slashcode-development — This list is for development of slashcode, no install questions please.

Re: [Slashcode-development] Q: Can Slash discussions be served in a format other than HTML?

[CertIndex.com W. <web...@ce...>]

Even if this isn't a direct function of Slash, I think you could easily get this same end by writing
a perl/php/whatever script to spider the site, parse the articles' comments and generate .xml files.
What do you think of this?

ps: kudos on working in such an interesting field, glad to help ya wherever possible.

----- Original Message -----
From: "Nick Matsakis" <mat...@ai...>
To: <sla...@li...>
Sent: Monday, 16 April, 2001 12:58
Subject: [Slashcode-development] Q: Can Slash discussions be served in a format other than HTML?


>
> Greetings,
>
> My name is Nick Matsakis, and I'm a graduate student at MIT's AI
> Laboratory. Lately, I've been working on some problems in information
> retrieval, and have become very interested in Slash-based sites as a
> potential source of data to try out some techniques.
>
> I've know that Slashdot (and others) serve lists of headlines in XML, and
> these are (presumably) the basis of many of the headline viewing programs
> out there.  However, what I am interested in is the threaded comment
> discussions that each new article spawns. I've looked around a bit, and
> have not been able to find out whether or not Slash is able to provide
> third parties direct access to these discussions, rather than as HTML.
>
> Ideally, what I would want would be an XML-like document which lists each
> comment with a unique identifier, the identifier of the comment it is a
> response to, and the text of the comment.  Other metadata (author, rating,
> subject) is useful but not necessary.
>
> I realize that this question is off-topic for a developers list, but it
> seemed like the most direct way to get an answer to the yes or no question
> of "is this possible?"
>
> Regards,
>
> Nick Matsakis
>
>
>
> _______________________________________________
> Slashcode-development mailing list
> Sla...@li...
> http://lists.sourceforge.net/lists/listinfo/slashcode-development
>

Re: [Slashcode-development] comment question

[CertIndex.com W. <web...@ce...>]

thanks a lot for taking the time to address this.

let me give a scenario just to make sure i 'got it'.

you have 1000 comments in a story.

user sets his commentspill to 500 and his commentlimit to 100

since commentspill is at 500, only 500 comments are 'acknowledged'. from that 500, they are divided
into pages, 100 comments each. so the user ends up with 5 pages of 100 questions each of 500 of the
1000 comments.

correct?

(if so, is it the 500 newest or oldest? or is that determined by how they sort the story.)

thanks again


----- Original Message -----
From: "Ask Slashdot Guy" <cl...@sl...>
To: <sla...@li...>
Sent: Tuesday, 17 April, 2001 10:42
Subject: Re: [Slashcode-development] comment question


>
> On Mon, 16 Apr 2001 21:22:53 -0700, CertIndex.com Webmaster said:
>
> >  anyone know what commentspill in users_comments does? commentlimit determines the number of
> >  comments to display per page, but what does commentspill do?
> >
> >  thanks. been beating my head against this one and playing around on /. to try to figure it out
> >  for a while now.
>
> A users commentlimit is fine for light stories (stories between 100-150
> comments, I think), but for larger stories, the system needs to establish an
> override for this limit (if 50 users ask for a 500 comment article and display
> all 500 comments, you can figure out how badly the machine will get loaded. Do
> the math for yourself and see...)....
>
> So commentspill is the system limit to prevent this from occurring...at least
> as I've determined, as I was in the same spot as you are about a month ago.
>
> So commentspill takes effect when stories hit a certain number of comments and
> if commentlimit > commentspill.
>
> Hope this explains it.
>
> - Cliff
>
>
> _______________________________________________
> Slashcode-development mailing list
> Sla...@li...
> http://lists.sourceforge.net/lists/listinfo/slashcode-development
>

Re: [Slashcode-development] comment question

[Ask S. G. <cl...@sl...>]

On Mon, 16 Apr 2001 21:22:53 -0700, CertIndex.com Webmaster said:

>  anyone know what commentspill in users_comments does? commentlimit determines the number of
>  comments to display per page, but what does commentspill do?
>  
>  thanks. been beating my head against this one and playing around on /. to try to figure it out
>  for a while now.

