Problem with document.lastModified in Mozilla

B

Bart

Hallo,

I'm have a problem with the following script:


function wr(s)
{
//Just got tired of writing document.write,
//so I created a shorthand version
document.write(s);
}


function lm()
{
//I know this can be shorter but the original script
//tried to format the outputstring, wichh did not
//work in Mozilla, so I commented a lot out
//and this is what's left of it...
var s = document.lastModified;
return s;
}

function lmstring()
{
var s = "This page was last modified at: ";
s += lm();
s += ".";
return s;
}


<html>

...

<script type="text/JavaScript">
<!--
wr(lmstring());
// -->
</script>

....
</html>


This _should_ return the date and time the document was last modified
(saved), however in Mozilla (1.4; Windows ME) it will return the
actual current date and time (as if I'd used "new Date()").
In IE (5.50) it behaves as I expect.

What am I doing wrong (or is it a Mozilla bug ???).

(The script is in an external .js file, if that should make any
difference)
 
D

Dr John Stockton

JRS: In article <[email protected]>, seen in
news:comp.lang.javascript said:
This _should_ return the date and time the document was last modified
(saved), however in Mozilla (1.4; Windows ME) it will return the
actual current date and time (as if I'd used "new Date()").
In IE (5.50) it behaves as I expect.

What am I doing wrong (or is it a Mozilla bug ???).
Click to expand...

For testing, and even more so for getting advice, throw all that code
away. It wastes our time.

document.write('"', document.lastModified, '"')

can be understood immediately, and shows exactly what the string
contains.

By reading the FAQ, you could have found
<URL:http://www.merlyn.demon.co.uk/js-date3.htm#lM>, q.v.


You do not say whether you are accessing the file as a file on your
local disc, as a Web page but from a server on your computer, or as a
Web page from a remote server.



The HTTP header line Last-Modified is well-specified; and I suspect
that, where implemented, it is correctly implemented.

But the javascript document.lastModified string appears to be totally
unspecified, or differently specified in different DOMs. If it is to be
correctly interpreted by all browsers, world-wide, it must be carefully
treated.

The reliable, sensible, economical easy approach is to write the date,
in an international standard form, in HTML on the page, and to update it
when you edit the page significantly. Like I do.


Consider

if (new Date() - (D = new Date(document.lastModified)) > 3e4)
LMdisplay(document.lastModified)

Remember that the string may be GMT.
 
@

@SM

Bart a ecrit :
Hallo,

I'm have a problem with the following script:
Click to expand...

function wr() { document.write(lmstring()); }

function wr(s)
{
//Just got tired of writing document.write,
//so I created a shorthand version
document.write(s);
}

function lm()
{
//I know this can be shorter but the original script
//tried to format the outputstring, wichh did not
//work in Mozilla, so I commented a lot out
//and this is what's left of it...
var s = document.lastModified;
return s;
}

function lmstring()
{
var s = "This page was last modified at: ";
s += lm();
s += ".";
return s;
}

<html>

..

<script type="text/JavaScript">
<!--
wr(lmstring());
// -->
</script>

...
</html>

This _should_ return the date and time the document was last modified
(saved), however in Mozilla (1.4; Windows ME) it will return the
actual current date and time (as if I'd used "new Date()").
In IE (5.50) it behaves as I expect.

What am I doing wrong (or is it a Mozilla bug ???).

(The script is in an external .js file, if that should make any
difference)
Click to expand...

--
******** (enlever/remove [OTER_MOI] du/from reply url) *******
Stéphane MORIAUX : mailto:[email protected]
Aide aux Pages Perso (images & couleurs, formulaire, CHP, JS)
http://perso.wanadoo.fr/stephane.moriaux/internet/
**************************************************************
 
B

Bart

function wr() { document.write(lmstring()); }
Click to expand...
Actually I use the wr() function in all my pages instead of
document.write, so this won't solve anything for me.
I think you misunderstood my question.
As fa as I can trace, document.lastModified does not return the date
the document was last modified, but the actual current date & time...

Bart
 
B

Bart

Op Sun, 01 Feb 2004 18:30:26 +0100 schreef Bart
<[email protected]>:

Just tested this (Mozilla 1.4 on Windows ME)

<script type="text/JavaScript">
<!--
document.write(document.lastModified);
// -->
</script>

This will return the actual time and date of my computer clock, _NOT_,
as one should ecxpect, the date & time the document was last modified.

(saved the HTML opened in browser: gave me "february 02, 18:36";
then changed computer clock to december 2nd 2004, reloaded the page
and it gave me "december 02, 2004 18:36" ....)

Netscape 6.0 behaves correctly on the same script...

Can anyone reproduce this ????

Bart
 
R

Richard Cornford

This will return the actual time and date of my computer
clock, _NOT_, as one should ecxpect, the date & time the
document was last modified.
Click to expand...
Can anyone reproduce this ????
Click to expand...

The browser gets its last modified date from the HTTP headers sent with
the HTML when it is downloaded, and only if the server sends that
information. When testing from the file system there are no HTTP
headers, IE may know how to get the information from the (Microsoft) OS;
Mozilla may not (or care to try).

If you want to see if the problem is reproducible you will have to go
into more details on the context. Perhaps a URL to a test file on a
server somewhere.

Richard.
 
@

@SM

Bart a ecrit :
As fa as I can trace, document.lastModified does not return the date
the document was last modified, but the actual current date & time...
Click to expand...

If document is not on a server your experencies are of no interest.

On My Mac IE and NC doesn't give same date/hour.

Of course, I didn't try playing with the computer clock !
Some secondes or minutes of waiting is enough to see it is realy
a last time modified, even if the time is false ...

--
******** (enlever/remove [OTER_MOI] du/from reply url) *******
Stéphane MORIAUX : mailto:[email protected]
Aide aux Pages Perso (images & couleurs, formulaire, CHP, JS)
http://perso.wanadoo.fr/stephane.moriaux/internet/
**************************************************************
 
B

Bart

Op Mon, 2 Feb 2004 18:37:06 -0000 schreef "Richard Cornford"
The browser gets its last modified date from the HTTP headers sent with
the HTML when it is downloaded, and only if the server sends that
information. When testing from the file system there are no HTTP
headers, IE may know how to get the information from the (Microsoft) OS;
Mozilla may not (or care to try).

If you want to see if the problem is reproducible you will have to go
into more details on the context. Perhaps a URL to a test file on a
server somewhere.

Richard.
Click to expand...

The behaviour is the same on a server (my homepage) or on my local
computer.
I recently tested the same page in Mozilla 1.2.1 and it behaved
normally. (I seem to remeber that Mozilla 1.3 also behaved as I
expected, since I tested all my pages under Mozilla 1.3, before
upgrading to 1.4).

This seems odd...

Thanx anyway !

Bart
 
B

Bart

If document is not on a server your experencies are of no interest.
Click to expand...

Checked it. You're right (that is today, yesterday I checked and it
wasn't. Same file, same Browser). Then again, it's always nice to
blame Bill if my computer behaves badly ;-)


However Mozilla 1.2.1 did this allright.
The site I'm working on will be a gift to some friends, and will be on
a CDROM.
(Anyhow most of them use IE, and they will see the proper date. It's
just that I prefer Mozilla, and it would be nice to show the date of
last modification in Mozilla as well...)

Thanks!

Bart
 
R

rgee_f

Bart said:
Checked it. You're right (that is today, yesterday I checked and it
wasn't. Same file, same Browser). Then again, it's always nice to
blame Bill if my computer behaves badly ;-)


However Mozilla 1.2.1 did this allright.
The site I'm working on will be a gift to some friends, and will be on
a CDROM.
(Anyhow most of them use IE, and they will see the proper date. It's
just that I prefer Mozilla, and it would be nice to show the date of
last modification in Mozilla as well...)

Thanks!

Bart
Click to expand...


Bart:

I have run into this problem where the page is displaying the current
time...this is driving me nuts. I read your thread and it looks like
you solved it but am unclear as to how this was done.

I have an external file that looks like this:

var day = new Array
("Sunday",
"Monday",
"Tuesday",
"Wednesday",
"Thursday",
"Friday",
"Saturday");

var mon = new Array
("January",
"February",
"March",
"April",
"May",
"June",
"July",
"August",
"September",
"October",
"November",
"December");

var now = new Date();
var last_mod = new Date(document.lastModified);
var mod_month = mon[now.getMonth()];
var mod_date = last_mod.getDate();
var mod_hour = last_mod.getHours();
var mod_minute = last_mod.getMinutes();
var mod_second = last_mod.getSeconds();
var mod_year = last_mod.getYear();
if (mod_year < 2000) {
var mod_year = mod_year + 1900 //because netscape does not evaluate
date properly
}

My html page looks like this:
<html>
<head>
<script src="my_lastModDate.js" type="text/javascript">
</script>
<title>
</title>
</head>
<body>
blah, blah, and more blah...
<SCRIPT TYPE="TEXT/JAVASCRIPT">
<!-- Hide

document.write ("Last modified on: <br>");
document.write (mod_month + " " + mod_date + "\, " + mod_year);
document.write(" " + mod_hour + ":" + mod_minute + ":" +
mod_second);

-->
</SCRIPT>

</body>
</html>

TIA,
rgee_f
 
D

Dr John Stockton

JRS: In article <[email protected]>, seen
in news:comp.lang.javascript said:
I have run into this problem where the page is displaying the current
time...this is driving me nuts.
Click to expand...



