[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: w3m man file : option -m, document encoding

to continue a batch of mails

Justin B Rye schrieb am 29. Oct 2014 um 12:36

> Sorry about the out-of-sequence replies; Gmail decided this one was
> spam for some reason...

No problem. I am at the edge of loosing control of the
correspondence. Hoping that nothing gets lost.

> > Justin B Rye schrieb am 28. Oct 2014 um 15:47
> [...]
> >>  w3m -m nntp://news.demon.co.uk/linux.debian.curiosa/3511
> >> 
> >> It works!  But I have no idea why anyone would want to do this; it
> >> isn't threaded, it doesn't support marking messages as "seen" (let
> >> alone spam-scoring), and the message numbers don't correspond to
> >> anything useful.
> > 
> > Apparently, we have two different versions of w3m. I get an error message:
> > 
> >   hiereth@lune:/tmp$  w3m -m nntp://news.demon.co.uk/linux.debian.curiosa/3511
> >   w3m: Can't load nntp://news.demon.co.uk/linux.debian.curiosa/3511.
> > 
> > Or is it necessary to open a special port for nntp communication?
> The news.demon.co.uk server only allows connections from *.demon.co.uk
> accounts.  There are open-access ones, but most of them require at
> least a registration.
> > Version and system information here:
> > 
> >   3$ w3m -version
> >   w3m version w3m/0.5.3+cvs-1.1055, options
> >   lang=en,m17n,image,color,ansi-color,mouse,gpm,menu,cookie,ssl,ssl-verify,
> >   external-uri-loader,w3mmailer,nntp,gopher,ipv6,alarm,mark,migemo
>                                   ^^^^
> So it should work if you can find a server!  Try
>   w3m -m nntp://news.aioe.org/linux.debian.curiosa/

This server served. 

> >   Package: w3m
> >   Version: 0.5.3-8
> > 
> >   Debian Release: 7.6
> You are in fact slightly out of date - 7.7 was released on the 18th,
> but w3m hasn't needed a stable update since I-don't-know-when.
> >     APT prefers stable
> >     APT policy: (500, 'stable')
> I also have stable-updates listed here.

I know that the best-before date vor 7.6 is already over. But I am
reluctant with new versions. Really would prefer to upgrade only from
7 to 8.

> >   Architecture: i386 (i686)
> >   
> >   Kernel: Linux 3.2.0-4-486
> Not 3.2.0-4-686-pae?  I've still got one remaining twentieth century
> machine in service (with dual 450MHz Intel Pentium II Xeon CPUs), and
> even that can run the -686-pae flavour...

OK. If you want to motivate me to replace or to install 686-pae as
extra kernel. pass a link what improvements can be expected.

> >   Locale: LANG=de_DE, LC_CTYPE=de_DE (charmap=ISO-8859-1)
> And I thought I was being slow when I switched over from ISO-8859-15
> to UTF-8 in Etch (Debian 4.0)!

I know. I wanted to avoid having text files with two different
encodings in my archive. Searching for keywords with German "Umlaute"
äüöÖÄÜ would get more complicated.

> >   Shell: /bin/sh linked to /bin/dash
> >   
> >   Versions of packages w3m depends on:
> >   ii  libc6        2.13-38+deb7u2
> Update available:    2.13-38+deb7u6
> >   ii  libgc1c2     1:7.1-9.1
> >   ii  libgpm2      1.20.4-6
> >   ii  libssl1.0.0  1.0.1e-2+deb7u11
> Update available:    1.0.1e-2+deb7u13
> >   ii  libtinfo5    5.9-10
> >   ii  zlib1g       1:1.2.7.dfsg-13
> >   
> >   Versions of packages w3m recommends:
> >   ii  ca-certificates  20130119
> Update available:        20130119+deb7u1
> >   Versions of packages w3m suggests:
> >   ii  man-db        2.6.2-1
> >   pn  menu          <none>
> >   pn  migemo        <none>
> >   ii  mime-support  3.52-1
> >   pn  w3m-el        <none>
> >   ii  w3m-img       0.5.3-8
> >   
> >   -- no debconf information
> I recommend subscribing to debian-announce@lists.debian.org and adding
> wheezy-updates in your APT sources.

I already thought about it. But with respect to w3m, it would change
nothing. Version 0.5.3-17 is part of "instable". 

>>> I still don't see any sign of the -m making a difference, but I
>>> can imagine on a nineties Japanese newsgroup where different
>>> messages use a variety of different encodings signalled in the
>>> headers, the -m might have fixed problems that I would have
>>> trouble testing for.  Except... if so, why not just make that
>>> mystery behaviour the default for nntp:// arguments?

>> All these questions favour a decision to omit -m in the OPTIONS
>> section and to mention it with the other dubious options only in the
>> NOTES section.
>> What freedoms are we allowed? All this work would justify that you
>> appear as author of the last version. I would take the honour of
>> introducing myself as author of the German version.

In version_3 you'll find remarks on options presently 'not
recommended' due to the danger of spoiling time.

> I still hadn't given up hope that we could come up with something we
> could plausibly send upstream, in which case I'd leave it up to them
> to decide what goes in the AUTHORS section.

Yesterday I got an answer of a former member of the project team,
Karsten Schoelzel. I would have appreciated his support, he could have
give me explanations in German. But he wrote I is to long ago that he
was busy with w3m.

>>> And what's the email version?  Can w3m be compiled with IMAP support?
>> There is no string "IMAP" in w3m-0.5.3/configure. 

>Nor indeed "POP3".  But while I've got the sources unpacked... ah, -m
>seems to set "SearchHeader = search_header = TRUE;"... I can't follow
>how that's used, but it does seem to be about parsing message headers
>for encoding information.

Character encoding problems. They are an adequate punishment for users
that explicitly made their computer use ISO-8859-1. Like me :-) .

>>> JBR   A long time ago this practice was followed, especially in the
>>>       upper classes, but today even the children of the lower classes
>>>       perform no executions, and this is extreme negligence.
>>>                               - "Hagakure", Yamamoto Tsunetomo (1716)

>> A strange idea about the things that children should do.
>It's a book of really bad advice.
Chaqun à son goût. Jeder nach seinem Geschmack.

To be continued

Reply to: