gmane.network.wwwoffle.user

How to view a page with HTTPS images offline inchromium

2012-01-31T17:44:07

How to view a page with HTTPS images offline in chromium.
$ chromium http://code.google.com/apis/chart/image/docs/gallery/line_charts.html
Left click first broken image icon and view in new tab.
Click "Proceed anyway".
Now refresh line_charts.html.
This must be done each time you want to view the page.

wwwoffle to be removed from Debian unless a newmaintainer steps forth

2011-11-07T12:20:34

Gentlemen, wwwoffle is in danger of being removed from Debian
http://news.gmane.org/group/gmane.linux.debian.devel.general/thread=166494/
unless a new maintainer steps forth.
(Me? I'm too old. http://www.youtube.com/watch?v=Kq5WPkLCAiY&list=PL6E40919035151385 )
It would perhaps be quite simple to maintain if one just used pretty
much the official upstream package without any local add ons.

Re: WWWOFFLE HTTPS now unusable

Andrew M. Bishop — 2011-10-04T09:07:28


I believe that these problems are all fixed in the new version of
WWWOFFLE (2.9h) which I released yesterday.

The problems were caused by the combination of an incompatible change
in gnutls (now fixed, but this new version of WWWOFFLE has a
work-around) and an apparent change in gnutls behaviour which exposed
a pre-existing bug in WWWOFFLE (now fixed).

Re: website broken with wwwoffle

Matus UHLAR - fantomas — 2011-05-10T14:26:57

On 07.05.11 17:04, Andrew M. Bishop wrote:

funny, Now I'm getting the same result. I turned proxy of and on repeatedly
to be sure it happend with wwwoffle.


yes, I see it now. 


no difference now. I'm going to watch this more precisely.

Re: website broken with wwwoffle

Andrew M. Bishop — 2011-05-07T16:04:05


When I request this page through WWWOFFLE I get the same result as
when I request it with wget.

There is something unusual about the data though, it is using 16-bits
per character and the first two bytes have the values hex values FF
and FE.

When displayed in less it looks like this:

<FF><FE><^< at >!^< at >D^< at >O^< at >C^< at >T^< at >Y^< at >P^< at >E^< at > ^< at >H^< at >T^< at >M^< at >L^< at > ^< at >P^< at >U^< at >B^< at >L^< at >I^< at >C^< at > ^< at >"^< at >-^< at >/^< at >/^< at >W^< at >3^< at >C^< at >/^< at >/^< at >D^< at >T^< at >D^< at > ^< at >H^< at >T^< at >M^< at >L^< at > ^< at >4^< at >.^< at >0^< at >1^< at > ^< at >T^< at >r^< at >a^< at >n^< at >s^< at >i

You can see that if you remove the zero bytes (shown as '^< at >') then you
get normal HTML.


Do you get a normal web page when you are not using WWWOFFLE and only
this strange result when you use WWWOFFLE?  What do you see when you
use wget:

wget -Y off http://www.hokej-megacamp.sk/hokej-sen/hokej.html

website broken with wwwoffle

Matus UHLAR - fantomas — 2011-04-28T17:14:17

Hello,

when I'm trying to fetch 
http://www.hokej-megacamp.sk/hokej-sen/hokej.html
using wwwoffle, I get broken content:

GET http://www.hokej-megacamp.sk/hokej-sen/hokej.html HTTP/1.0

HTTP/1.0 200 OK
Date: Thu, 28 Apr 2011 17:10:23 GMT
Server: Apache
Last-Modified: Mon, 27 Dec 2010 14:32:01 GMT
ETag: "2cd148-2a44-4986532f6da40"
Accept-Ranges: bytes
Content-Type: text/html
Connection: close
Proxy-Connection: close

!<h<h<m<t</<m<m<m<m<m<l</<b<t<t<t<i</<t<t<t<t<a<i</</</<t<t</</<t<t<t<t<t</</<t<t</<t<a</<a</<a</<a</<a</<a</</<t</</</</</</</<t<t<t<t<i</</</<t<t<t</<t<h</<h</<b<t<t<t<p<b</</</<t<p<b</</</</<t<t<p<b</</</<t<p<b<b</</</</<t<t<p<b</</</<t<p<b</<b</</</</<t<t<p<b</</</<t<p<b</</</</<t<t<p<b</</</<t<p<b</</</</<t<t<p<b</</</<t<p<b</</</</<t<t<p<b</</</<t<p<b</</</</<t<t<p<b</</</<t<p<b</</</</<t<t<p<b</</</<t<p<b</</</</</</<t</</</</</</<d<s</<s<a</<a</<a</<a</<a</<a</</</</</Connection closed by foreign host.


I use debian-patched 2.9g (unpacked debian-patched 2.9f and applied diff
manually)

Re: Request Already Recorded message doesn'thammer the point in enough

Andrew M. Bishop — 2011-01-24T17:01:29


The reason that I use the word "Already" rather than "Previously" is
because "Already" tells you that the request is currently recorded
while "Previously" only tells you that it was once recorded (which
includes the case that "Already" implies).

I don't think that there is any point in telling the user that they
are wasting their time.  The chances are that they clicked the link
without knowing (or being sure) if they have already requested it.  I
wouldn't want to alienate the users by telling them not to do it
again (even if in a humorous way).


The two mesages are identical apart from the word "already" which
appears twice in the second page and not at all in the first.

Request Already Recorded message doesn't hammerthe point in enough

2011-01-21T22:37:50

Actually this isn't clear enough for my English:

  WWWOFFLE Request Already Recorded
  Your request for URL
  http://www...
  has already been recorded for download.

I would say:

  WWWOFFLE Request Previously Recorded
  Your request for URL
  http://www...
  has already been recorded for download.
  So there is no point in you asking this additional time, Holmes :-)

Or something.

Hmmm, 'previously' might not imply still on the queue though.
Anyway, 'already' for me is the common word companies use to say 'really
in the queue'.

Yes, if one compares the
WWWOFFLE Request Recorded and
WWWOFFLE Request Already Recorded
messages, one will know the difference.
However the latter means the former if one has never encountered the former.

Re: Warning: Failed to get name/IP address forhost ... [Name or service not known].

Andrew M. Bishop — 2010-07-12T17:42:32


The message "Warning: Failed to get name/IP address for host" is only
a warning so this isn't the problem and can be ignored (WWWOFFLE is
trying to get a fully qualified domain name and jidanni4 isn't one).
The WWWOFFLE server seems to have started OK - there are no errors.

The problem is with the wwwoffle program trying to connect to hostname
0.0.0.0 but this value should only appear in the configuration file as
the 'bind-ipv4' option and the WWWOFFLE program won't use this value.

What value do you have as the first entry in the LocalHost section of
the configuration file?  This is the hostname that the wwwoffle
program will use and it should be set to jidanni4.

Warning: Failed to get name/IP address for host... [Name or service not known].

2010-07-11T00:06:59

Help. On one of my machines, if one tries to do ANYTHING with wwwoffle,
from the time of boot until first connecting to the Internet, one gets

# wwwoffled -c /etc/wwwoffle/wwwoffle.conf
wwwoffled[3769] Warning: Failed to get name/IP address for host 'jidanni4' [Name or service not known].
wwwoffled[3769] Important: WWWOFFLE Demon Version 2.9f (with ipv6,with zlib,with gnutls) started.
wwwoffled[3769] Information: WWWOFFLE Read Configuration File '/etc/wwwoffle/wwwoffle.conf'.
wwwoffled[3769] Information: Running with uid=13, gid=13.
wwwoffled[3769] Information: Read in 0 trusted certificates.
wwwoffled[3769] Important: Detached from terminal and changed pid to 3772.
# wwwoffle -online -c /etc/wwwoffle/wwwoffle.conf
wwwoffle[3775] Warning: Failed to get name/IP address for host 'jidanni4' [Name or service not known].
wwwoffle[3775] Warning: Unknown host '0.0.0.0' for server [Address family for hostname not supported].
wwwoffle[3775] Fatal: Cannot open connection to wwwoffle server 0.0.0.0 port 8081.
# hostname
jidanni4
# host `hostname`
jidanni4 has address 127.0.0.1

After connecting to the Internet all works normally.

The problem occurs no matter what one puts in wwwoffle.conf, no matter
what one does with /etc/hosts, no matter what one does with pdnsd. Help.

Re: Stopping an ongoing "wwwoffle -fetch" action?

2010-05-30T02:02:49

Perhaps also see http://jidanni.org/comp/wwwoffle/wwwoffle-swat .