A users commentlimit is fine for light stories (stories between 100-150
comments, I think), but for larger stories, the system needs to establish an
override for this limit (if 50 users ask for a 500 comment article and display
all 500 comments, you can figure out how badly the machine will get loaded. Do
the math for yourself and see...)....

So commentspill is the system limit to prevent this from occurring...at least
as I've determined, as I was in the same spot as you are about a month ago.

So commentspill takes effect when stories hit a certain number of comments and 
if commentlimit > commentspill.

Hope this explains it.

- Cliff

[Slashcode-development] comment question

[CertIndex.com W. <web...@ce...>]

 anyone know what commentspill in users_comments does? commentlimit determines the number of
comments to display per page, but what does commentspill do?

thanks. been beating my head against this one and playing around on /. to try to figure it out for a
while now

[Slashcode-development] Oracle Port Progress, April 16: hello world

[Stephen C. <the...@wa...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I end my work on Slash today with milestone #2: the front page displays in all
its glory.  Doesn't get too far past that :)  But it's still nice to see the
meat of the system working.

The foolish inconsistency of the day involves fetchrow_hashref.  I was seeing
successful template fetches but the final output came out completely blank.  The
problem stems from fetchrow_hashref using (by default) $sth->{NAME} as the
column list to get the hash keys.  In Oracle, the keys returned by $sth->{NAME}
are always upper case, and as you have probably already guessed, all the key
requests in Slash::Display::Provider are lower case.

The fix is to change the calls to fetchrow_hashref in Slash::DB::Utility to:

... ->fetchrow_hashref('NAME_lc');

What I'm curious about is if this change breaks anything on the MySQL end, since
I've made it in Slash::DB::Utility (and I'm still too lazy to get MySQL set up
locally to actually test these things myself).  If anyone cares to patch
- --by-hand Slash::DB::Utility (there are three occurences of fetchrow_hashref in
there), I'd be curious to know the results.

Well, this was a welcome break, but sadly my real job summons me to perform late
night server upgrades.  More progress tomorrow (we hope :)

- -- 
Stephen Clouse <the...@wa...>
warpcore.org Founder, Chief Megalomaniac, and Evil Overlord

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8

iQA/AwUBOtumZt1EXk7JbKbMEQJ/NACeOmJowLG4E7O9tWq+kInN7fbrjx0An1NS
z3jyNwQ+jeNI42xgd4kVM76Q
=7ayH
-----END PGP SIGNATURE-----

[Slashcode-development] Q: Can Slash discussions be served in a format other than HTML?

[Nick M. <mat...@ai...>]

Greetings,

My name is Nick Matsakis, and I'm a graduate student at MIT's AI
Laboratory. Lately, I've been working on some problems in information
retrieval, and have become very interested in Slash-based sites as a
potential source of data to try out some techniques.

I've know that Slashdot (and others) serve lists of headlines in XML, and
these are (presumably) the basis of many of the headline viewing programs
out there.  However, what I am interested in is the threaded comment
discussions that each new article spawns. I've looked around a bit, and
have not been able to find out whether or not Slash is able to provide
third parties direct access to these discussions, rather than as HTML.  

Ideally, what I would want would be an XML-like document which lists each
comment with a unique identifier, the identifier of the comment it is a
response to, and the text of the comment.  Other metadata (author, rating,
subject) is useful but not necessary.  

I realize that this question is off-topic for a developers list, but it
seemed like the most direct way to get an answer to the yes or no question
of "is this possible?"

