Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[philo]

--- weigelt@metux.de wrote:

From: Enrico Weigelt <weigelt@metux.de>
To: Fans of the OS Plan 9 from Bell Labs <9fans@cse.psu.edu>
Subject: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]
Date: Tue, 11 Mar 2008 18:01:05 +0100

* Vinícius de Figueiredo Silva <viniciusfs@gmail.com> wrote:

Hi folks,

> Abaco's website is now hosted at http://abaco.oitobits.net ;)

I've just tried it on plan9port, but running mk failed due some
missing fields in Irule.
After some futher looks I discovered another source within the
"abaco" subdir, which seems to build fine. 
It seems there are two different versions in the tree, which
should be cleaned up.

Trying to run abaco didn't work - it says it cannot connect
webfs. In fact, webfs doesn't seem to create an server socket
(in /tmp/ns.$LOGNAME.:0/).

Could someone give me an advise, what's wrong here ?



Hello:
I am a Plan 9 newbie with very little experience but have recently used Abaco.
I had checked some prior posts and and saw someone else had trouble compliling...
so I ended up merely downloading the pre-compiled version

It worked fine ...but prior to running it I had to run   webfs

Running Abaco by itself does not automatically invoke webfs

and for some sites it was also necessary to enable cookies



Also: I am making the assumption the your net card was properly detected
and you started your ip connection and bound DNS

_____________________________________________________________
-. www.tuol.org .-

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Pietro Gagliardi]

Try:

	webcookies
	webfs
	abaco

It should work then.

I think abaco should be made to host its own webcookies/webfs. Why   
not? It's this easy:

	if ((cookiesproc = fork()) == 0) {
		execl("/bin/webcookies", "webcookies", 0);
		sysfatal("fork");
		break;
	} else if (cookiesproc < 0)
		sysfatal("fork");
	if ((fsproc = fork()) == 0) {
		execl("/bin/webfs", "webfs", 0);
		sysfatal("fork");
		break;
	} else if (fsproc < 0)
		sysfatal("fork");

On Mar 11, 2008, at 6:29 PM, philo wrote:

>
>
> --- weigelt@metux.de wrote:
>
> From: Enrico Weigelt <weigelt@metux.de>
> To: Fans of the OS Plan 9 from Bell Labs <9fans@cse.psu.edu>
> Subject: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco  
> sources?]
> Date: Tue, 11 Mar 2008 18:01:05 +0100
>
> * Vinícius de Figueiredo Silva <viniciusfs@gmail.com> wrote:
>
> Hi folks,
>
>> Abaco's website is now hosted at http://abaco.oitobits.net ;)
>
> I've just tried it on plan9port, but running mk failed due some
> missing fields in Irule.
> After some futher looks I discovered another source within the
> "abaco" subdir, which seems to build fine.
> It seems there are two different versions in the tree, which
> should be cleaned up.
>
> Trying to run abaco didn't work - it says it cannot connect
> webfs. In fact, webfs doesn't seem to create an server socket
> (in /tmp/ns.$LOGNAME.:0/).
>
> Could someone give me an advise, what's wrong here ?
>
>
>
> Hello:
> I am a Plan 9 newbie with very little experience but have recently  
> used Abaco.
> I had checked some prior posts and and saw someone else had trouble  
> compliling...
> so I ended up merely downloading the pre-compiled version
>
> It worked fine ...but prior to running it I had to run   webfs
>
> Running Abaco by itself does not automatically invoke webfs
>
> and for some sites it was also necessary to enable cookies
>
>
>
> Also: I am making the assumption the your net card was properly  
> detected
> and you started your ip connection and bound DNS
>
> _____________________________________________________________
> -. www.tuol.org .-

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Steve Simon]

> I think abaco should be made to host its own webcookies/webfs. Why
> not?

I think this is a bad idea, what if you want to use an alternate
webfs (on a different NIC), or an non-standard cookies file? do you
want to wait whilst webcookies rescans it databse at startup and
webfs rescans its cache (work in progress)?

If we continue this way why not put the code for webfs and and webcookies
in abaco, and why not include upas and nntpfs too; I guess you can see where
this is leading...

I think fgb's simple shell script is an elegant solution, if this is what
you want (sh to rc translation not withstanding) but keeping webfs and webcookies
as long lived external servers has significant benefits - it is The plan9 way™
after all.

-Steve

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[erik quanstrom]

> I think this is a bad idea, what if you want to use an alternate
> webfs (on a different NIC), or an non-standard cookies file? do you
> want to wait whilst webcookies rescans it databse at startup and
> webfs rescans its cache (work in progress)?
>
> If we continue this way why not put the code for webfs and and webcookies
> in abaco, and why not include upas and nntpfs too; I guess you can see where
> this is leading...
>
> I think fgb's simple shell script is an elegant solution, if this is what
> you want (sh to rc translation not withstanding) but keeping webfs and webcookies
> as long lived external servers has significant benefits - it is The plan9 way™
> after all.

standard slippery slope argument.

i think that abaco (by inheratance from webfs) may confuse
elegance with unfriendliness.  for example, why do i have
to type "http://"?  why can't i type "g $query" to google
something?  why doesn't esc in the tag highlight like acme?

abaco is great stuff.  this is why i took the time to add a few
of these things to my version.  they might not be the height
of elegance.  they may not be added in the right place.  perhaps
webfs should do this.  but usability is important, too.