Re: Stopping an ongoing "wwwoffle -fetch" action?

Andrew M. Bishop — 2010-05-19T17:34:12


If you set WWWOFFLE offline then it will stop fetching (when the
current requests have finished).


You can use the wwwoffle-ls program to get a list of the cache
contents and current requests.  It is very easy to use

wwwoffle-ls outgoing

You might need to use the '-c' option to specify the configuration
filename so that the program can find the files.


The easiest way to manipulate the list of requests is using the web
page provided:

http://localhost:8080/index/outgoing?sort=alpha

Stopping an ongoing "wwwoffle -fetch" action?

Zrajm C Akfohg — 2010-05-16T20:41:56

I have involontarily started a "wwwoffle -fetch" at several times -- are
there any way of interrupting that? When I simply wish to stop downloading
pages, but not to flush any download queue, or anything of the sort.

(This mostly happens when I'm on my laptop, running on battery power.)

BTW is there some way of listing and manipulating the current download
queue?

/zrajm

Re: Debian Bug#564475: last NMU forgot to compilein libgnutls, HTTPS now broken

2010-01-22T20:41:56

A> I see no reason that anything is *unsafe* in downgrading.
OK, I'll use the old old version until someone fixes the old version.
# dpkg -i /var/cache/apt/archives/wwwoffle_2.9d-3_i386.deb
dpkg: warning: downgrading wwwoffle from 2.9d-3.1 to 2.9d-3.
Checking for htdig package...
Fixing ownership of spooldirs in the background.
# killall chown
# aptitude forbid-version wwwoffle

Re: request URL, but not if already in cache

Albert Reiner — 2009-11-18T11:07:15

Hi Andrew,

thanks for your response:

[amb-Uxr6IM1mbv2TY6FTCsQk+9Bc4/FLrbF6< at >public.gmane.org (Andrew M. Bishop), Mon, 16 Nov 2009 17:46:44 +0000]:

Yes, I knew it was an old version (and three years is not that much
for mature software, is it?).  As it turns out, I always followed the
announcements, but so far WWWOFFLE works absolutely wonderfully for
what I need.  The improvements simply never mattered enough for me to
make me want to re-install WWWOFFLE on two different computers,
especially as these installations need to be match (as I am browsing
on a computer without any internet connection, carrying files back and
forth via memory sticks or, initially, even 3.5" floppies).

...

which is essentially the same as what I mentioned in my e-mail, except
that `xargs -i` is replaced by `wwwoffle -n 1`.

Thanks again for your work, as well as for your reply -

Albert.

Re: request URL, but not if already in cache

Andrew M. Bishop — 2009-11-16T17:46:44


Did you realise that this version of WWWOFFLE that you are using is 3
years old?  The current version is 2.9e and that is nearly a year old.



I think that -<new-option> actually exists and is -o.

The following should do what you want (you need to throw away the
output of the "wwwoffle -o" command and only pass in one URL at a
time).

cat <list-of-URLs> | xargs wwwoffle -n 1 -o > /dev/null

request URL, but not if already in cache

Albert Reiner — 2009-11-13T09:56:05

Consider the following situation (running WWWOFFLE 2.9a under Linux):

Suppose I have a list of URLs, say, URL1 URL2 URL3..., some of which
are already present in the cache.  Now I want to make sure that, once
I am online again, those and only those URLi will be fetched that are
not already in the cache.

- One way to do this would be to start the browser and request each
  one of the URLi, which would do the right thing (show those pages
  that are already in the cache, mark for fetching those that are
  missing) but is extremely tedious if there are many URLs.

- The simplest I have come up with is

    cat <list-of-URLs> | xargs -i wwwoffle -o '{}' > /dev/null

  , which seems to work.

But what I would really like is some additional option such that

    cat <list-of-URLs> | xargs wwwoffle -<new-option>

will do the right thing, i.e., `wwwoffle -<new-option> URL1 URL2...`
would behave like `wwwoffle URL1 URL2...` for URLi not already cached
but skips those that are not present (and outputs some message like
'URLi is already in the cache, will not be fetched again' on stderr).

Would this be a valuable feature to add?  And is there any chance this
will actually be added in the future?  Comments?

Best regards, and many thanks to Andrew for WWWOFFLE which has been in
constant use here for many years -

Albert.