Regards,

Nick Matsakis

Re: [Slashcode-development] Oracle Port Progress

[Ask S. G. <cl...@sl...>]

On Sun, 15 Apr 2001 20:01:49 -0500, Stephen Clouse said:

> -----BEGIN PGP SIGNED MESSAGE-----
>  Hash: SHA1
>  
>  Yee-haw, as they say down south.  install-slashsite now survives without a
>  single error (using the slashcode theme, anyway).
>  
>  The placeholder idea I described last week is in here as well, and works as
>  expected.  All the Slash::DB::MySQL code that hasn't been overridden for Oracle
>  continues to work, which makes the transition quite a bit easier.  Of course,
>  cleaning up the MySQL code with the newly-extended functions is still in the
>  works.  I need to actually fire things up with a MySQL server and make sure
>  nothing has broken horribly.  I might post a patch with just the
>  Slash::DB::Utility changes and see if I can get some QA on it.
>  
>  The most annoying problem at the moment is dealing with the column names that
>  were altered due to Oracle reserved word conflicts.  I'm very tempted to try 
>  hacking this in Slash::DB::Utility (letting it surreptitiously alter the field
>  names passed to the SQL generator functions), because I'm having to override
>  entire subroutines (that would work otherwise) to change three letters.  Is it
>  at all possible to synchronize the column names between the different ports?  Or 
>  are we too far along at this point?  The changes were fairly minor (this ripped
>  >from the notes at the top of my schema create script):
>  
>  #       - *(uid) => user_id
>  #       - *(comment) => comment_text
>  #       - *(mode) => comment_mode
>  #       - comments(date) => comment_date
>  #       - pollquestions(date) => poll_date
>  #       - sessions(session) => session_id
>  
>  OTOH, there's no guarantee that the next database deserving of a port won't have
>  conflicts as well, so it's probably a moot point.  I'll try to find some way of 
>  munging the field names transparently in the sql* functions.
>  
>  This week I'll actually fire up Apache and see what falls apart -- most likely a
>  lot :)  Even so, I'm feeling rather triumphant at the moment.  You couldn't 
>  finish installing on Oracle a week ago :)

Steven,

Congratulations! That's one heck of an accomplishment!

My thoughts on the field names: aside from the table creation scripts (these
must be made with the proper name for your backend), it shouldn't be too hard
to have the Backend Specific module take key names (some methods of the
Slash::DB objects refer to key names and these must be converted as well!) and
field names and maps them to something appropriate to the backend. 

It's best to leave these on a per module basis, since not all modules will need
this. Maybe we can have some facility in Slash::Utility that can handle this if
necessary. I'll look into this sometime this week and see if the team can't
figure out a clean way to handle this.

I would appreciate a preview copy of your patch, though, so I can see what's
been done so far (and fit what fixes I may need to make into the framework of
your patch).

Thanks a lot for the hard work!

- Cliff

[Slashcode-development] Oracle Port Progress

[Stephen C. <the...@wa...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yee-haw, as they say down south.  install-slashsite now survives without a
single error (using the slashcode theme, anyway).

The placeholder idea I described last week is in here as well, and works as
expected.  All the Slash::DB::MySQL code that hasn't been overridden for Oracle
continues to work, which makes the transition quite a bit easier.  Of course,
cleaning up the MySQL code with the newly-extended functions is still in the
works.  I need to actually fire things up with a MySQL server and make sure
nothing has broken horribly.  I might post a patch with just the
Slash::DB::Utility changes and see if I can get some QA on it.

The most annoying problem at the moment is dealing with the column names that
were altered due to Oracle reserved word conflicts.  I'm very tempted to try 
hacking this in Slash::DB::Utility (letting it surreptitiously alter the field
names passed to the SQL generator functions), because I'm having to override
entire subroutines (that would work otherwise) to change three letters.  Is it
at all possible to synchronize the column names between the different ports?  Or 
are we too far along at this point?  The changes were fairly minor (this ripped
from the notes at the top of my schema create script):

#       - *(uid) => user_id
#       - *(comment) => comment_text
#       - *(mode) => comment_mode
#       - comments(date) => comment_date
#       - pollquestions(date) => poll_date
#       - sessions(session) => session_id

OTOH, there's no guarantee that the next database deserving of a port won't have
conflicts as well, so it's probably a moot point.  I'll try to find some way of 
munging the field names transparently in the sql* functions.

