apc-aa-coders — Discussion only useful to active developers.

[Apc-aa-coders] PHPdoc like comments

[Jakub A. <jak...@se...>]

Mitra,
 
your code in include/zids.php3 is nice formatted and commented. I only
do not understand why you did not use the PHPdoc formatting described in
this list? Are you going to use it in the future or do you have some
reason against?
 
And I didn't find a remark about the author (which I assume is just you)
of the file.
 
Jakub

RE: [Apc-aa-coders] Abandoning aadb.sql

[Marek T. <ma...@gn...>]

> But ... lets not lose site of the original question - can we delete 
> aadb.sql from the system and remove it from the documentation!   We 
> don't need the wizzard before we do that.

I'd say "YES"

YES. YES YES YES !!!!!!!


Yes 	   Yes    
 Yes    Yes     
  Yes  Yes      
   YesYes 
    Yes
     Yes 
      Yes
       Yes 

Is there anything more I could do ? 

[Apc-aa-coders] change in views conditons?

[Norbert B. <br...@ch...>]

Mitra,=20

I am testing current AA code and it seems that there were some changes =
in view conditions introduced (perhaps related to zids?).

Here is link to a view which works (it just displays _#HEADLINE field =
from its slice)
http://www.changenet.sk/aa/view.php3?vid=3D302&cmd[302]=3Dx-302-44786&noc=
ache=3D1

This view used to display nothing when I passed no short item id to it:
http://www.changenet.sk/aa/view.php3?vid=3D302&cmd[302]=3Dx-302-&nocache=3D=
1

but now (with fresh code from CVS) I am getting error message:

Database error: Invalid SQL: SELECT * FROM item WHERE short_id IN ()
MySQL Error: 1064 (You have an error in your SQL syntax near ')' at line =
1)
Please contact te...@ch... and report the exact error message.
Session halted.



norbert
br...@ch...

Re: [Apc-aa-coders] Ids of strange lineths.

[Mitra <mi...@mi...>]

Well spotted!

I get similar results (and those two "huh" should be "huhl", 've got


Array
(
     [16] =3D> 3652
     [15] =3D> 15
)
packed

Array
(
     [32] =3D> 3652
     [30] =3D> 15
)
unpacked

- Mitra


At 2:43 PM +0100 25/1/03, Honza Malik wrote:
>Hi Mitra,
>
>    I did some more testing on ids and now I know, where the problem is.
>The new_id() is OK - it returns allways 32 character long id, but
>sometimes it return id with ...20 at the end. Packed '20' is space and
>this is problem, because if you insert 16 characters long packed id
>ending with space into MySQL and then you get it back, the id is 15
>character long.
>
>MySQL REMOVES ENDING SPACES FROM INSERTED STRING!
>
>I will fix new_id() to not generate such ids.
>
>
>On P=E1, 2003-01-24 at 12:28, Mitra wrote:
>>  Ok - stranger .... my guess was wrong, md5 returns a 32 character
>>  string, which cannot contain 00 or 27 so cannot end up as a shorter
>>  string.
>>
>>  To test, I ran a loop through 99999 samples of new_id, which should
>>  have generated about 400 bad ids by your statistics.
>>
>>  I think the bad ids in your database must be the result of some buggy
>>  earlier code.
>>
>>  Lets try three things.
>>  1: Please check my test - appended at the end of this message.
>
>I test it and all 99999 ids are 32 characters long.
>
>>  2: Please re-run your test, see if the number of bad ids has gone up,
>>  if there is something bad then as long as you have created more than
>>  256 ids then statistically it should have increased by 1 or more.
>
>Increased: [16] =3D> 92254 [15] =3D> 367 [14] =3D> 4
>
>>  3: Please send the code for your test, I'll run it on my installation
>>  and see how many odd length ids I have.
>
>require "include/config.php3";
>require $GLOBALS[AA_INC_PATH]."util.php3";
>require $GLOBALS[AA_INC_PATH]."locsessi.php3";
>
>$db =3D new DB_AA;
>
>$db->query( "SELECT id from item" );
>while($db->next_record()) {
>     $fromdb[$len_p =3D strlen($db->f('id'))]++;
>     $unpdb[$len_u =3D strlen(unpack_id($db->f('id')))]++;
>//    if( $len_p<16 )
>//        echo "<br>".$db->f('id');
>}   
>
>huh($fromdb, 'packed');
>huh($unpdb, 'unpacked');
>
>echo "
>
>        Regards
>                  Honza ";
>
>
>  >
>  > - Mitra
>  >
>  >
>  > At 11:55 PM +0100 22/1/03, Honza Malik wrote:
>  > >  >3) Your assumption that packed ids are EXACTLY 16 characters long (=
and
>>  >  >unpacked are EXACTLY 32 characters long) is unfortunately wrong. I c=
an't
>>  >  >say you why - we generate the ids by new_id() function, but in our
>>  >  >database are packed ids for item 16 (in 91907 items), 15 (366) and 1=
4
>>  >  >(4) characters long.
>>  >
>>  >  Strange - because new_id uses "md5" which has to return 128 bits.
>>  >
>>  >  I wonder if these were ids which generated a leading ascii zero, the
>>  >  ratios look reasonable for this, i.e. 919007/366 =3D 251, which is
>>  >  close to a one in 256 chance, i.e. something that occurred when 8
>>  >  bits (2^8=3D256) all were zero.    This makes sense if we assume that
>>  >  md5 returns a number which the ereg in new_id casts as a string, and
>>  >  so ignores the leading zero.
>>  >
>>  >  I've fixed zids.php3 to allow 12..16 character packed ids, and 24..32
>>  >  long, but this could break my tagged ids. I've also fixed
>>  >  util.php3.new_id to only return a string of exactly 16 chars.
>>  >
>>  >  > >5) In zids.php3 you wrote:
>>  >>  >
>>  >>  >function id_t2l(&$str) {
>>  >>  >#TODO: Look online for quicker way to substr last 32 chars - mitra
>>  >>  >         if (ereg("(.*)([0-9a-f]{32})",$str,$parts))
>>  >>  >                 return $parts[2];
>>  >>  >
>>  >>  >The quicker way is: substr($str, -32), but see 3)
>>  >>
>>  >>  Ok - I'll make this 24..32, and I guess we can't use tags that end i=
n
>>  >>  a hex character [0-9a-f)
>>  >
>>  >Or you can use some delimeter (tag:unpacked_id). As I write before, we
>>  >never use tags so far, so the syntax is on your choice.
>>
>>
>>  require "../include/config.php3";
>>  require $GLOBALS[AA_INC_PATH]."util.php3";
>>
>>  $i =3D 0;
>>  $j=3D 99999;
>>  print "<br>Testing start, loops=3D$j: sample=3D";
>  > $foo =3D new_id();
>>  print $foo;
>>  while ($i++ < $j) {
>>  $foo =3D new_id();
>>  if (strlen($foo) !=3D 32) print "<br>FOO:$i:$foo:strlen=3D".strlen($foo)=
;
>>  if (strlen(pack_id($foo)) !=3D 16) print "<br>FOO:$i:$foo:packed
>>  strlen=3D".strlen(\
>>  pack_id($foo));
>>  }
>>  print "<br>Testing end sample =3D ";
>>  $foo =3D new_id();
>>  print $foo;
>>
>>  print "Done\n";
>>
>
>
>
>-------------------------------------------------------
>This SF.NET email is sponsored by:
>SourceForge Enterprise Edition + IBM + LinuxWorld http://www.vasoftware.com
>_______________________________________________
>Apc-aa-coders mailing list
>Apc...@li...
>https://lists.sourceforge.net/lists/listinfo/apc-aa-coders


-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved

Re: [Apc-aa-coders] Ids of strange lineths.

