Closed
Bug 271652
Opened 20 years ago
Closed 17 years ago
History and cache does not refresh cascading style sheet CSS after relaunching browser.
Categories
(Firefox :: Bookmarks & History, defect)
Tracking
()
RESOLVED
INCOMPLETE
People
(Reporter: mmarcus, Unassigned)
Details
(Whiteboard: CLOSEME 07/14)
User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.7.5) Gecko/20041107 Firefox/1.0
Build Identifier: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.7.5) Gecko/20041107 Firefox/1.0
I updated the CSS for one of my web sites. After closing and then relaunching
Firefox, the browser does not retrieve the latest external CSS which I had
uploaded to the web site earlier. Instead, Firefox uses the cached CSS (out of
date) when displaying the site. I had to right-click This Frame > Reload Frame
in order to get the frame to render using the latest uploaded external style
sheet. This seems to be a bug related to frames in which the frames all use the
same external CSS.
Reproducible: Always
Steps to Reproduce:
1.
2.
3.
Expected Results:
Firefox should have ignored the cached copy of the site's external CSS since the
cached copy of the CSS was an earlier date than the recently uploaded (newer) CSS.
Comment 1•20 years ago
|
||
I can confirm that I'm seeing this happen on my site as well.
We did took a more css-friendly redesign public last week and some (firefox)
users complained. Asking them to shift-reload solved the problem.
Reporter | ||
Updated•20 years ago
|
Version: unspecified → 1.0 Branch
Reporter | ||
Comment 2•19 years ago
|
||
Shift-reload is a poor solution for this bug. Firefox, while rendering a page,
should check the file date for the css. This should be a transparent background
check since web sites usually are updated fairly infrequently and we don't want
to slow the rendering speed of Firefox. Thus, on the occasional instance were
Firefox detects that the css has been updated, the page should be automatically
refreshed using the updated css. Note that this check must also consider if the
css is used for a frames site which uses the same css for all frames.
Accordingly, each frame must be re-rendered using the updated css if Firefox
(while performing the background check of the date/time of the css) detects that
the css has been updated.
Comment 3•19 years ago
|
||
Are all next conditions met?.
(1) Server doesn't return Cache-Control:max-age header nor Expires: header.
(2) Cache option is one of next.
"When the page is out of date"(browser.cache.check_doc_frequency=3,default)
"Once per session" (browser.cache.check_doc_frequency=0)
If yes, and if NOT "issue only when CSS of a HTML in IFRAMEs", you are possibly seeing Bug 328605 (Case-2 in Bug 221036 Comment #13) or seeing Bug 221036.
Check HTTP header, your cache option settings and about:cache.
Get HTTP log and attach log file to bug(Mime-Type=text/plain), if HTTP and cache related issue. (Shift+Reload, Reload, Normal load by link click or bookmark, and "Reload Frame" cases)
( See http://www.mozilla.org/projects/netlib/http/http-debugging.html )
Reporter | ||
Comment 4•19 years ago
|
||
(In reply to comment #3)
> Are all next conditions met?.
> (1) Server doesn't return Cache-Control:max-age header nor Expires: header.
> (2) Cache option is one of next.
> "When the page is out of
> date"(browser.cache.check_doc_frequency=3,default)
> "Once per session" (browser.cache.check_doc_frequency=0)
> If yes, and if NOT "issue only when CSS of a HTML in IFRAMEs", you are possibly
> seeing Bug 328605 (Case-2 in Bug 221036 Comment #13) or seeing Bug 221036.
> Check HTTP header, your cache option settings and about:cache.
>
> Get HTTP log and attach log file to bug(Mime-Type=text/plain), if HTTP and
> cache related issue. (Shift+Reload, Reload, Normal load by link click or
> bookmark, and "Reload Frame" cases)
> ( See http://www.mozilla.org/projects/netlib/http/http-debugging.html )
>
Yep, I checked my config and browser.cache.check_doc_frequency is set to 3. That is obviously the problem. A friend confirms that IE behaves somewhat similarly; he has to blow out IE's cache to be sure he is viewing the latest animation on a weather page. For the average computer user who is not very savy using a computer or the Internet, I would suggest the following tweaks to Firefox regarding this "bug":
1. Netscape used to have a menu for selecting how often to check for updates to web pages. It would be nice to have this feature back under Options with the option to check for updates every time, but the following items are features which I consider to be "mandatory".
2. Lets assume that the average non-savy computer user realizes that the web site he or she is viewing isn't up to date. Firefox should ALWAYS CHECK for newer versions of any web site if Firefox has been restarted -- period! After all, a user might not launch Firefox and browse the Internet after several days since the user last closed Firefox. Thus Firefox should make sure that all info presented on any web site is up to date after Firefox has been restarted. That seems logical to me.
3. Web developers love to check the appearance of their web site using the local copy of their web site. Firefox SHOULD ALWAYS check for more recent versions of files if the URLs are local (within the local network). This feature would also be very beneficial for corporate employees who check their company's local intranet web sites for the latest info. This would be a "one up" with regards to IE and could become a selling point for Firefox to corporations. Corporate management must "know" that Firefox will ALWAYS check for and display the latest revisions of their internal web sites if their employees check their internal web sites for the latest information, regardless of whether or not Firefox has been restarted or the page has been reloaded.
Finally, I have a few non-related suggestions regarding Firefox and JAVA which I consider to be VERY IMPORTANT!!! Thus...
1. If Firefox crashes for any reason, Firefox should automatically KILL JAVA and ALL running JAVA apps no matter what! I hate to have to suggest this, but Firefox has to be retroactive regarding security issues within Sun's JAVA interpreter.
2. Firefox should automatically detect JAVA code which causes the CPU to run continuously or nearly continuously at 100% and which corrupts the OS timekeeping. DON'T LOAD THE FOLLOWING LINK UNLESS YOU REBOOT YOUR COMPTUER AFTER VIEWING THIS LINK! I came across a web page (at a college http://physics.gmu.edu/~hgeller/observatory.html) which uses JAVA within the page for their scrolling messages. Even after I closed Firefox, the JAVA applet somehow remained running in the background and my computer's timekeeping slowed down by two hours before I caught this problem about four hours later! There is some sort of JAVA security at issue here. You would think that once I closed Firefox (and JAVA appeared to close as well) that my computer's timekeeping would have returned to normal. It didn't until I rebooted.
-- Michael
Comment 5•19 years ago
|
||
(In reply to comment #4)
> 1. Netscape used to have a menu for selecting how often to check for updates to
> web pages. It would be nice to have this feature back under Options with the
> option to check for updates every time
Mozilla Suite and Seamonkey also have UI in Preference(and fortunately I use Seamonkey.)
UI request for Firefox was already decided to WONTFIX several times.
See Bug 304953(closed as DUP of Bug 188720 on 2005/8/17) and Bug 188720(closed as WONTFIX on 2003/02/01).
So set browser.cache.check_doc_frequency=1 (Every time I view the page) thru about:config or user.js or prefs.js.
Updated•18 years ago
|
Assignee: bross2 → nobody
QA Contact: mozilla → history
Comment 6•17 years ago
|
||
Reporter, do you still see this problem with the latest Firefox 2? If not, can you please close this bug as WORKSFORME. Thanks!
Whiteboard: CLOSEME 07/14
Comment 7•17 years ago
|
||
-> incomplete, feel free to reopen if you still ee this problem with the current version
Status: UNCONFIRMED → RESOLVED
Closed: 17 years ago
Resolution: --- → INCOMPLETE
Component: History → Bookmarks & History
QA Contact: history → bookmarks
You need to log in
before you can comment on or make changes to this bug.
Description
•