274741 - NS_ERROR_FILE_UNRECOGNIZED_PATH exception

Status: RESOLVED INVALID

(Other Applications Graveyard :: Venkman JS Debugger, defect)

Description

[Robert Brown III]

1. I have Version 0.9.84 of Venkman
2. I have an HTML file, which in turn has a <script> tag that refers to a
Javascript file. The javascript file resides in a subdirectory called "jscript".

3. I open the HTML file using Firefox.

4. I start Venkman

5. In the "Open Windows" view, I select my HTML file ->Files->my Javascript file
(in this case, called calendarcode.js).

6. I get the following error message:

Error loading URL <file:///Jscripts/calendarcode.js>: [Exception... "Component
returned failure code: 0x80520001 (NS_ERROR_FILE_UNRECOGNIZED_PATH)
[nsIChannel.asyncOpen]" nsresult: "0x80520001 (NS_ERROR_FILE_UNRECOGNIZED_PATH)"
location: "JS frame :: chrome://venkman/content/venkman-url-loader.js ::
loadURLAsync :: line 75" data: no].

Note the following:

a. The javascript loads and operates without problems in Firefox; in other
words, the browser has no problem finding the Javascript file. This means that
the <script> tag in my HTML is correct.

b. The path which Venkman seems unable to recognize was apparently created by
Venkman. This suggests that (despite [e-mail]'s comments on
https://bugzilla.mozilla.org/show_bug.cgi?id=274623) that it isn't me who is
creating an erroneous path to the JS file, but it is Venkman.

c. This behavior is the same on Versions .8, .9, and 1.0 of Firefox, on Windows
2000, Windows XP, Red Hat Linux 9, and SuSe Linux. This suggests that Venkman
has a bug forming paths to referenced JS files which has nothing to do with the
operating system it is residing on.

   One other problem: this particular error may "not have been seen before" but
it clearly exists and you are seeing it now. It is, ultimately, a cryptic error
which should be trapped, processed, and displayed to the user in English (or
whatever localized language is appropriate). In its current form, it tells the
user a lot of nothing. It is one of many problems with Venkman that make it an
unusable, low- quality application.

Comment 1

[James Ross]

Thanks for missing out the CRUTIAL information you smart-arse. We would have to
see the page in question, or at least the <script> tag, to be able to do
anything here.

So then, please fix our 'low-quality' application.

Also, FYI, inernal errors are just that. INTERNAL. Yes, it means something broke
somewhere, but no they do not all need to be painstakingly turned into some
'pretty' message that ultimatly would tell the developer NOTHING.

Comment 2

[Robert Brown III]

Um... Mr. Ross:

First: the word is spelled "Crucial" not Crutial. If you want to make pathetic 
attempts at insults, do spell your words correctly.

Second: if you want to see the page, or the <script> tag, all you have to do 
is ask. I'll gladly include it if you ask nicely.

Third: In none of my bug reports have I suggested making a "pretty" error 
message. I am suggesting making a "proper" one. Proper error handling involves 
creating an error message that is informative and tells the user exactly what 
the problem is. The current error message documented here does not do so.

Of course, I was assuming that you are a competent engineer who understands 
error handling 101. I apologize if I was wrong; after all, I wouldn't want to 
ask something that is beyond your capabilities.

Comment 3

[James Ross]

It wasn't an insult, mearly pointing out how silly it was you didn't include the
one bit of information that could help when you seemed to include a lot of
information that didn't help.

"Third: In none of my bug reports have I suggested making a "pretty" error
message. I am suggesting making a "proper" one."

I used quotes around pretty for a reason.

"Proper error handling involves creating an error message that is informative
and tells the user exactly what the problem is. The current error message
documented here does not do so."

Of course it doesn't! If it explain what was wrong, we'd have worked out what
the original problem was and fixed it, wouldn't we? (it's rhetorical, so don't
bother answering)

Comment 4

[Robert Brown III]

You still don't get it, do you?

You clearly have no comprehension of what "error handling" is about.

Forget it. By now you should have seen my last bug entry. And I do apologize; 
I did indeed ask too much from you. If you are so ignorant that you cannot 
comprehend what a "proper" error/failure message is, then you -- and this 
pathetic debugger of yours -- are without hope.

Goodbye.

Comment 5

[James Ross]

Robert Brown III, we would love to fix the bug, though I'm not sure you would
care any more, but you spent far too much time accusing us of knowing nothing or
being unable to fix it without actually giving us any opportunity (or nessessary
information) to do so.

Anyway, since the reported has apparently left forever without giving us the
<script> in question, I don't see how we can fix this. CANTFIX would be better,
but INVALID is all we've got. :(

If anyone knows what the <script> is that causes this error, please re-open it
giving us an example so we can investigate and fix whatever is broken.