- erik

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Iruata Souza]

On Wed, Mar 12, 2008 at 10:36 AM, erik quanstrom <quanstro@coraid.com> wrote:
> > I think this is a bad idea, what if you want to use an alternate
>  > webfs (on a different NIC), or an non-standard cookies file? do you
>  > want to wait whilst webcookies rescans it databse at startup and
>  > webfs rescans its cache (work in progress)?
>  >
>  > If we continue this way why not put the code for webfs and and webcookies
>  > in abaco, and why not include upas and nntpfs too; I guess you can see where
>  > this is leading...
>  >
>  > I think fgb's simple shell script is an elegant solution, if this is what
>  > you want (sh to rc translation not withstanding) but keeping webfs and webcookies
>  > as long lived external servers has significant benefits - it is The plan9 way™
>  > after all.
>
>  standard slippery slope argument.
>
>  i think that abaco (by inheratance from webfs) may confuse
>  elegance with unfriendliness.  for example, why do i have
>  to type "http://"?  why can't i type "g $query" to google
>  something?  why doesn't esc in the tag highlight like acme?
>
>  abaco is great stuff.  this is why i took the time to add a few
>  of these things to my version.  they might not be the height
>  of elegance.  they may not be added in the right place.  perhaps
>  webfs should do this.  but usability is important, too.
>

have you sent your patches to fgb?

iru

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[erik quanstrom]

>
> have you sent your patches to fgb?
>
> iru

i think so.  but it's been a very long time.
here's the entire patch to urls.c  some fixups
may be required due to this fact.  attached
is the whole patch to urls.c.

- erik

// tired of typing http://, tired of going to google first.
void
justgoogleit(Url *u)
{
	Rune *s;

	s = ucvt(u->src.r+2);
	free(u->src.r);
	u->src.r = runesmprint("http://www.google.com/search?hl=en&ie=UTF-8&q=%S", s);
	free(s);
	u->src.nr = runestrlen(u->src.r);
}

void
addhttp(Url *u)
{
	Rune *s;
	if(validurl(u->src.r))
		return;
	s = u->src.r;
	u->src.r = runesmprint("http://%S", u->src.r);
	free(s);
	u->src.nr = runestrlen(u->src.r);
}

struct{
	void	(*f)(Url*);
	Rune	*lead;
	int	len;
} ctab[] = {
	justgoogleit,	L"g ",		2,
	addhttp,		L"",		0,
};

void
urlconvience(Url *u)
{
	int i;

	for(i = 0; u->src.nr >= ctab[i].len && runestrncmp(u->src.r, ctab[i].lead, ctab[i].len) != 0; i++)
		;
	ctab[i].f(u);
}

int
urlopen(Url *u)
{
	char buf[BUFSIZE];
	int cfd, fd, conn, n;

	urlconvience(u);

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Federico G. Benavento]

hola,

"http://9fans.net" is an URL, "9fans.net" is not.

as for google, I added a "Google string" command sometime
ago.

--
Federico G. Benavento

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Pietro Gagliardi]

Here's another thing I'd like to see: preview on-disk pages:

	abaco $home/x.html

On Mar 12, 2008, at 1:54 PM, Federico G. Benavento wrote:

> hola,
>
> "http://9fans.net" is an URL, "9fans.net" is not.
>
> as for google, I added a "Google string" command sometime
> ago.
>
> --
> Federico G. Benavento
>

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[erik quanstrom]

> Here's another thing I'd like to see: preview on-disk pages:
>
> 	abaco $home/x.html
>

that's not abaco -- that's webfs.  webfs doesn't know what to do with "file://".

- erik

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Federico G. Benavento]

hola,

>         abaco $home/x.html

maybe, abaco file://$home/x.html

feel free to add the file protocol to webfs.

>  On Mar 12, 2008, at 1:54 PM, Federico G. Benavento wrote:
>
>  > hola,
>  >
>  > "http://9fans.net" is an URL, "9fans.net" is not.
>  >
>  > as for google, I added a "Google string" command sometime
>  > ago.
>  >
>  > --
>  > Federico G. Benavento
>  >
>
>
>



--
Federico G. Benavento

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[erik quanstrom]

> is there anything that breaks if we take an adhoc approach to ignoring
> standards? who needs URLs anyway if you are too lazy for them. you are
> not too lazy to complain.

i don't see how providing code & and argument for the code
is complaining.  i didn't bring this subject up.

as i see it, convienence takes text that is nor a valid uri and
converts it to a valid uri.  how does this qualify as ignoring the
standard?

- erik

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Bruce Ellis]

is there anything that breaks if we take an adhoc approach to ignoring
standards? who needs URLs anyway if you are too lazy for them. you are
not too lazy to complain.