[Honza M. <hon...@ec...>]

Hi Mitra,

   I did some more testing on ids and now I know, where the problem is.
The new_id() is OK - it returns allways 32 character long id, but
sometimes it return id with ...20 at the end. Packed '20' is space and
this is problem, because if you insert 16 characters long packed id
ending with space into MySQL and then you get it back, the id is 15
character long.=20

MySQL REMOVES ENDING SPACES FROM INSERTED STRING!

I will fix new_id() to not generate such ids.


On P=E1, 2003-01-24 at 12:28, Mitra wrote:
> Ok - stranger .... my guess was wrong, md5 returns a 32 character=20
> string, which cannot contain 00 or 27 so cannot end up as a shorter=20
> string.
>=20
> To test, I ran a loop through 99999 samples of new_id, which should=20
> have generated about 400 bad ids by your statistics.
>=20
> I think the bad ids in your database must be the result of some buggy=20
> earlier code.
>=20
> Lets try three things.
> 1: Please check my test - appended at the end of this message.

I test it and all 99999 ids are 32 characters long.

> 2: Please re-run your test, see if the number of bad ids has gone up,=20
> if there is something bad then as long as you have created more than=20
> 256 ids then statistically it should have increased by 1 or more.

Increased: [16] =3D> 92254 [15] =3D> 367 [14] =3D> 4=20

> 3: Please send the code for your test, I'll run it on my installation=20
> and see how many odd length ids I have.

require "include/config.php3";
require $GLOBALS[AA_INC_PATH]."util.php3";
require $GLOBALS[AA_INC_PATH]."locsessi.php3";=20

$db =3D new DB_AA;

$db->query( "SELECT id from item" );
while($db->next_record()) {
    $fromdb[$len_p =3D strlen($db->f('id'))]++;
    $unpdb[$len_u =3D strlen(unpack_id($db->f('id')))]++;
//    if( $len_p<16 )
//        echo "<br>".$db->f('id');
}   =20

huh($fromdb, 'packed');
huh($unpdb, 'unpacked');

echo "

       Regards
                 Honza ";


>=20
> - Mitra
>=20
>=20
> At 11:55 PM +0100 22/1/03, Honza Malik wrote:
> >  >3) Your assumption that packed ids are EXACTLY 16 characters long (=
and
> >  >unpacked are EXACTLY 32 characters long) is unfortunately wrong. I =
can't
> >  >say you why - we generate the ids by new_id() function, but in our
> >  >database are packed ids for item 16 (in 91907 items), 15 (366) and =
14
> >  >(4) characters long.
> >
> >  Strange - because new_id uses "md5" which has to return 128 bits.
> >
> >  I wonder if these were ids which generated a leading ascii zero, the
> >  ratios look reasonable for this, i.e. 919007/366 =3D 251, which is
> >  close to a one in 256 chance, i.e. something that occurred when 8
> >  bits (2^8=3D256) all were zero.    This makes sense if we assume tha=
t
> >  md5 returns a number which the ereg in new_id casts as a string, and
> >  so ignores the leading zero.
> >
> >  I've fixed zids.php3 to allow 12..16 character packed ids, and 24..3=
2
> >  long, but this could break my tagged ids. I've also fixed
> >  util.php3.new_id to only return a string of exactly 16 chars.
> >
> >  > >5) In zids.php3 you wrote:
> >>  >
> >>  >function id_t2l(&$str) {
> >>  >#TODO: Look online for quicker way to substr last 32 chars - mitra
> >>  >         if (ereg("(.*)([0-9a-f]{32})",$str,$parts))
> >>  >                 return $parts[2];
> >>  >
> >>  >The quicker way is: substr($str, -32), but see 3)
> >>
> >>  Ok - I'll make this 24..32, and I guess we can't use tags that end =
in
> >>  a hex character [0-9a-f)
> >
> >Or you can use some delimeter (tag:unpacked_id). As I write before, we
> >never use tags so far, so the syntax is on your choice.
>=20
>=20
> require "../include/config.php3";
> require $GLOBALS[AA_INC_PATH]."util.php3";
>=20
> $i =3D 0;
> $j=3D 99999;
> print "<br>Testing start, loops=3D$j: sample=3D";
> $foo =3D new_id();
> print $foo;
> while ($i++ < $j) {
> $foo =3D new_id();
> if (strlen($foo) !=3D 32) print "<br>FOO:$i:$foo:strlen=3D".strlen($foo=
);
> if (strlen(pack_id($foo)) !=3D 16) print "<br>FOO:$i:$foo:packed=20
> strlen=3D".strlen(\
> pack_id($foo));
> }
> print "<br>Testing end sample =3D ";
> $foo =3D new_id();
> print $foo;
>=20
> print "Done\n";
>=20

[Apc-aa-coders] m_gettext conversion incomplete! jakub - are you still working on this?

[Mitra <mi...@mi...>]

it looks like the m_gettext conversion is incomplete, I keep hitting 
places where L_ constants aren't defined, and a quick grep finds 156 
occurances

Of these  76 in oldDb2Db which I presume are intentionally not 
converted  - some are even in places like admin/itemedit.php3.

What do we need to do about this?

For now, I've restored the en_site_lang.php3 file, since its catching 
me in the site module. If other people are seeing L_ strings 
appearing all over the place they might want to do the same thing!

I've also put en_news_lang.php3 back into the CVS distribution so 
that at least we have a reference to what the other strings are 
supposed to be, I'll probably put it back into config.php3 on my 
sites as well.

Jakub - are you still working on fixing these,

- Mitra

-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722

Life is a Mystery to be Lived, not a Problem to be Solved

Re: [Apc-aa-coders] Abandoning aadb.sql

[Mitra <mi...@mi...>]

This sounds like a good idea.

Since in practice the admin script probably won't be deleted / moved, 
I would suggest that the script generates config_new.php3, and that 
this has to be renamed as config.php3 manually.

But ... lets not lose site of the original question - can we delete 
aadb.sql from the system and remove it from the documentation!   We 
don't need the wizzard before we do that.