This week I'll actually fire up Apache and see what falls apart -- most likely a
lot :)  Even so, I'm feeling rather triumphant at the moment.  You couldn't 
finish installing on Oracle a week ago :)

- -- 
Stephen Clouse <the...@wa...>
warpcore.org Founder, Chief Megalomaniac, and Evil Overlord

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8

iQA/AwUBOtpEe91EXk7JbKbMEQKDOgCgiX27RLCINORa2w4l5ljyEU1tkGkAoKFL
WEi+ljbw/6rcVLS22atbKEOX
=s3Gq
-----END PGP SIGNATURE-----

Re: [Slashcode-development] IndexHandler

[Eric D. <eri...@ja...>]

I haven't seen any problems that I experienced in older versions. Seems to work
fine.

Chris Nandor wrote:

> At 10:18 -0400 2001.04.11, shane wrote:
> >At 12:38 AM 4/11/2001 -0400, you wrote:
> >>At 23:23 -0400 2001.04.10, shane wrote:
> >> >At 04:42 PM 4/10/2001 -0400, you wrote:
> >> >>OK, several people had problems with IndexHandler working.  If you want to
> >> >>try out the latest from CVS and give it a shot, please do and report back
> >> >>to us if it is working or not.  Just update and install the new code, it
> >> >>crosses a few different modules.  Thanks,
> >> >
> >> >Looks like it's fixed! Thanks!
> >>
> >>Great, thanks a lot for your help.
> >
> >Out of curiosity, what ended up being the fix for this?
>
> It was basically just a little bit of weirdness in the API, I think.  We
> had to se $r->uri along with $r->filename, or else Apache would get upset
> with us.
>
> --
> Chris Nandor                      pu...@po...    http://pudge.net/
> Open Source Development Network    pu...@os...     http://osdn.com/
>
> _______________________________________________
> Slashcode-development mailing list
> Sla...@li...
> http://lists.sourceforge.net/lists/listinfo/slashcode-development

--
Back up my hard disk? I can't find the reverse switch!

Eric Dannewitz - Adventurer, saxophonist, good-timer (crook? quite possibly),
clarinetist, manic self-publicist, part-time flautist(flutist?), macintosher, and
often thought to be completely out to lunch. http://www.jazz-sax.com

Re: [Slashcode-development] IndexHandler

[Chris N. <pu...@po...>]

At 10:18 -0400 2001.04.11, shane wrote:
>At 12:38 AM 4/11/2001 -0400, you wrote:
>>At 23:23 -0400 2001.04.10, shane wrote:
>> >At 04:42 PM 4/10/2001 -0400, you wrote:
>> >>OK, several people had problems with IndexHandler working.  If you want to
>> >>try out the latest from CVS and give it a shot, please do and report back
>> >>to us if it is working or not.  Just update and install the new code, it
>> >>crosses a few different modules.  Thanks,
>> >
>> >Looks like it's fixed! Thanks!
>>
>>Great, thanks a lot for your help.
>
>Out of curiosity, what ended up being the fix for this?

It was basically just a little bit of weirdness in the API, I think.  We
had to se $r->uri along with $r->filename, or else Apache would get upset
with us.

-- 
Chris Nandor                      pu...@po...    http://pudge.net/
Open Source Development Network    pu...@os...     http://osdn.com/

Re: [Slashcode-development] IndexHandler

[shane <sh...@lo...>]

At 12:38 AM 4/11/2001 -0400, you wrote:
>At 23:23 -0400 2001.04.10, shane wrote:
> >At 04:42 PM 4/10/2001 -0400, you wrote:
> >>OK, several people had problems with IndexHandler working.  If you want to
> >>try out the latest from CVS and give it a shot, please do and report back
> >>to us if it is working or not.  Just update and install the new code, it
> >>crosses a few different modules.  Thanks,
> >
> >Looks like it's fixed! Thanks!
>
>Great, thanks a lot for your help.

Out of curiosity, what ended up being the fix for this?

