo copy/paste
successfully, read my CodeRescue article !
Luc Pattyn [Forum Guidelines] [My Articles]
this weeks tips:
- make Visual display line numbers: Tools/Options/TextEditor/...
- show exceptions with ToString() to see all information
- before you ask a question here, search CodeProject, then Google
|
|
|
|
|
At least in Lounge, where we discourage programming questions, we can call it as differently emphasizing Bold and Bright. Wouldn't that solve your need?
|
|
|
|
|
I just got hit by a 1 for my new article[^] and feel quite a bit cheated.
The average was 5 with 2 votes and then someone comes along and posts a 1 (without comment!) dropping me to 3.4 and straight to the bottom of the category.
I understand that no voting system is perfect and that it's difficult to solve all kinds of problems but such an obvious abuse should just not be possible.
What annoys me most is that this is the second time this happened to me (just not so obvious). Of course these spoiler votes get evened out over-time but as I don't expect more than 10 votes for the article I can be happy if the article raises to something above 4 in the years to come. Incidents like this take all the fun and ambition out of being a CPian .
I think that these extreme low/high samples should be kicked out of the calculation but then I'm not very good at statistics so I wouldn't know how to do it properly. But I still think something should be done....
btw: see also the Voting - perhaps not?[^] thread.
|
|
|
|
|
Welcome to CodeProject.
"Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass..." - Dale Earnhardt, 1997 ----- "...the staggering layers of obscenity in your statement make it a work of art on so many levels." - Jason Jystad, 10/26/2001
|
|
|
|
|
Hi... i would like to ask about the meanings of terms found in the profile realted to function timing created for any Visual C++ with OpenGL program, these terms are: Child Time and Hit Count, and in what we can use these terms?
regards
maria
|
|
|
|
|
I'd like to suggest that at the top of each article category page (e.g., http://www.codeproject.com/treectrl), there should be a list of the last ten updates made in that category, just like on the main page.
|
|
|
|
|
Excellent idea.
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
I would suggest if CP could look into providing the following features for the articles:
(1) Surfer statistics
(2) Download statistics
Perhaps something like SiteMeter does. That way it would be helpful for both CP (to calculate the bandwidth used etc and manage effectively) and also for the Article Author to fine tune it according to the region where the article is being searched and reached widely.
Or there can be provision in CP to enable ArticleAuthor to embed SiteMeter like code. Isn't it?
|
|
|
|
|
Vasudevan Deepak Kumar wrote: Download statistics
I have suggested this already and as far as I know it has been added to to-do-list
|
|
|
|
|
I actually made a quick search in 'Search' link at the right. But for some reason, since it couldn't pop out, I just thought of sharing out this suggestion.
Was it a relatively old suggestion and hence I think the default period could not retrieve the post?
|
|
|
|
|
When an ASP Error or Duplicate Post error occurs, 'Try Again' link is shown. But it just shows 'forums.asp'. Instead, if we could link the full HTTP_REFERER including the querystring, it would facilitate getting back to the original page, without duplicate clicks.
|
|
|
|
|
When a message is bieng voted, there is an AJAX call being sent and then a message to confirm this is updated in the little SPAN. But the existing vote count still does not get incremented and this increment happens only page refresh. When you return the success message, this count also can be automatically incremented. Isn't it?
|
|
|
|
|
Yes
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Hi,
I just received the 06 august newsletter; it awards most popular article an article with
bad formatting, that I am not even willing to read since that would require horizontal
scrolling for every line of text (although my screen is 1920 pixels wide !).
I strongly suggest you exclude badly formatted articles from every "most ..." list,
and make this clearly known to all; this may help rectify the situation. Awarding and
mentioning them certainly does not.
Luc Pattyn [Forum Guidelines] [My Articles]
this weeks tips:
- make Visual display line numbers: Tools/Options/TextEditor/AllLanguages/General
- show exceptions with ToString() to see all information
- before you ask a question here, search CodeProject, then Google
|
|
|
|
|
At the top of each article is a link "Report Broken Article". You can use this to report the article and we will then make sure the article is cleaned up and readable.
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Hi Chris,
I am aware of that.
I would use it only if something fails that I would not blame the author for.
Things the author did wrong I can always report to him with a comment message.
I don't like to use the "broken article link" for formatting problems; it would be an
invitation to cheating the editing queue: publish an unedited and unformatted article,
report it as broken, get it edited right away, rather than waiting your turn in the queue.
In my opinion people should take care of their own formatting, if they don't care
about it, why would I trust the content of the article or message ? And why would
they deserve being awarded anything ?
{added: here is what I suggest: when an unedited article is submitted, check it
for a resaonable ratio of P-tags versus file length; if insufficient, don't publish
it at all, just insert it at the end of the editor's queue. They will learn it
eventually.]
Luc Pattyn [Forum Guidelines] [My Articles]
this weeks tips:
- make Visual display line numbers: Tools/Options/TextEditor/AllLanguages/General
- show exceptions with ToString() to see all information
- before you ask a question here, search CodeProject, then Google
|
|
|
|
|
Luc Pattyn wrote: cheating the editing queue: publish an unedited and unformatted article,
report it as broken, get it edited right away, rather than waiting your turn in the queue
Bad way to bypass the queue and get preempt others who are waiting in the queue. But we can block this link from being used by the author (or his IP) like in Vote. Or we can ask for a mandatory reasons. Either of the reasons or both combined should stall the problem or at least 'check' the problem and 'arrest'/'contain' it right?
|
|
|
|
|
|
When you login to CodeProject, it would be good to show 'Last Known Best Login' or 'Number of Incorrect Logins' (like http://www.eml.cc/[^] does)
Also, in the user profile, you can show whether the user is online or offline and his last login timestamp.
|
|
|
|
|
No - this is a breach of privacy.
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Hi Chris,
I first understood it as showing the name of who ever did the most recent login;
but that does not make much sense to me.
Now I think VDK meant information on login attempts with the current user's account;
that's what several services do (such as banks over here), it may help in noticing
break-in attempts.
Luc Pattyn [Forum Guidelines] [My Articles]
this weeks tips:
- make Visual display line numbers: Tools/Options/TextEditor/AllLanguages/General
- show exceptions with ToString() to see all information
- before you ask a question here, search CodeProject, then Google
|
|
|
|