- Mitra



At 9:59 PM +0530 24/1/03, Ram Prasad wrote:
>Few applications like movabletype, have a wizard to do the initial
>configuration. After which the sysadmin is required to rename/delete the
>wizard file.
>
>We could have a setup_init.php3, for example, which will create the initial
>config.php3 ( it could ask for the DB parameters .. etc). After completing
>( or after a test of connectivity to the DB), the script could be deleted or
>moved.
>
>- Ram
>----- Original Message -----
>From: "Jakub Adamek" <jak...@se...>
>To: "'Ram Prasad'" <ra...@pa...>; "'Mitra'" <mi...@mi...>;
><apc...@so...>
>Sent: Friday, January 24, 2003 9:52 PM
>Subject: RE: [Apc-aa-coders] Abandoning aadb.sql
>
>
>>  Perhaps this approach may work: Let some script generate code into a
>>  text area, which the superadmin copies manually into config.php3.
>>
>>  Jakub
>>
>>  > -----Original Message-----
>>  > From: apc...@li...
>>  > [mailto:apc...@li...] On Behalf
>>  > Of Ram Prasad
>>  > Sent: Thursday, January 23, 2003 8:00 PM
>>  > To: 'Mitra'; apc...@so...
>>  > Subject: RE: [Apc-aa-coders] Abandoning aadb.sql
>>  >
>>  >
>>  > Also, it would be nice if we have a page to set up the initial
>>  > configuration. This way, the config.php3 can be set using the web
>>  > interface, which can then be used to configure apc-aa.
>>  >
>>  > - Ram
>>  >
>>  > -----Original Message-----
>>  > From: apc...@li...
>>  > [mailto:apc...@li...] On Behalf Of Mitra
>>  > Sent: Thursday, January 23, 2003 1:24 PM
>>  > To: apc...@so...
>>  > Subject: [Apc-aa-coders] Abandoning aadb.sql
>>  >
>>  >
>>  > I'm just installing APC-AA on a machine where I do not have any kind
>>  > of shell access. This is going to be the case for many potential
>>  > users of apc-aa.
>>  >
>>  > So of course, I could not follow the directions, which depend on shell
>>  > access.
>  > >
>  > > Typically all of doc/install-3.html can be skipped, since any
>  > > competent ISP will have a reasonably recent version of Apache, PHP
>  > > and MySQL installed.
>  > >
>  > > My first install attempt failed since the sysadmin hadn't given ALTER
>>  > or DROP permissions to MYSQL, but the second attempt worked fine -
>>  > using sql_update.php3 and then setup.php3 to get it all going.
>>  >
>>  > Here are some suggestions
>>  > 1: Create some test scripts, so that a user can tell if the system is
>>  > set up correctly, and if not knows what to ask for, I've started with
>>  > admin/test.htm and admin/test.php which test PHP installation, and
>>  > Server Side Includes, but it would be great to have something which
>>  > excercised other common functions and maybe checked PHP options etc?
>>  >
>>  > 2: Stop any dependency on aadb.sql, I think that all it does is
>>  > replicate what sql_update.php3 does anyway, so why have the
>>  > functionality in two places?   Maybe just have setup.php3 run
>>  > sql_update if the tables don't exist?
>>  >
>>  > - Mitra
>>  >
>>  >
>>  >
>>  > --
>>  > Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
>>  > 02-6684-8096  or 0414-648-0722
>>  >
>>  > Life is a Mystery to be Lived, not a Problem to be Solved
>>  >
>>  >
>>  >
>>  > -------------------------------------------------------
>>  > This SF.net email is sponsored by: Scholarships for Techies! Can't
>>  > afford IT training? All 2003 ictp students receive scholarships. Get
>>  > hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more.
>>  > www.ictp.com/training/sourceforge.asp
>>  > _______________________________________________
>>  > Apc-aa-coders mailing list
>  > > Apc...@li...
>>  > https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
>>  >
>>  >
>>  >
>>  >
>>  > -------------------------------------------------------
>>  > This SF.NET email is sponsored by:
>>  > SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
>>  > http://www.vasoftware.com
>>  > _______________________________________________
>>  > Apc-aa-coders mailing list
>>  > Apc...@li...
>>  > https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
>>  >
>>
>>
>>
>>  -------------------------------------------------------
>>  This SF.NET email is sponsored by:
>>  SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
>>  http://www.vasoftware.com
>>  _______________________________________________
>>  Apc-aa-coders mailing list
>>  Apc...@li...
>>  https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
>>
>>
>
>
>
>-------------------------------------------------------
>This SF.NET email is sponsored by:
>SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
>http://www.vasoftware.com
>_______________________________________________
>Apc-aa-coders mailing list
>Apc...@li...
>https://lists.sourceforge.net/lists/listinfo/apc-aa-coders


-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved

Re: [Apc-aa-coders] GD library and image widths

[Mitra <mi...@mi...>]

At 4:00 PM +0100 24/1/03, Norbert Brazda wrote:
>Hi Mitra,
>
>From: "Mitra" <mi...@mi...>
>>  Your image upload library looks really useful, timely as I need
>>  something like ti for a project I'm working on for a community radio
>>  station.
>
>We plan to start internet radio (using MUSE and IceCAST) so if you 
>have any recommendations regarding www interface for listeners or 
>code to share, I will be happy to hear.
>
>
>>  The question .... are GIF's supported - your code recognizes them as
>>  a supported type but your email said they weren't supported?
>
>GD library does not support GIF due to some licencing problems. Thus 
>GIF images could be uploaded into AA but there is no way how to 
>resize them, etc.

OK - that is fine, but it appeared that there was a bug that 
imagefunc.GetSupportedTypes was pretending it supported GIF's - I was 
mistaken, the code looks fine.

>  > Problem 1: Are there simple instructions we could give people (in the
>>  FAQ and installation docs) for installing GD, since it seems that it
>>  is not present on most PHP installations.
>
>It was a bit tricky to install on our server, so we do not have any 
>general FAQ. But, homepage of GD library is at 
>http://www.boutell.com/gd/ where GD lib authors wrote:
>
>Updated PHP Information, 12/27/02: php 4.3.0 is available, and it 
>includes a version of gd as "standard equipment." I have only taken 
>a brief preliminary look at it, but it seems to include mainstream 
>gd fixes up through version 2.0.2, and also includes an alternative 
>set of image rotation functions. If you wish to use gd with PHP, it 
>is probably best to get php 4.3.0 and use the included gd library. 
>However, if you wish to continue using php 4.2.3 but need gd support 
>for it, see the patching PHP for gd 2.0.8 and up page for important 
>information. The php maintainers included a statement saying that 
>their improvements would be contributed back to the "mainstream" gd, 
>so I look forward to doing so in the near future. I will also look 
>into preparing a patch for PHP that allows the use of the resulting 
>release of gd.
>
>Here is link to the "patching PHP for gd 2.0.8 and up":
>http://www.boutell.com/gd/phppatch.html

See earlier email ...

>  > Problem 2:
>>  .... lets take a typical example .
>>
>>  I'll setup file upload to have parameters    image/*:200:100
>>
>>  So lets say the user uploads an image 400*100, it will get resampled
>>  to 200*50 which is great.
>>
>>  But ... how do we generate the output HTML , what I'd like to output is
>>  <IMG SRC="/apc-aa-uploads/12345/foo.jpg" HEIGHT=50 WIDTH=200>
>>  but I have no way to know that the height and width actually are 50 and 200.
>
>You are right. Width and height is not supported by Stano`s img 
>function. Do you really need them?


Jakub said
>The function is standard in PHP and does not even need GD. It is 
>getimagesize().


Yes - image support, for uploaded images, has been really hard in 
APC, Stano's function gets us really close to having easy to use 
image support.  Any good HTML code should set HEIGHT and WIDTH fields 
so that the HTML can render while waiting for the images to download, 
so we need an easy way for system administrators to handle and work 
with images that are uploaded at runtime and so they don't know the 
WIDTH and HEIGHT of.  Jakub has pointed me at the getimagesize() 
standard PHP function, I can put that into imagefunc, although I was 
leaving that code to Stano in case he was still working on it.

>
>
>  > I can think of several ways around this.
>  > 1: Have the upload function stuff the resampled height in the
>  > img_width..... and img_height..... fields
>  > this is not so clever as there might be multiple images on a form.
>>
>>  2: Provide an output expansion that gets what is required, I think
>>  the ideal might be
>>  {img:width:apc-aa-uploads/12345/foo.jpg} and
>>  {img:height:apc-aa-uploads/12345/foo.jpg}
>>
>>  So that the output expression becomes simpyl
>>  <IMG SRC="{img_url.........}" HEIGHT={img:height:{img_url.........}}
>>  WIDTH={img:width:{img_url.........}}>
>  > This would give maximum flexibility to output anything extra without
>>  requiring lots of switches and math in each output?
>>
>>  What do you think - could you provide a couple of functions that take
>>  a file name and return the width and height ?  if so then I'll code
>>  them into the string expansion code.
>
>It would be much simpler to manipulate images on output size (when 
>generating html code for the client), but we wanted to avoid 
>unnecessary load on server. This is why we decided to resize images 
>on input (upload).

I think its great you are resizing on upload. BUT to keep aspect 
ratio you can only know for sure ONE of the dimensions - e.g. 
resizing a 400*200 image into a 200*200 box ends up with a 200*100 
image and you have to know this when you output the <IMG SRC=** 
HEIGHT=** WIDTH=**>. So we need a way INSIDE THE VIEW to find this 
height and width.


>
>Caching in AA is really good, so perhaps there is no need to 
>calculate image dimension on upload side. What others think about 
>this?

Yes - I agree, that we can calculate it when we do the output - 
although we DO have to worry about speed in AA, if you run AA with 
standard 10 minute caching, and some part of your site is accessed 
less often than this, then the speed of AA is EXTREMELY slow, I had 
to set caching to 2 days, after one client complained that their site 
ran so slowly.

- Mitra



-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved

RE: [Apc-aa-coders] GD library and image widths

[Mitra <mi...@mi...>]

Thanks

It would be helpful to have instructions for upgrading earlier 
versions that support GD, our current install instructions say we 
support even PHP 3.*, we were going to change it to say it supports 
4.0.6 and above (to get array_map) but I guess we have to put in the 
instructions that APC-AA is only fully supported on 4.3.0 and above.

For example, I'm installing apc-aa for a community radio station, and 
their ISP is running PHP4.2.3 -
http://bayfm.org/apc-aa/admin/test.php3

- Mitra



At 2:20 PM +0100 24/1/03, Marek Tichy wrote:
>  > The question .... are GIF's supported - your code recognizes them as
>>  a supported type but your email said they weren't supported?
>No longer, and GD authors are quite strict:
>
>"PLEASE DON'T ASK FOR GIF-BASED VERSIONS OF GD. HARANGUES ABOUT GIF 
>WILL BE DELETED UNREAD. Thank you! "
>http://www.boutell.com/gd/
>
>>  Problem 1: Are there simple instructions we could give people (in the
>>  FAQ and installation docs) for installing GD, since it seems that it
>>  is not present on most PHP installations.
>
>php 4.3.0 is available, and it includes a version of gd as "standard 
>equipment."
>
>Best
>  Marek
>
>
>-------------------------------------------------------
>This SF.NET email is sponsored by:
>SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
>http://www.vasoftware.com
>_______________________________________________
>Apc-aa-coders mailing list
>Apc...@li...
>https://lists.sourceforge.net/lists/listinfo/apc-aa-coders


-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved

Re: [Apc-aa-coders] Abandoning aadb.sql

[Ram P. <ra...@pa...>]

Few applications like movabletype, have a wizard to do the initial
configuration. After which the sysadmin is required to rename/delete the
wizard file.

We could have a setup_init.php3, for example, which will create the initial
config.php3 ( it could ask for the DB parameters .. etc). After completing
( or after a test of connectivity to the DB), the script could be deleted or
moved.

- Ram
----- Original Message -----
From: "Jakub Adamek" <jak...@se...>
To: "'Ram Prasad'" <ra...@pa...>; "'Mitra'" <mi...@mi...>;
<apc...@so...>
Sent: Friday, January 24, 2003 9:52 PM
Subject: RE: [Apc-aa-coders] Abandoning aadb.sql


> Perhaps this approach may work: Let some script generate code into a
> text area, which the superadmin copies manually into config.php3.
>
> Jakub
>
> > -----Original Message-----
> > From: apc...@li...
> > [mailto:apc...@li...] On Behalf
> > Of Ram Prasad
> > Sent: Thursday, January 23, 2003 8:00 PM
> > To: 'Mitra'; apc...@so...
> > Subject: RE: [Apc-aa-coders] Abandoning aadb.sql
> >
> >
> > Also, it would be nice if we have a page to set up the initial
> > configuration. This way, the config.php3 can be set using the web
> > interface, which can then be used to configure apc-aa.
> >
> > - Ram
> >
> > -----Original Message-----
> > From: apc...@li...
> > [mailto:apc...@li...] On Behalf Of Mitra
> > Sent: Thursday, January 23, 2003 1:24 PM
> > To: apc...@so...
> > Subject: [Apc-aa-coders] Abandoning aadb.sql
> >
> >
> > I'm just installing APC-AA on a machine where I do not have any kind
> > of shell access. This is going to be the case for many potential
> > users of apc-aa.
> >
> > So of course, I could not follow the directions, which depend on shell
> > access.
> >
> > Typically all of doc/install-3.html can be skipped, since any
> > competent ISP will have a reasonably recent version of Apache, PHP
> > and MySQL installed.
> >
> > My first install attempt failed since the sysadmin hadn't given ALTER
> > or DROP permissions to MYSQL, but the second attempt worked fine -
> > using sql_update.php3 and then setup.php3 to get it all going.
> >
> > Here are some suggestions
> > 1: Create some test scripts, so that a user can tell if the system is
> > set up correctly, and if not knows what to ask for, I've started with
> > admin/test.htm and admin/test.php which test PHP installation, and
> > Server Side Includes, but it would be great to have something which
> > excercised other common functions and maybe checked PHP options etc?
> >
> > 2: Stop any dependency on aadb.sql, I think that all it does is
> > replicate what sql_update.php3 does anyway, so why have the
> > functionality in two places?   Maybe just have setup.php3 run
> > sql_update if the tables don't exist?
> >
> > - Mitra
> >
> >
> >
> > --
> > Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
> > 02-6684-8096  or 0414-648-0722
> >
> > Life is a Mystery to be Lived, not a Problem to be Solved
> >
> >
> >
> > -------------------------------------------------------
> > This SF.net email is sponsored by: Scholarships for Techies! Can't
> > afford IT training? All 2003 ictp students receive scholarships. Get
> > hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more.
> > www.ictp.com/training/sourceforge.asp
> > _______________________________________________
> > Apc-aa-coders mailing list
> > Apc...@li...
> > https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
> >
> >
> >
> >
> > -------------------------------------------------------
> > This SF.NET email is sponsored by:
> > SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> > http://www.vasoftware.com
> > _______________________________________________
> > Apc-aa-coders mailing list
> > Apc...@li...
> > https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
> >
>
>
>
> -------------------------------------------------------
> This SF.NET email is sponsored by:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
> _______________________________________________
> Apc-aa-coders mailing list
> Apc...@li...
> https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
>
>