Re: [Slashcode-development] IndexHandler

[Chris N. <pu...@po...>]

At 22:31 -0700 2001.04.10, Eric Dannewitz wrote:
>Do we need to reinstall the CPAN modules as well?

No, that shouldn't be necessary.

Although, personally, I peridoically do upgrade all my modules.  I can't
recommend that, though, because occasionally an upgrade breaks things.  :-)
However, for those that care, this is my normal procedure:

* run CPAN shell
* type "r" and return
* get list of all modules that can be upgraded
* run "install MODULE" for each one I want to upgrade

-- 
Chris Nandor                      pu...@po...    http://pudge.net/
Open Source Development Network    pu...@os...     http://osdn.com/

Re: [Slashcode-development] IndexHandler

[Eric D. <eri...@ja...>]

Do we need to reinstall the CPAN modules as well?

Chris Nandor wrote:

> OK, several people had problems with IndexHandler working.  If you want to
> try out the latest from CVS and give it a shot, please do and report back
> to us if it is working or not.  Just update and install the new code, it
> crosses a few different modules.  Thanks,
>
> --
> Chris Nandor                      pu...@po...    http://pudge.net/
> Open Source Development Network    pu...@os...     http://osdn.com/
>
> _______________________________________________
> Slashcode-development mailing list
> Sla...@li...
> http://lists.sourceforge.net/lists/listinfo/slashcode-development

--
Back up my hard disk? I can't find the reverse switch!

Eric Dannewitz - Adventurer, saxophonist, good-timer (crook? quite possibly),
clarinetist, manic self-publicist, part-time flautist(flutist?), macintosher,
and often thought to be completely out to lunch. http://www.jazz-sax.com

Re: [Slashcode-development] IndexHandler

[Chris N. <pu...@po...>]

At 23:23 -0400 2001.04.10, shane wrote:
>At 04:42 PM 4/10/2001 -0400, you wrote:
>>OK, several people had problems with IndexHandler working.  If you want to
>>try out the latest from CVS and give it a shot, please do and report back
>>to us if it is working or not.  Just update and install the new code, it
>>crosses a few different modules.  Thanks,
>
>Looks like it's fixed! Thanks!

Great, thanks a lot for your help.

-- 
Chris Nandor                      pu...@po...    http://pudge.net/
Open Source Development Network    pu...@os...     http://osdn.com/

Re: [Slashcode-development] IndexHandler

[shane <sh...@lo...>]

At 04:42 PM 4/10/2001 -0400, you wrote:
>OK, several people had problems with IndexHandler working.  If you want to
>try out the latest from CVS and give it a shot, please do and report back
>to us if it is working or not.  Just update and install the new code, it
>crosses a few different modules.  Thanks,

Looks like it's fixed! Thanks!

[Slashcode-development] IndexHandler

[Chris N. <pu...@po...>]

OK, several people had problems with IndexHandler working.  If you want to
try out the latest from CVS and give it a shot, please do and report back
to us if it is working or not.  Just update and install the new code, it
crosses a few different modules.  Thanks,

-- 
Chris Nandor                      pu...@po...    http://pudge.net/
Open Source Development Network    pu...@os...     http://osdn.com/

Re: [Slashcode-development] RFC: SQL Placeholders

[Stephen C. <the...@wa...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

The extension I've come up with to implement placeholders is described
below.  The goal was not so much elegance as it was to keep from recoding half
of the Slash::DB tree, something I know you're not wanting to do so close 
to release.  Even so I think it meshes quite nicely, and gives a pretty good
illusion of elegance, assuming you don't think it is :)

Basically, instead of passing SQL with prefilled values to the generator
functions in Slash::DB::Utility, you would pass the statement elements with
placeholders, followed by an array reference containing the bind values.  So
something like this (from Slash::DB::MySQL::createComment):

	my($maxCid) = $self->sqlSelect(
		"max(cid)", "comments", "sid=$sid_db"
	);

becomes:

	my($maxCid) = $self->sqlSelect(
		"max(cid)", "comments", "sid=?", [$sid_db]
	);

Meanwhile, Slash::DB::Utility::sqlSelect looks like this:

sub sqlSelect {
	my $self = shift;
	my $sql = '';
	foreach my $elem ('SELECT ',' FROM ',' WHERE ',' ') {
		last if !@_ or ref($_[0]);
		$sql .= $elem . shift;
	}
	my $sth = $self->{_dbh}->prepare_cached($sql);
	$self->sqlConnect();
	if (!$sth->execute($_[0] ? @$_[0] : ()) {
		errorLog($sql);
		$self->sqlConnect;
		return undef;
	}
	my @r = $sth->fetchrow;
	$sth->finish;
	return @r;
}

which builds the SQL statement assuming the old parameter format until it
encounters the bind value arrayref, then passes that to execute as the bind
value set.

I picked a very simplistic example, but the syntax savings are in more complex
statements farther down that entail heavy parameter quoting.  Since bind values
don't need quoting, the excessive variable copying going on now can be
completely eliminated and the parameters passed verbatim in the arrayref.  That
alone is a huge cleanup to the MySQL code.  In the case of Oracle and PG (and
other databases that support bind values natively), any necessary quoting is
done in the server's internal API, making them work even faster.

The real necessary extension comes when binding values for insert.  Inserts
already use a hashref to carry the values over, so nothing needs to change there
(other than adjusting the internals of sqlInsert to generate a placeholder-based
statement instead).  The extension here is a second hashref containing the extra
database-dependent information.  So my current bane (the template insert):

	$self->sqlInsert('templates', $hash);

can become:

	$self->sqlInsert('templates', $hash,
		{ template => { ora_type => ORA_CLOB } }
	);

And that extra pragma for the template column would go into a bind_param call to
properly set up the Oracle statement.  The hashref extension should also be
available to the select function; although I've never run into a select
statement that needs the extra parameters, other databases might need
them for something.  And it doesn't change anything if it's not there anyway.

Admittedly this isn't the best product pitch I've done :)  But hopefully you can
extrapolate the benefits.  I've freed up my evenings this week for this task
(changing Slash::DB::Utility and Slash::DB::MySQL) if you approve of it.  I
don't expect it to take long, and the old MySQL code should keep working
anyway, since the parameters to sql* haven't really changed (just optional
ones tacked on).  I could even have Oracle doing something quasi-functional by 
this weekend, which would be a huge accomplishment, IMO.

In case you're curious about my drive for this, I'm wanting Slash2 to power my
soon-to-reactivate site, which has been floundering for well over a year and is
now beckoning me to give it a new drive core :)  Corporate politics and personal
SQL bigotry keep me from running MySQL on our main server cluster, so Oracle it
is.  I think support is close, though, once these minor design hurdles are
cleared.

- -- 
Stephen Clouse <the...@wa...>
warpcore.org Founder, Chief Megalomaniac, and Evil Overlord

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8

iQA/AwUBOtJLod1EXk7JbKbMEQI9HQCfaEybkJ4sYs2hk8uvR+wYjQX3igYAoPQt
UwSJ9PWWYM6DxDzjv8WaFRJ9
=nWpz
-----END PGP SIGNATURE-----

Re: [Slashcode-development] RFC: SQL Placeholders

[Chris N. <pu...@po...>]

Stephen, please do let us know the details of the required changes.

-- 
Chris Nandor                      pu...@po...    http://pudge.net/
Open Source Development Network    pu...@os...     http://osdn.com/

Re: [Slashcode-development] slashcode-2.0.0-pre1 debian package?

[Ask S. G. <cl...@sl...>]

On Mon, 09 Apr 2001 18:10:01 +0900, Taku YASUI said:

> Hi, all.
>  
>  I want to use slashcode-2.0.0-pre1 on debian system (stable and
>  unstable). Does anyone have its debian package?
>  
>  If no one made it, I would like to make some debs.
>  
>  Regards,

Deb packages for Slash are in the works right now.

- Cliff

[Slashcode-development] [ slashcode-Patches-414881 ] Adds mysql->postgresql sql functions

[<no...@so...>]

Patches item #414881, was updated on 2001-04-09 06:37
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=304421&aid=414881&group_id=4421

Category: MAIN branch
Group: None
Status: Open
Priority: 5
Submitted By: Alessio Bragadini (alessio)
Assigned to: Nobody/Anonymous (nobody)
Summary: Adds mysql->postgresql sql functions

Initial Comment:
This patch to the slashschema_create SQL script for
PostgreSQL adds an handful of functions to mimic
existing MySQL functions (from_unixtime, to_days, etc.)
and fixes a bug in a table definition.

The functions have been kindly contributed by Zachary
Beane <xa...@xa...> and are in the public domain.


----------------------------------------------------------------------

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=304421&aid=414881&group_id=4421

[Slashcode-development] slashcode-2.0.0-pre1 debian package?

[<ta...@de...>]

Hi, all.

I want to use slashcode-2.0.0-pre1 on debian system (stable and
unstable). Does anyone have its debian package?

If no one made it, I would like to make some debs.

Regards,
-- 
Taku YASUI <ta...@de...>

RE: [Slashcode-development] can't upload a file with multipart/form-data submissions

[Blake D. <bl...@ar...>]

$r = shift; simply means that $r will get the value of the first element of
the @_ array.  In the case below, $r will have the value of the first
argument passed to the main function.  Also, shift completely removes the
first element of the array and shifts the other values down to take its
place.

Blake Day
CTO, SpunkNetwork
bl...@sp... <mailto:bl...@sp...>
home: (770) 338-1589
mobile: (678) 485-0519


-----Original Message-----
From: sla...@li...
[mailto:sla...@li...]On Behalf Of
James Edward Marca
Sent: Sunday, April 08, 2001 12:14 AM
To: slashcode list
Subject: Re: [Slashcode-development] can't upload a file with
multipart/form-data submissions


As is always the case, I think I solved my problem.
I copied file_upload.pl from libapreq/eg/perl into my plugin's perl
code, and made it work fine by editing the print commands.

some lingering black magic---I have no idea what $r=shift means, as in:

sub main(){
...
my $r=shift;                       ## huh?
				   ## is this getting a reference to myself?
my $apr=Apache::Request->new($r);  ## doesn't seem to clash with other
				   ## calls to getCurrentForm()
my $upload=$apr->upload;           ## works great.
...
}

Sorry about the noise,
james


_______________________________________________
Slashcode-development mailing list
Sla...@li...
http://lists.sourceforge.net/lists/listinfo/slashcode-development

Re: [Slashcode-development] can't upload a file with multipart/form-data submissions

[James E. M. <jm...@tr...>]

As is always the case, I think I solved my problem.
I copied file_upload.pl from libapreq/eg/perl into my plugin's perl
code, and made it work fine by editing the print commands.  

some lingering black magic---I have no idea what $r=shift means, as in:

sub main(){  
...
my $r=shift;                       ## huh?  
				   ## is this getting a reference to myself?
my $apr=Apache::Request->new($r);  ## doesn't seem to clash with other 
				   ## calls to getCurrentForm()
my $upload=$apr->upload;           ## works great.
...
}

Sorry about the noise,
james

Re: [Slashcode-development] RFC: SQL Placeholders

[Stephen C. <the...@wa...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sat, Apr 07, 2001 at 08:11:03PM -0700, Eric Dannewitz wrote:
> Why not use VARCHAR2 in Oracle? From what I remember of Oracle, that is a much
> better variable than VARCHAR

Currently VARCHAR == VARCHAR2 in Oracle.  Makes you wonder why they have two
different names, but then there are lots of things you tend to wonder about
Oracle.  Officially the docs say they may develop differences in a future 
version, but they also said that 8i would finally break the infuriatingly 
annoying '' == NULL behavior (which is also mucking up its share of things in
my porting attempts).  I'm not looking for either change to happen soon.

In any case, the maximum length of VARCHAR* is 4000 bytes; beyond that you must 
use *LOB.  The templates are well beyond that limit, and inevitably some user 
comments will be as well, so the use of LOB is unavoidable here.  My original 
problem remains.

- -- 
Stephen Clouse <the...@wa...>
warpcore.org Founder, Chief Megalomaniac, and Evil Overlord

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8

iQA/AwUBOs/kvt1EXk7JbKbMEQINbwCgjITyQNrtg1oQEQTKFoWXMX1ZZhsAoPxp
y6U7l/03EO7Zv8yLkoX96ght
=JIu4
-----END PGP SIGNATURE-----

Re: [Slashcode-development] RFC: SQL Placeholders

[Eric D. <eri...@ja...>]

Why not use VARCHAR2 in Oracle? From what I remember of Oracle, that is a much
better variable than VARCHAR

Stephen Clouse wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> As of late I have been working on getting Slash2 to work with Oracle (going
> ahead on my own since there wasn't much interest (read: none) a while back when
> I asked).  So far it's been quite successful -- the initial database creates and
> loads without error.  Now it's down to messing with the DB interface
> itself.  The biggest show-stopper so far is while loading the templates in:
>
> DBD::Oracle::db do failed: ORA-01704: string literal too long
>
> because it's trying to embed the entire string into the SQL statement, which you
> can't do in Oracle if it's over 4000 bytes (the boundary between VARCHAR and
> LOB).  The variable has to be explicitly typed as a LOB when preparing the
> statement, and that requires the use of DBI placeholders for bind
> values.  However, there isn't a single placeholder in the current code.  I found
> this rather surprising.
>
> Aside from the obvious technical issue above, there's a performance one as
> well.  Oracle isn't able to cache any of the SQL with the variables embedded
> inside, so performance will suffer when I finally do manage to get it
> working.  This is actually an issue with MySQL as well, since Slash2 currently
> makes quite a bit of use of prepare_cached, a use that's practically worthless
> (and a large waste of memory) without using placeholders.
>
> My proposal is to get the SQL internals using placeholders and bind values to
> facilitate current and future porting, as well as making the current MySQL core
> substantially cleaner and more memory-efficient.  I've spent all day looking
> over the Slash::DB::Utility interface and the sql* methods and I've devised an
> extension to the current interface that would make the changes to the current
> code minimal (an additional argument to sql* calls with embedded variables, and
> some fairly minor recoding where such calls are).  The conversion would almost
> be painless.
>
> Even so, this is a pretty radical change to the code, and I wouldn't go ahead
> with it without some kind of "blessing" from the core developers and ensuring
> that this is a direction they want to head.  However, my main goal is seeing
> this thing run on Oracle, and I don't see any way of getting it there without
> A) these changes to allow Oracle more seamless plugging into the existing code,
> or B) a substantial rewrite Slash::DB::* for Oracle, which would almost be a
> code fork since the interface would have to be radically different.  That's the
> last place I want to head.
>
> There are other porting issues I've come across, but they're pathetic compared
> to this.  Comments are welcome.
>
> - --
> Stephen Clouse <the...@wa...>
> warpcore.org Founder, Chief Megalomaniac, and Evil Overlord
>
> -----BEGIN PGP SIGNATURE-----
> Version: PGP 6.5.8
>
> iQA+AwUBOs/Lt91EXk7JbKbMEQKX8gCYsdgaJOPlKxDvE+XB3g7EBLielACeLYzY
> vleJuQHrdCaBJn4gtYttfow=
> =y90I
> -----END PGP SIGNATURE-----
>
> _______________________________________________
> Slashcode-development mailing list
> Sla...@li...
> http://lists.sourceforge.net/lists/listinfo/slashcode-development

--
Back up my hard disk? I can't find the reverse switch!

Eric Dannewitz - Adventurer, saxophonist, good-timer (crook? quite possibly),
clarinetist, manic self-publicist, part-time flautist(flutist?), macintosher, and
often thought to be completely out to lunch. http://www.jazz-sax.com