On Thu, Mar 13, 2008 at 10:08 AM, erik quanstrom <quanstro@quanstro.net> wrote:
> > hola,
> >
> > "http://9fans.net" is an URL, "9fans.net" is not.
> >
> > as for google, I added a "Google string" command sometime
> > ago.
>
> what's the advantage of taking that position?  is there something
> that breaks if "9fans.net" is interpreted as "http://9fans.net"?
>
> "Google string" is unsatisfactory to me for several reasons.  first,
> it opens up a new window.  second, it is much more cumbersome
> because you have to select the whole search.  with "g term" you
> can double click at the end of an existing url and type "g term<enter>"
> and you're done.  no extra trip to the mouse.
>
> but that's just what works for me, and i'm lazy.
>
> - erik
>
>

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[erik quanstrom]

> hola,
>
> "http://9fans.net" is an URL, "9fans.net" is not.
>
> as for google, I added a "Google string" command sometime
> ago.

what's the advantage of taking that position?  is there something
that breaks if "9fans.net" is interpreted as "http://9fans.net"?

"Google string" is unsatisfactory to me for several reasons.  first,
it opens up a new window.  second, it is much more cumbersome
because you have to select the whole search.  with "g term" you
can double click at the end of an existing url and type "g term<enter>"
and you're done.  no extra trip to the mouse.

but that's just what works for me, and i'm lazy.

- erik

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Enrico Weigelt]

* Federico G. Benavento <benavento@gmail.com> wrote:
> #!/bin/sh
> if [ ! -f $HOME/.webcookies ]; then
>   touch $HOME/.webcookies
> fi
> if [ ! -e `namespace`/web ]; then
>   webfs -s web
> fi
> # export PLAN9=/usr/share/abacoabaco.bin $*

ah, the "-s web" option was missing :)

now another problem: abaco expects some fonts in
/usr/local/plan9/fonts.

Seems we need some convenient way for relocating fonts.
Maybe an even an fontserver ? ;-)


cu
--
---------------------------------------------------------------------
 Enrico Weigelt    ==   metux IT service - http://www.metux.de/
---------------------------------------------------------------------
 Please visit the OpenSource QM Taskforce:
 	http://wiki.metux.de/public/OpenSource_QM_Taskforce
 Patches / Fixes for a lot dozens of packages in dozens of versions:
	http://patches.metux.de/
---------------------------------------------------------------------

Re: [9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Federico G. Benavento]

#!/bin/sh
if [ ! -f $HOME/.webcookies ]; then
	touch $HOME/.webcookies
fi
if [ ! -e `namespace`/web ]; then
	webfs -s web
fi

# export PLAN9=/usr/share/abaco
abaco.bin $*


On Tue, Mar 11, 2008 at 2:01 PM, Enrico Weigelt <weigelt@metux.de> wrote:
> * Vinícius de Figueiredo Silva <viniciusfs@gmail.com> wrote:
>
>  Hi folks,
>
>  > Abaco's website is now hosted at http://abaco.oitobits.net ;)
>
>  I've just tried it on plan9port, but running mk failed due some
>  missing fields in Irule.
>  After some futher looks I discovered another source within the
>  "abaco" subdir, which seems to build fine.
>  It seems there are two different versions in the tree, which
>  should be cleaned up.
>
>  Trying to run abaco didn't work - it says it cannot connect
>  webfs. In fact, webfs doesn't seem to create an server socket
>  (in /tmp/ns.$LOGNAME.:0/).
>
>  Could someone give me an advise, what's wrong here ?
>
>
>  cu
>  --
>  ---------------------------------------------------------------------
>   Enrico Weigelt    ==   metux IT service - http://www.metux.de/
>  ---------------------------------------------------------------------
>   Please visit the OpenSource QM Taskforce:
>         http://wiki.metux.de/public/OpenSource_QM_Taskforce
>   Patches / Fixes for a lot dozens of packages in dozens of versions:
>         http://patches.metux.de/
>  ---------------------------------------------------------------------
>
>



-- 
Federico G. Benavento

[9fans] abaco @ plan9port [WAS: Hi and, plan9-native abaco sources?]

[Enrico Weigelt]

* Vinícius de Figueiredo Silva <viniciusfs@gmail.com> wrote:

Hi folks,

> Abaco's website is now hosted at http://abaco.oitobits.net ;)

I've just tried it on plan9port, but running mk failed due some
missing fields in Irule.
After some futher looks I discovered another source within the
"abaco" subdir, which seems to build fine. 
It seems there are two different versions in the tree, which
should be cleaned up.

Trying to run abaco didn't work - it says it cannot connect
webfs. In fact, webfs doesn't seem to create an server socket
(in /tmp/ns.$LOGNAME.:0/).

Could someone give me an advise, what's wrong here ?


cu
-- 
---------------------------------------------------------------------
 Enrico Weigelt    ==   metux IT service - http://www.metux.de/
---------------------------------------------------------------------
 Please visit the OpenSource QM Taskforce:
 	http://wiki.metux.de/public/OpenSource_QM_Taskforce
 Patches / Fixes for a lot dozens of packages in dozens of versions:
	http://patches.metux.de/
---------------------------------------------------------------------