RE: [Apc-aa-coders] Abandoning aadb.sql

[<jak...@se...>]

Perhaps this approach may work: Let some script generate code into a
text area, which the superadmin copies manually into config.php3.

Jakub

> -----Original Message-----
> From: apc...@li... 
> [mailto:apc...@li...] On Behalf 
> Of Ram Prasad
> Sent: Thursday, January 23, 2003 8:00 PM
> To: 'Mitra'; apc...@so...
> Subject: RE: [Apc-aa-coders] Abandoning aadb.sql
> 
> 
> Also, it would be nice if we have a page to set up the initial
> configuration. This way, the config.php3 can be set using the web
> interface, which can then be used to configure apc-aa.
> 
> - Ram
> 
> -----Original Message-----
> From: apc...@li...
> [mailto:apc...@li...] On Behalf Of Mitra
> Sent: Thursday, January 23, 2003 1:24 PM
> To: apc...@so...
> Subject: [Apc-aa-coders] Abandoning aadb.sql
> 
> 
> I'm just installing APC-AA on a machine where I do not have any kind 
> of shell access. This is going to be the case for many potential 
> users of apc-aa.
> 
> So of course, I could not follow the directions, which depend on shell
> access.
> 
> Typically all of doc/install-3.html can be skipped, since any 
> competent ISP will have a reasonably recent version of Apache, PHP 
> and MySQL installed.
> 
> My first install attempt failed since the sysadmin hadn't given ALTER 
> or DROP permissions to MYSQL, but the second attempt worked fine - 
> using sql_update.php3 and then setup.php3 to get it all going.
> 
> Here are some suggestions
> 1: Create some test scripts, so that a user can tell if the system is 
> set up correctly, and if not knows what to ask for, I've started with 
> admin/test.htm and admin/test.php which test PHP installation, and 
> Server Side Includes, but it would be great to have something which 
> excercised other common functions and maybe checked PHP options etc?
> 
> 2: Stop any dependency on aadb.sql, I think that all it does is 
> replicate what sql_update.php3 does anyway, so why have the 
> functionality in two places?   Maybe just have setup.php3 run 
> sql_update if the tables don't exist?
> 
> - Mitra
> 
> 
> 
> -- 
> Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
> 02-6684-8096  or 0414-648-0722
> 
> Life is a Mystery to be Lived, not a Problem to be Solved
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Scholarships for Techies! Can't
> afford IT training? All 2003 ictp students receive scholarships. Get
> hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more.
> www.ictp.com/training/sourceforge.asp
> _______________________________________________
> Apc-aa-coders mailing list
> Apc...@li...
> https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.NET email is sponsored by:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
> _______________________________________________
> Apc-aa-coders mailing list
> Apc...@li...
> https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
> 

[Apc-aa-coders] Minor changes

[<jak...@se...>]

I comitted a few minor changes into CVS, 

01/24/01 - bug fix in slicedit.php3, mistyped URL sliceEdit.php3 (Jakub)
01/24/01 - changed design of se_fields.php3 (Jakub)
01/24/01 - added H2 to admin.css (Jakub)

Jakub

RE: [Apc-aa-coders] GD library and image widths

[Marek T. <ma...@gn...>]

> The question .... are GIF's supported - your code recognizes them as 
> a supported type but your email said they weren't supported?
No longer, and GD authors are quite strict:

"PLEASE DON'T ASK FOR GIF-BASED VERSIONS OF GD. HARANGUES ABOUT GIF WILL BE DELETED UNREAD. Thank you! "
http://www.boutell.com/gd/
 
> Problem 1: Are there simple instructions we could give people (in the 
> FAQ and installation docs) for installing GD, since it seems that it 
> is not present on most PHP installations.

php 4.3.0 is available, and it includes a version of gd as "standard equipment." 

Best 
 Marek 

RE: [Apc-aa-coders] Abandoning aadb.sql

[Mitra <mi...@mi...>]

I'm happy to rewrite the installation instructions - in fact, I 
already did this to give the alternative way of doing it when you 
don't have shell access (such as with my current project).

What I'm hoping for is a consensus to drop aadb.sql so we no longer 
document it and have to keep supporting it.

- Mitra


At 12:44 PM +0100 24/1/03, Marek Tichy wrote:
>Hi Mitra,
>yes, yes, you are indeed right. The aadb.sql file is there for 
>historical reasons, this file has been there ever since, 
>while sql_update.php3 was introduced with 1.8 ? (guessing - doesn't 
>really matter). I think you are just the first one who has noticed 
>that you can use sql update as well, even if you don't have any 
>database structure yet. Don't be ashamed of it - great spotting !
>I can rewrite the installation instructions if you want me to do do.
>Best
>  Marek
>
>-----Original Message-----
>From: apc...@li... 
>[mailto:apc...@li...]On Behalf Of Mitra
>Sent: 23 January 2003 22:43
>To: Honza Malik; apc...@so...
>Subject: RE: [Apc-aa-coders] Abandoning aadb.sql
>
>I agree that we don't want write access to the scripts.
>
>Honza - I'm probably missunderstanding something - what is the point 
>of aadb.sql ? What function does aadb.sql provide that 
>sql_update.php3 doesn't?  What advantages are there of configuring 
>apc-aa with aadb.sql rather than via sql_update.php3?
>
>- Mitra
>
>
>At 9:52 PM +0100 23/1/03, Honza Malik wrote:
>
>>I agree with droping aadb.sql, mainly if we can quite easily write an
>>option for sql_update.php3, which will generate aadb.sql.
>>
>>I'm not sure about configuration using web interface. It would require
>>write access to config.php3. I think we should never give webserver the
>>rights to write to any AA script.
>>
>>    Honza
>>
>>
>>On =89t, 2003-01-23 at 19:59, Ram Prasad wrote:
>>>  Also, it would be nice if we have a page to set up the initial
>>>  configuration. This way, the config.php3 can be set using the web
>>>  interface, which can then be used to configure apc-aa.
>>>
>>>  - Ram
>>>
>>>  -----Original Message-----
>>>  From: apc...@li...
>>>  [mailto:apc...@li...] On Behalf Of Mitra
>>>  Sent: Thursday, January 23, 2003 1:24 PM
>  > To: apc...@so...
>  > Subject: [Apc-aa-coders] Abandoning aadb.sql
>  >
>  >
>  > I'm just installing APC-AA on a machine where I do not have any kind
>  > of shell access. This is going to be the case for many potential
>>  users of apc-aa.
>>
>>  So of course, I could not follow the directions, which depend on shell
>>  access.
>>
>>  Typically all of doc/install-3.html can be skipped, since any
>>  competent ISP will have a reasonably recent version of Apache, PHP
>>  and MySQL installed.
>>
>>  My first install attempt failed since the sysadmin hadn't given ALTER
>>  or DROP permissions to MYSQL, but the second attempt worked fine -
>>  using sql_update.php3 and then setup.php3 to get it all going.
>>
>>  Here are some suggestions
>>  1: Create some test scripts, so that a user can tell if the system is
>>  set up correctly, and if not knows what to ask for, I've started with
>>  admin/test.htm and admin/test.php which test PHP installation, and
>>  Server Side Includes, but it would be great to have something which
>>  excercised other common functions and maybe checked PHP options etc?
>
>  >
>
>  > 2: Stop any dependency on aadb.sql, I think that all it does is
>
>  > replicate what sql_update.php3 does anyway, so why have the
>
>  > functionality in two places?   Maybe just have setup.php3 run
>>  sql_update if the tables don't exist?
>>
>>  - Mitra
>>
>>
>
>
>
>-------------------------------------------------------
>This SF.NET email is sponsored by:
>SourceForge Enterprise Edition + IBM + LinuxWorld http://www.vasoftware.com
>_______________________________________________
>Apc-aa-coders mailing list
>Apc...@li...
>https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
>
>
>
>--
>Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
>02-6684-8096  or 0414-648-0722 
>
>Life is a Mystery to be Lived, not a Problem to be Solved