Showing all that code is largely pointless - the prime question is "what
does the lastModified string itself hold?".

If it holds some version of the current date/time, even in FFF, then one
cannot hope to re-interpret it as anything else.

If it holds a date/time equivalent to 1970-01-01 00:00:00 GMT, even in
FFF, then one cannot hope to re-interpret it as anything else.

If it holds a date/time which is reasonable for a Web page, i.e. after
TB-L's invention but appreciably before current, even in FFF, then one
can hope to re-interpret it as a respectable representation of that.

The re-interpretation code can, and should, be tested independently, on
other date strings of expected formats,



BTW, what happens with pages generated dynamically on the server - do
they have a Last-Modified header representing the instant of generation,
or something else, or not?
 
R

rgee_f

Dr John Stockton said:
JRS: In article <[email protected]>, seen




Showing all that code is largely pointless - the prime question is "what
does the lastModified string itself hold?".

If it holds some version of the current date/time, even in FFF, then one
cannot hope to re-interpret it as anything else.

If it holds a date/time equivalent to 1970-01-01 00:00:00 GMT, even in
FFF, then one cannot hope to re-interpret it as anything else.

If it holds a date/time which is reasonable for a Web page, i.e. after
TB-L's invention but appreciably before current, even in FFF, then one
can hope to re-interpret it as a respectable representation of that.

The re-interpretation code can, and should, be tested independently, on
other date strings of expected formats,



BTW, what happens with pages generated dynamically on the server - do
they have a Last-Modified header representing the instant of generation,
or something else, or not?
Click to expand...


John:

Thanks for the reply. As far as I can tell the lastModified string is
hoding current and time. I am sorry but I really feel stupid because
I really do not quite understand your three above 'if'
statements...When I run my lastModified code and press the reload
button I can see the seconds ticking away. If I leave the page alone
for a day it will display the current date and time. I went to your
page and am still unclear if this is a coding problem or a server
problem, in that the server is not sending HTTP header data back to
the browser...

TIA.
rgee_f
 
D

Dr John Stockton

JRS: In article <[email protected]>, seen
Thanks for the reply. As far as I can tell the lastModified string is
hoding current and time. I am sorry but I really feel stupid because
I really do not quite understand your three above 'if'
statements...When I run my lastModified code and press the reload
button I can see the seconds ticking away. If I leave the page alone
for a day it will display the current date and time. I went to your
page and am still unclear if this is a coding problem or a server
problem, in that the server is not sending HTTP header data back to
the browser...
Click to expand...


A confusing report.

What you need to do is to put the following code on your server at a
known time of day, wait for at least a few times as long as the largest
possible difference between server time and real time, then fetch the
page and show us the exact result, the time of upload, and the time of
fetch.

You need also to say, unless looking at the result tells you all you
need to know, how your own local time differs from GMT; and, in case
your server is wrongly configured, it could be useful to know server's
local time difference from GMT. Always give locality information in any
case where it may matter.

<script>
document.write('Result: "', document.lastModified, '"')
</script>

Using my test page locally, I get Result: "02/09/04 19:53:54" which
is, in American, exactly when the page was changed.

I expect that the string will always represent a GMT time, and that it
may contain the letters GMT.

You may get the time of upload by the server clock, the time of fetch by
your computer's clock, "0", "", or something unpredictable (I don't
expect "0" or "", if I understand you correctly.

There is NO POINT in examining this string with any more code until the
string itself is understood.


You could, of course, also tell us all the URL of the page, so that we
can look at it.


IIRC, you can check the page with http://malch.elsop.com/quick.cgi -
IIRC, the full report includes the headers, and may show a Last
Modified: line.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Similar Threads

Problem with android and scrolling with <input textarea 5
Issue with textbox script? 0
Help with a function 8
Need help with this script 4
Help with code 0
JSON problem on Supervideo.tv remote upload app 4
Collect Excel Data from Website 5
Only one table shows up with the information 2

Members online

No members online now.
Total: 23 (members: 1, guests: 22)
Robots: 120

Forum statistics

Threads
473,776
Messages
2,569,603
Members
45,188
Latest member
Crypto TaxSoftware

Latest Threads

Top