Closed Bug 187930 Opened 22 years ago Closed 14 years ago

Not handing URL with mix of DOS & UNIX slash

Categories

(SeaMonkey :: General, defect)

x86
Windows NT
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED DUPLICATE of bug 64488

People

(Reporter: pmarvin, Assigned: asa)

References

Details

User-Agent: Mozilla/5.0 (Windows; U; WinNT4.0; en-US; rv:1.3b) Gecko/20030102 Build Identifier: Mozilla/5.0 (Windows; U; WinNT4.0; en-US; rv:1.3b) Gecko/20030102 I have tested this with both Windows NT 4.0 SP6 and Red Hat Linux 7.3 installs of mozilla. The Red Hat version nightly build 2003010605. I am drilling through web links fine until I get to an application built page. The application built page as a graphic defined as <img src="images%5Cdiagram0.jpeg". This is producing a URL similar to http://root/level1/level2/images\diagram0.jpeg. I do not have control over this generated code but when I changed "%5C" to a unix slash (/) to test that this is the problem, everything worked fine. If I start with the HTML code that calls this graphic then without edits everything works fine. When I replaced the %5C with a physical DOS slash (\) it stilled failed. Reproducible: Always Steps to Reproduce: 1. 2. 3. Actual Results: The graphic would fail to load display. Expected Results: converted the dos slash (\) to a unix slash (/) before getting the graphic.
"\" is not illigeal for URLs. Mozilla accepts this only for users who are typing URLs with "\" in the URL Bar but Mozilla doesn't support broken Links with "\". -> invalid
Status: UNCONFIRMED → RESOLVED
Closed: 22 years ago
Resolution: --- → INVALID
verified. '\' is a valid character in a filename and is _not_ a path separator in urls.
Status: RESOLVED → VERIFIED
OK. After reading what I submitted I understand why I got the response I did. I have a third party application that generates HTML code. It puts a reference to a critical image file into the code as: <img src="images%5Cdiagram0.jpeg". When I access this page the image file fails to load because it cannot find the file. Examinining the properties on the broken image icon shows a URL similar to http://root/level1/level2/images\diagram0.jpeg. Unless I manually edit every generated page and change images%5Cdiagram0 to images/diagram0, I cannot use Mozilla. IE works without any edits.
See http://www.ietf.org/rfc/rfc2396.txt for the standard that defines what a URL is, and how to specify it (paragraph 1.6 and 3.3). It's IE that is broken by accepting '\' instead of a '/'. But that doesn't help on webpages where a real '\' is part of the segment (normally a directory-name), and IE is not allowed to replace them. If the application generated a proper '/', it would work in all browsers. That's where the bug is, not in Mozilla.
Paul, I understand your situation, and it's not a happy one... but what you're asking us to do is to put in a workaround for the bug in the app you're using; a workaround that will break some perfectly correctly authored pages (as IE does)....
*** Bug 190592 has been marked as a duplicate of this bug. ***
*** Bug 190785 has been marked as a duplicate of this bug. ***
*** Bug 199442 has been marked as a duplicate of this bug. ***
*** Bug 238199 has been marked as a duplicate of this bug. ***
*** Bug 262339 has been marked as a duplicate of this bug. ***
*** Bug 262339 has been marked as a duplicate of this bug. ***
Product: Browser → Seamonkey
*** Bug 359355 has been marked as a duplicate of this bug. ***
Status: VERIFIED → RESOLVED
Closed: 22 years ago14 years ago
Resolution: INVALID → DUPLICATE
You need to log in before you can comment on or make changes to this bug.