-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved

[Apc-aa-coders] GD library and image widths

[Mitra <mi...@mi...>]

Stanislav

Your image upload library looks really useful, timely as I need 
something like ti for a project I'm working on for a community radio 
station.

One question, and two problems

The question .... are GIF's supported - your code recognizes them as 
a supported type but your email said they weren't supported?

Problem 1: Are there simple instructions we could give people (in the 
FAQ and installation docs) for installing GD, since it seems that it 
is not present on most PHP installations.

Problem 2:
.... lets take a typical example .

I'll setup file upload to have parameters    image/*:200:100

So lets say the user uploads an image 400*100, it will get resampled 
to 200*50 which is great.

But ... how do we generate the output HTML , what I'd like to output is
<IMG SRC="/apc-aa-uploads/12345/foo.jpg" HEIGHT=50 WIDTH=200>
but I have no way to know that the height and width actually are 50 and 200.

I can think of several ways around this.
1: Have the upload function stuff the resampled height in the 
img_width..... and img_height..... fields
this is not so clever as there might be multiple images on a form.

2: Provide an output expansion that gets what is required, I think 
the ideal might be
{img:width:apc-aa-uploads/12345/foo.jpg} and 
{img:height:apc-aa-uploads/12345/foo.jpg}

So that the output expression becomes simpyl
<IMG SRC="{img_url.........}" HEIGHT={img:height:{img_url.........}} 
WIDTH={img:width:{img_url.........}}>
This would give maximum flexibility to output anything extra without 
requiring lots of switches and math in each output?

What do you think - could you provide a couple of functions that take 
a file name and return the width and height ?  if so then I'll code 
them into the string expansion code.

- Mitra



-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722

Life is a Mystery to be Lived, not a Problem to be Solved

RE: [Apc-aa-coders] Abandoning aadb.sql

[Marek T. <ma...@gn...>]

RE: [Apc-aa-coders] Abandoning aadb.sqlHi Mitra,
yes, yes, you are indeed right. The aadb.sql file is there for historical
reasons, this file has been there ever since, while sql_update.php3 was
introduced with 1.8 ? (guessing - doesn't really matter). I think you are
just the first one who has noticed that you can use sql update as well, even
if you don't have any database structure yet. Don't be ashamed of it - great
spotting !
I can rewrite the installation instructions if you want me to do do.
Best
 Marek
  -----Original Message-----
  From: apc...@li...
[mailto:apc...@li...]On Behalf Of Mitra
  Sent: 23 January 2003 22:43
  To: Honza Malik; apc...@so...
  Subject: RE: [Apc-aa-coders] Abandoning aadb.sql


  I agree that we don't want write access to the scripts.


  Honza - I'm probably missunderstanding something - what is the point of
aadb.sql ? What function does aadb.sql provide that sql_update.php3 doesn't?
What advantages are there of configuring apc-aa with aadb.sql rather than
via sql_update.php3?


  - Mitra




  At 9:52 PM +0100 23/1/03, Honza Malik wrote:
    I agree with droping aadb.sql, mainly if we can quite easily write an
    option for sql_update.php3, which will generate aadb.sql.

    I'm not sure about configuration using web interface. It would require
    write access to config.php3. I think we should never give webserver the
    rights to write to any AA script.

       Honza


    On ât, 2003-01-23 at 19:59, Ram Prasad wrote:
    > Also, it would be nice if we have a page to set up the initial
    > configuration. This way, the config.php3 can be set using the web
    > interface, which can then be used to configure apc-aa.
    >
    > - Ram
    >
    > -----Original Message-----
    > From: apc...@li...
    > [mailto:apc...@li...] On Behalf Of Mitra
    > Sent: Thursday, January 23, 2003 1:24 PM
    > To: apc...@so...
    > Subject: [Apc-aa-coders] Abandoning aadb.sql
    >
    >
    > I'm just installing APC-AA on a machine where I do not have any kind
    > of shell access. This is going to be the case for many potential
    > users of apc-aa.
    >
    > So of course, I could not follow the directions, which depend on shell
    > access.
    >
    > Typically all of doc/install-3.html can be skipped, since any
    > competent ISP will have a reasonably recent version of Apache, PHP
    > and MySQL installed.
    >
    > My first install attempt failed since the sysadmin hadn't given ALTER
    > or DROP permissions to MYSQL, but the second attempt worked fine -
    > using sql_update.php3 and then setup.php3 to get it all going.
    >
    > Here are some suggestions
    > 1: Create some test scripts, so that a user can tell if the system is
    > set up correctly, and if not knows what to ask for, I've started with
    > admin/test.htm and admin/test.php which test PHP installation, and
    > Server Side Includes, but it would be great to have something which
    > excercised other common functions and maybe checked PHP options etc?
    >
    > 2: Stop any dependency on aadb.sql, I think that all it does is
    > replicate what sql_update.php3 does anyway, so why have the
    > functionality in two places?   Maybe just have setup.php3 run
    > sql_update if the tables don't exist?
    >
    > - Mitra
    >
    >



    -------------------------------------------------------
    This SF.NET email is sponsored by:
    SourceForge Enterprise Edition + IBM + LinuxWorld
http://www.vasoftware.com
    _______________________________________________
    Apc-aa-coders mailing list
    Apc...@li...
    https://lists.sourceforge.net/lists/listinfo/apc-aa-coders




--
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
  02-6684-8096  or 0414-648-0722

  Life is a Mystery to be Lived, not a Problem to be Solved

[Apc-aa-coders] Ids of strange lineths.

[Mitra <mi...@mi...>]

Ok - stranger .... my guess was wrong, md5 returns a 32 character 
string, which cannot contain 00 or 27 so cannot end up as a shorter 
string.

To test, I ran a loop through 99999 samples of new_id, which should 
have generated about 400 bad ids by your statistics.

I think the bad ids in your database must be the result of some buggy 
earlier code.

Lets try three things.
1: Please check my test - appended at the end of this message.
2: Please re-run your test, see if the number of bad ids has gone up, 
if there is something bad then as long as you have created more than 
256 ids then statistically it should have increased by 1 or more.
3: Please send the code for your test, I'll run it on my installation 
and see how many odd length ids I have.

- Mitra


At 11:55 PM +0100 22/1/03, Honza Malik wrote:
>  >3) Your assumption that packed ids are EXACTLY 16 characters long (and
>  >unpacked are EXACTLY 32 characters long) is unfortunately wrong. I can't
>  >say you why - we generate the ids by new_id() function, but in our
>  >database are packed ids for item 16 (in 91907 items), 15 (366) and 14
>  >(4) characters long.
>
>  Strange - because new_id uses "md5" which has to return 128 bits.
>
>  I wonder if these were ids which generated a leading ascii zero, the
>  ratios look reasonable for this, i.e. 919007/366 = 251, which is
>  close to a one in 256 chance, i.e. something that occurred when 8
>  bits (2^8=256) all were zero.    This makes sense if we assume that
>  md5 returns a number which the ereg in new_id casts as a string, and
>  so ignores the leading zero.
>
>  I've fixed zids.php3 to allow 12..16 character packed ids, and 24..32
>  long, but this could break my tagged ids. I've also fixed
>  util.php3.new_id to only return a string of exactly 16 chars.
>
>  > >5) In zids.php3 you wrote:
>>  >
>>  >function id_t2l(&$str) {
>>  >#TODO: Look online for quicker way to substr last 32 chars - mitra
>>  >         if (ereg("(.*)([0-9a-f]{32})",$str,$parts))
>>  >                 return $parts[2];
>>  >
>>  >The quicker way is: substr($str, -32), but see 3)
>>
>>  Ok - I'll make this 24..32, and I guess we can't use tags that end in
>>  a hex character [0-9a-f)
>
>Or you can use some delimeter (tag:unpacked_id). As I write before, we
>never use tags so far, so the syntax is on your choice.


require "../include/config.php3";
require $GLOBALS[AA_INC_PATH]."util.php3";

$i = 0;
$j= 99999;
print "<br>Testing start, loops=$j: sample=";
$foo = new_id();
print $foo;
while ($i++ < $j) {
$foo = new_id();
if (strlen($foo) != 32) print "<br>FOO:$i:$foo:strlen=".strlen($foo);
if (strlen(pack_id($foo)) != 16) print "<br>FOO:$i:$foo:packed 
strlen=".strlen(\
pack_id($foo));
}
print "<br>Testing end sample = ";
$foo = new_id();
print $foo;

print "Done\n";


-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved

Re: [Apc-aa-coders] CountHits

[Mitra <mi...@mi...>]

This sounds like a trivial change, but I'm unclear where are all the 
places that need changing if a change like this is to be made to a 
key table like view?

Maybe this is something you could put in the Developers part of the 
=46AQ, I know that if I do it, I'll probably miss one place and break 
something.

-  Mitra


At 9:57 PM +0100 23/1/03, Honza Malik wrote:
>It's quite hard to decide, when to count hit and when not by script. The
>x-vid-item_id commend is mainly used for fulltext item display, so used
>aproximation is not so bad, but I agree it could be improved. The idea
>of view parameter sounds good.
>
>    Honza
>
>On P=E1, 2003-01-17 at 08:39, Mitra wrote:
>>  I'm doing some work on the flexible Ids that we discussed.
>>
>>  Looking at view.php3, it is very strange in the way it Counts hits -
>>  where parameters are passed with cmd[]=3Dx-vid-123-456 -
>>
>>  it increments the first item in the view, but not the others, this is
>>  going to make the first item look much more popular than the others?
>>
>>  Maybe this should be a parameter of the view, overridable in the
>>  code?  So for example as a site editor I might consider that showing
>>  a list of headlines didn't count as a "display", but showing the
>>  abstracts did?
>>
>>  - Mitra
>
>
>
>-------------------------------------------------------
>This SF.NET email is sponsored by:
>SourceForge Enterprise Edition + IBM + LinuxWorld http://www.vasoftware.com
>_______________________________________________
>Apc-aa-coders mailing list
>Apc...@li...
>https://lists.sourceforge.net/lists/listinfo/apc-aa-coders


-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved

RE: [Apc-aa-coders] Abandoning aadb.sql

[Mitra <mi...@mi...>]

I agree that we don't want write access to the scripts.

Honza - I'm probably missunderstanding something - what is the point 
of aadb.sql ? What function does aadb.sql provide that 
sql_update.php3 doesn't?  What advantages are there of configuring 
apc-aa with aadb.sql rather than via sql_update.php3?

- Mitra


At 9:52 PM +0100 23/1/03, Honza Malik wrote:
>I agree with droping aadb.sql, mainly if we can quite easily write an
>option for sql_update.php3, which will generate aadb.sql.
>
>I'm not sure about configuration using web interface. It would require
>write access to config.php3. I think we should never give webserver the
>rights to write to any AA script.
>
>    Honza
>
>
>On =E2t, 2003-01-23 at 19:59, Ram Prasad wrote:
>>  Also, it would be nice if we have a page to set up the initial
>>  configuration. This way, the config.php3 can be set using the web
>>  interface, which can then be used to configure apc-aa.
>>
>>  - Ram
>>
>>  -----Original Message-----
>>  From: apc...@li...
>>  [mailto:apc...@li...] On Behalf Of Mitra
>>  Sent: Thursday, January 23, 2003 1:24 PM
>>  To: apc...@so...
>>  Subject: [Apc-aa-coders] Abandoning aadb.sql
>>
>>
>>  I'm just installing APC-AA on a machine where I do not have any kind
>>  of shell access. This is going to be the case for many potential
>>  users of apc-aa.
>>
>>  So of course, I could not follow the directions, which depend on shell
>>  access.
>>
>>  Typically all of doc/install-3.html can be skipped, since any
>>  competent ISP will have a reasonably recent version of Apache, PHP
>>  and MySQL installed.
>>
>>  My first install attempt failed since the sysadmin hadn't given ALTER
>>  or DROP permissions to MYSQL, but the second attempt worked fine -
>>  using sql_update.php3 and then setup.php3 to get it all going.
>>
>>  Here are some suggestions
>>  1: Create some test scripts, so that a user can tell if the system is
>>  set up correctly, and if not knows what to ask for, I've started with
>>  admin/test.htm and admin/test.php which test PHP installation, and
>>  Server Side Includes, but it would be great to have something which
>>  excercised other common functions and maybe checked PHP options etc?
>  >
>  > 2: Stop any dependency on aadb.sql, I think that all it does is
>  > replicate what sql_update.php3 does anyway, so why have the
>  > functionality in two places?   Maybe just have setup.php3 run
>>  sql_update if the tables don't exist?
>>
>>  - Mitra
>>
>>
>
>
>
>-------------------------------------------------------
>This SF.NET email is sponsored by:
>SourceForge Enterprise Edition + IBM + LinuxWorld http://www.vasoftware.com
>_______________________________________________
>Apc-aa-coders mailing list
>Apc...@li...
>https://lists.sourceforge.net/lists/listinfo/apc-aa-coders


-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved

Re: [Apc-aa-coders] CountHits

[Honza M. <hon...@ec...>]

It's quite hard to decide, when to count hit and when not by script. The
x-vid-item_id commend is mainly used for fulltext item display, so used=20
aproximation is not so bad, but I agree it could be improved. The idea
of view parameter sounds good.

