There was much discussion about a year ago about the cache performance on
RISC OS, and there were some code changes, but I would like to add the
results of some investigations of the Netsurf v3.4 cache on my Iyonix,
running RISC OS 5.23 (11 Oct 2015).
In the past I had problems with the cache taking large amounts of disc
space, and the resulting long backup times for !Boot, so my current
settings are 10MB space, expiring after 2 days.
However, the actual space usage was 45MB (as measured by Filer Count),
and it contained 210 files. What was more difficult to find was that
there were 7,298 directories with 8 levels, which occupied another 14MB,
of which 6,412 contained no files at any lower level. So only 886
directories actually contained the 210 files of cached data. Enumeration
of the cache took about 2 minutes.
I decided to delete all 6,412 directories that contained no data, saving
about 12MB of disc space. More importantly, counting or enumerating all
the cache now takes about 7 seconds. There are still the same number of
files and cached bytes.
Netsurf itself still seems to work, but I have not noticed any change in
So, some questions:
- When are cached files deleted to meet the configured size & expiry?
- Are directories included in the space used?
- Are directories ever deleted? If so, when?
- Will deletion of empty directories cause any problems for Netsurf?
I have looked at the help ... but that says that files are not deleted by
Netsurf, and makes no mention of directories. It also refers to a
'Perform maintenance' button which can be used to delete redundant files
... but this is nowhere to be seen!
I'm tinkering with something on my own website, an area that requires
the viewer to provide a username and password.
I successfully get in with my own username and password. Now I want
to test those of another user.
So I click the link that takes me to the subsite -- and I'm straight
in; I don't see any box asking for username and password.
Tried Ctrl-F5, menu Reload, tried deleting from the History file,
tried quitting and relaunching Netsurf. Still never see the login box
How do I "log out"? (If that's the correct term in this situation.
Maybe it's more like changing the "state".)
Jim Nagel www.archivemag.co.uk
I've just discovered that I can get Facebook up in NetSurf if I substitute
"m" for "www" in the URL.
That'll save a lot of wear and tear on the memory stick to get photos of
the new granddaughter into my RISC OS e-mails for the elderly relative (90)
that sends cheques, not to mention my time!
Forgive me if you already knew ...
| John Williams
Names for Soul Band:- Soul Suppliers *
> Date: Sat, 14 May 2016 14:24:46 +0100
> From: Jim Nagel <netsurf(a)abbeypress.co.uk>
> Subject: Re: how to make Netsurf forget its logged-in state
> Frank de Bruijn wrote on 14 May:
> > If I quit NetSurf it *does* forget the login details, as I would have
> > expected. The authentication window pops up again after a restart.
> > That's with NetSurf 3.6 (Dev CI #3538). Which version are you using?
> moment, if that's relevant). And I'll update today, I promise. I
> usually update much more frequently; the past month or two are
> You're right, though: quitting Netsurf does forget the login details.
> So that's obviously the simple solution to my original problem.
> Still would like to know where Netsurf stores this info -- cache? --
> and if there'd be any way to erase it (i.e., to "log out") without
> quitting everything else that might be open.
This sounds like the sort of session behaviour that (Oracle) single sign-on
provides. Once you sign-in the browser remembers until you close the whole
browser, this is how IE, Firefox behave with Oracle sso.
Thanks for explaining the login process. It looks like you make use of
Basic Auth for the login. I think you should be able to destroy the saved
state if you login using another username and password.
Could you try doing this ?
Open this URL, http://username:firstname.lastname@example.org in netsurf.
Could you try a couple of cases ?
One, a username password which is valid but different from the saved one.
A different (invalid) username password combo.
I am hoping the invalid combo forces the login window next time (or locks
On May 12, 2016 10:31 PM, "Jim Nagel" <netsurf(a)abbeypress.co.uk> wrote:
> Ashish Gupta wrote on 12 May:
> > It would help if you could elaborate on this login process that your
> > site relies on.
> It's a pretty simple PHP script that asks for a username and password.
> The user submits* these, the PHP checks against its list of authorized
> users, if there's a match you're in.
> The username and password are input via a standard dialogue box, which
> I guess is provided by the OS or by the browser -- the RiscOS box is
> most familiar to me.
> I also tested with Firefox on Windows and saw an equivalent box
> there, followed by a popup box that asked if I want Firefox to
> remember the details for next time; I said no. RiscOS had no such
> Also tested with the default browser on my Android phone: again an
> equivalent login box. Tried that one again just now (two or three
> hours after first access): I'm straight in without being asked for
> username and password.
> > I am also curious to know how a session is maintained as you mentioned
> > that cookies are not used.
> Exactly what I too am curious about.
> When I asked my PHP guru about this, he said HTML is "stateless" --
> meaning, I think, that HTML itself does not register the state of
> Jim Nagel www.archivemag.co.uk
>Date: Thu, 12 May 2016 21:33:22 +0100
>From: Michael Drake <tlsa(a)netsurf-browser.org>
>Michael Drake http://www.netsurf-browser.org/
This may be more complex.
While there is a small slow down across versions:
I was testing by opening the page and then timing how long it takes to
When I got to 3537 I timed the initial load and then the reload.
Initial load: 4:10
Reload : 9:29
The small change from the previous test of 3537 is probably due to page
I will do some more tests with other pages to see if the reload is
I've got two newsletters here. One from PermaJet and another from Wisdom
Books, both through Pluto.
The PermaJet newsletter loads into NetSurf OK via the HTML file attachment.
But, surprise, surprise, the HTML file attachment in the Wisdom Books
newsletter belongs to the PermaJet newsletter and loads the PermaJet
newsletter into NetSurf.???
Never ever seen this one before. My feeling is that the fault must lie
with Pluto but could the fault lie elsewhere, perhaps? How could this
Any explanations to this one, please.
Jeremy Nicoll - ml netsurf wrote on May 13:
> On 2016-05-12 12:27, Jim Nagel wrote:
>> Cookies: none involved -- it's my own site I'm talking about, and it
>> doesn't make any cookies.
> As you replied below (to an email that doesn't seem to have appeared
> here) ...
People have been replying to me privately -- I wish they would reply
to the list. Probably that's the cause of the confusion.
> I would think that the code you're using to do login sets the cookie
> even if you don't know about it; it's hard to see how else this could
Netsurf doesn't seem to know about it either. From iconbar menu,
"Open > Show cookies" shows nothing from "archivemag.co.uk".
I'll ask my PHP guru about it. To my eyes, nothing in the bit of PHP
he wrote for me looks like a cookie-setter.
Jim Nagel www.archivemag.co.uk
> Perhaps you can provide the CI build numbers of "working" and "slow"
> versions? the closer these values are together the easier it will be
> for a developer to attempt to diagnose the issue.
Testing on a complex page www.express.co.uk
Build 3242 required 5m 15s
Build 3537 required 8m 28s
At least it looks like it's not imaginary so I will try to narrow
down the version where the change occurs.