The two Servers

The two Servers

[Frank Schmitt]

Hello

I'm at the moment in the bad situation of reading news at university
from their nntp server and at home from my local newsserver. This is a
pain because the articles I read at university are unread at home and
vice versa.

Therefor I want a solution which solves this. That means articles
numbers are evil, Message-IDs are good. I thought of two solutions:

Either create a second .newsrc file (for example ~/newsrc.id) and when
an entry in ~/.newsrc is made an entry in ~/.newsrc.id is created which
says which Message-ID belongs to this article number.
When getting the overview one had to parse it for already known
Message-IDs.
This solution is probably much and hard work.

Second idea I had is this one: define something like gnus-get-ids which
gets the Message-IDs for all read articles and gnus-read-ids which gets
a list of Message-IDs, retrieves the article numbers of the corresponding
articles and writes those in ~/.newsrc.
This seems to be not to hard.

I searched the documentation and found that gnus-move.el does something
very similar but there is one thing I really don't understand: Why the
hell does it retrieve the whole head of the article? Wouldn't it be *much*
faster to use the stat command?

Anybody out there who has done something similar before or has an other
idea how to do it?

-- 
One Ring to rule them all, One Ring to find them,
One Ring to bring them all and in the darkness bind them
In the Land of Mordor where the Shadows lie.
19. Dezember 2001

Re: The two Servers

[Frank Schmitt]

Andre Canis <Andre.Canis@web.de> writes:

>>I wished it was this easy. Problem number one: At home I've got a dialup
>>connection and I don't like Gnus agent, so I've got to use a local nntp
>>server (That's hamster in my case).
>
>Does your Hamster fetch its news from the same server you use at
>university? Maybe it's possible to tell it to store its articles using
>the same numbers as the upstream server; then you could easily sync
>your .newsrc files.

I'm just writing an application to solve my problem, here's what it
does:

It gets as input the name of a server and the information if it should be
slave or master. 

If it's master then it opens a file consisting of pairs from
article-number and corresponding message-ids (for this server), which
was created by previous runs of the program.

Then it opens ~/.newsrc, looks up for which articles the message-id is
still missing, and gets those message-ids from the newsserver. After
that it creates a file including the message-ids from all articles
listed in ~/.newsrc.

If it's started as slave, it reads the file the master created (the file
with the message-id's of read articles), gets the corresponding article
numbers (like above, already known ones from a file, new ones from the
server) and creates a new ~/.newsrc.

-- 
One Ring to rule them all, One Ring to find them,
One Ring to bring them all and in the darkness bind them
In the Land of Mordor where the Shadows lie.
19. Dezember 2001

Re: The two Servers

[Andre Canis]

Frank Schmitt <usereplyto@Frank-Schmitt.net> writes:

> I wished it was this easy. Problem number one: At home I've got a dialup
> connection and I don't like Gnus agent, so I've got to use a local nntp
> server (That's hamster in my case).

Does your Hamster fetch its news from the same server you use at
university? Maybe it's possible to tell it to store its articles using
the same numbers as the upstream server; then you could easily sync
your .newsrc files.

-- 
"Unfortunately, Stupendous Man's cape is caught in Mild-Mannered
 Calvin's zipper! Curses!"
                                -- "Calvin and Hobbes", Bill Watterson

Re: The two Servers

[Raymond Scholz]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Oy.  Tough, tough.  Hm.  Does anyone else have suggestions?

I've asked for something similar about a year ago.  There are Slave
Gnusae, which solve the problem if you're reading from the same server
(using article numbers).  I wonder if it would be possible to extend
this approach to message IDs somehow.

Cheers, Ray
-- 
Look Ma, this man can twist his fingers as if they were made of rubber,
isn't that amazing? -- Not really, he's been using Emacs for years...!

Re: The two Servers

[Raymond Scholz]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Oy.  Tough, tough.  Hm.  Does anyone else have suggestions?

I've asked for something similar about a year ago.  There are Slave
Gnusae, which solve the problem if you're reading from the same server
(using article numbers).  I wonder if it would be possible to extend
this approach to message IDs somehow.

Cheers, Ray
-- 
Look Ma, this man can twist his fingers as if they were made of rubber,
isn't that amazing? -- Not really, he's been using Emacs for years...!

Re: The two Servers

[Kai Großjohann]

Frank Schmitt <usereplyto@Frank-Schmitt.net> writes:

> Number two: At university I've got no admin rights and have a quota of
> only 50mb which I exceed almost daily so I can't set up a newsserver.

Oy.  Tough, tough.  Hm.  Does anyone else have suggestions?

kai
-- 
Silence is foo!

Re: The two Servers

[Kai Großjohann]

Frank Schmitt <usereplyto@Frank-Schmitt.net> writes:

> Number two: At university I've got no admin rights and have a quota of
> only 50mb which I exceed almost daily so I can't set up a newsserver.

Oy.  Tough, tough.  Hm.  Does anyone else have suggestions?

kai
-- 
Silence is foo!

Re: The two Servers

[Frank Schmitt]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

>>I'm at the moment in the bad situation of reading news at university
>>from their nntp server and at home from my local newsserver. This is a
>>pain because the articles I read at university are unread at home and
>>vice versa.
>>
>>Therefor I want a solution which solves this. That means articles
>>numbers are evil, Message-IDs are good. I thought of two solutions:
>
>I suggest that you set up your own news server, for example with the
>program Leafnode.  Then you can tell Leafnode to fetch from both
>servers, and you can carry around the data directory with rsync or
>something like this.

I wished it was this easy. Problem number one: At home I've got a dialup
connection and I don't like Gnus agent, so I've got to use a local nntp
server (That's hamster in my case).
Number two: At university I've got no admin rights and have a quota of
only 50mb which I exceed almost daily so I can't set up a newsserver.

-- 
One Ring to rule them all, One Ring to find them,
One Ring to bring them all and in the darkness bind them
In the Land of Mordor where the Shadows lie.
19. Dezember 2001

Re: The two Servers

[Frank Schmitt]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

>>I'm at the moment in the bad situation of reading news at university
>>from their nntp server and at home from my local newsserver. This is a
>>pain because the articles I read at university are unread at home and
>>vice versa.
>>
>>Therefor I want a solution which solves this. That means articles
>>numbers are evil, Message-IDs are good. I thought of two solutions:
>
>I suggest that you set up your own news server, for example with the
>program Leafnode.  Then you can tell Leafnode to fetch from both
>servers, and you can carry around the data directory with rsync or
>something like this.

I wished it was this easy. Problem number one: At home I've got a dialup
connection and I don't like Gnus agent, so I've got to use a local nntp
server (That's hamster in my case).
Number two: At university I've got no admin rights and have a quota of
only 50mb which I exceed almost daily so I can't set up a newsserver.

-- 
One Ring to rule them all, One Ring to find them,
One Ring to bring them all and in the darkness bind them
In the Land of Mordor where the Shadows lie.
19. Dezember 2001

Re: The two Servers

[Kai Großjohann]

Frank Schmitt <usereplyto@Frank-Schmitt.net> writes:

> I'm at the moment in the bad situation of reading news at university
> from their nntp server and at home from my local newsserver. This is a
> pain because the articles I read at university are unread at home and
> vice versa.
>
> Therefor I want a solution which solves this. That means articles
> numbers are evil, Message-IDs are good. I thought of two solutions:

I suggest that you set up your own news server, for example with the
program Leafnode.  Then you can tell Leafnode to fetch from both
servers, and you can carry around the data directory with rsync or
something like this.

kai
-- 
Silence is foo!

Re: The two Servers

[Kai Großjohann]

Frank Schmitt <usereplyto@Frank-Schmitt.net> writes:

> I'm at the moment in the bad situation of reading news at university
> from their nntp server and at home from my local newsserver. This is a
> pain because the articles I read at university are unread at home and
> vice versa.
>
> Therefor I want a solution which solves this. That means articles
> numbers are evil, Message-IDs are good. I thought of two solutions:

I suggest that you set up your own news server, for example with the
program Leafnode.  Then you can tell Leafnode to fetch from both
servers, and you can carry around the data directory with rsync or
something like this.

kai
-- 
Silence is foo!

The two Servers

[Frank Schmitt]

Hello

I'm at the moment in the bad situation of reading news at university
from their nntp server and at home from my local newsserver. This is a
pain because the articles I read at university are unread at home and
vice versa.

Therefor I want a solution which solves this. That means articles
numbers are evil, Message-IDs are good. I thought of two solutions:

Either create a second .newsrc file (for example ~/newsrc.id) and when
an entry in ~/.newsrc is made an entry in ~/.newsrc.id is created which
says which Message-ID belongs to this article number.
When getting the overview one had to parse it for already known
Message-IDs.
This solution is probably much and hard work.

Second idea I had is this one: define something like gnus-get-ids which
gets the Message-IDs for all read articles and gnus-read-ids which gets
a list of Message-IDs, retrieves the article numbers of the corresponding
articles and writes those in ~/.newsrc.
This seems to be not to hard.

I searched the documentation and found that gnus-move.el does something
very similar but there is one thing I really don't understand: Why the
hell does it retrieve the whole head of the article? Wouldn't it be *much*
faster to use the stat command?

Anybody out there who has done something similar before or has an other
idea how to do it?

-- 
One Ring to rule them all, One Ring to find them,
One Ring to bring them all and in the darkness bind them
In the Land of Mordor where the Shadows lie.
19. Dezember 2001