   Honza

On P=E1, 2003-01-17 at 08:39, Mitra wrote:
> I'm doing some work on the flexible Ids that we discussed.
>=20
> Looking at view.php3, it is very strange in the way it Counts hits -=20
> where parameters are passed with cmd[]=3Dx-vid-123-456 -
>=20
> it increments the first item in the view, but not the others, this is=20
> going to make the first item look much more popular than the others?
>=20
> Maybe this should be a parameter of the view, overridable in the=20
> code?  So for example as a site editor I might consider that showing=20
> a list of headlines didn't count as a "display", but showing the=20
> abstracts did?
>=20
> - Mitra

RE: [Apc-aa-coders] Abandoning aadb.sql

[Honza M. <hon...@ec...>]

I agree with droping aadb.sql, mainly if we can quite easily write an
option for sql_update.php3, which will generate aadb.sql.

I'm not sure about configuration using web interface. It would require
write access to config.php3. I think we should never give webserver the
rights to write to any AA script.

   Honza


On =C8t, 2003-01-23 at 19:59, Ram Prasad wrote:
> Also, it would be nice if we have a page to set up the initial
> configuration. This way, the config.php3 can be set using the web
> interface, which can then be used to configure apc-aa.
>=20
> - Ram
>=20
> -----Original Message-----
> From: apc...@li...
> [mailto:apc...@li...] On Behalf Of Mitra
> Sent: Thursday, January 23, 2003 1:24 PM
> To: apc...@so...
> Subject: [Apc-aa-coders] Abandoning aadb.sql
>=20
>=20
> I'm just installing APC-AA on a machine where I do not have any kind=20
> of shell access. This is going to be the case for many potential=20
> users of apc-aa.
>=20
> So of course, I could not follow the directions, which depend on shell
> access.
>=20
> Typically all of doc/install-3.html can be skipped, since any=20
> competent ISP will have a reasonably recent version of Apache, PHP=20
> and MySQL installed.
>=20
> My first install attempt failed since the sysadmin hadn't given ALTER=20
> or DROP permissions to MYSQL, but the second attempt worked fine -=20
> using sql_update.php3 and then setup.php3 to get it all going.
>=20
> Here are some suggestions
> 1: Create some test scripts, so that a user can tell if the system is=20
> set up correctly, and if not knows what to ask for, I've started with=20
> admin/test.htm and admin/test.php which test PHP installation, and=20
> Server Side Includes, but it would be great to have something which=20
> excercised other common functions and maybe checked PHP options etc?
>=20
> 2: Stop any dependency on aadb.sql, I think that all it does is=20
> replicate what sql_update.php3 does anyway, so why have the=20
> functionality in two places?   Maybe just have setup.php3 run=20
> sql_update if the tables don't exist?
>=20
> - Mitra
>=20
>=20

RE: [Apc-aa-coders] Abandoning aadb.sql

[Ram P. <ra...@pa...>]

Also, it would be nice if we have a page to set up the initial
configuration. This way, the config.php3 can be set using the web
interface, which can then be used to configure apc-aa.

- Ram

-----Original Message-----
From: apc...@li...
[mailto:apc...@li...] On Behalf Of Mitra
Sent: Thursday, January 23, 2003 1:24 PM
To: apc...@so...
Subject: [Apc-aa-coders] Abandoning aadb.sql


I'm just installing APC-AA on a machine where I do not have any kind 
of shell access. This is going to be the case for many potential 
users of apc-aa.

So of course, I could not follow the directions, which depend on shell
access.

Typically all of doc/install-3.html can be skipped, since any 
competent ISP will have a reasonably recent version of Apache, PHP 
and MySQL installed.

My first install attempt failed since the sysadmin hadn't given ALTER 
or DROP permissions to MYSQL, but the second attempt worked fine - 
using sql_update.php3 and then setup.php3 to get it all going.

Here are some suggestions
1: Create some test scripts, so that a user can tell if the system is 
set up correctly, and if not knows what to ask for, I've started with 
admin/test.htm and admin/test.php which test PHP installation, and 
Server Side Includes, but it would be great to have something which 
excercised other common functions and maybe checked PHP options etc?

2: Stop any dependency on aadb.sql, I think that all it does is 
replicate what sql_update.php3 does anyway, so why have the 
functionality in two places?   Maybe just have setup.php3 run 
sql_update if the tables don't exist?

- Mitra



-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722

Life is a Mystery to be Lived, not a Problem to be Solved



-------------------------------------------------------
This SF.net email is sponsored by: Scholarships for Techies! Can't
afford IT training? All 2003 ictp students receive scholarships. Get
hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more.
www.ictp.com/training/sourceforge.asp
_______________________________________________
Apc-aa-coders mailing list
Apc...@li...
https://lists.sourceforge.net/lists/listinfo/apc-aa-coders

[Apc-aa-coders] Abandoning aadb.sql

[Mitra <mi...@mi...>]

I'm just installing APC-AA on a machine where I do not have any kind 
of shell access. This is going to be the case for many potential 
users of apc-aa.

So of course, I could not follow the directions, which depend on shell access.

Typically all of doc/install-3.html can be skipped, since any 
competent ISP will have a reasonably recent version of Apache, PHP 
and MySQL installed.

My first install attempt failed since the sysadmin hadn't given ALTER 
or DROP permissions to MYSQL, but the second attempt worked fine - 
using sql_update.php3 and then setup.php3 to get it all going.

Here are some suggestions
1: Create some test scripts, so that a user can tell if the system is 
set up correctly, and if not knows what to ask for, I've started with 
admin/test.htm and admin/test.php which test PHP installation, and 
Server Side Includes, but it would be great to have something which 
excercised other common functions and maybe checked PHP options etc?

2: Stop any dependency on aadb.sql, I think that all it does is 
replicate what sql_update.php3 does anyway, so why have the 
functionality in two places?   Maybe just have setup.php3 run 
sql_update if the tables don't exist?

- Mitra



-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722

Life is a Mystery to be Lived, not a Problem to be Solved

Re: [Apc-aa-coders] Re: [Apc-aa-general] CVS updates for zids

[Honza M. <hon...@ec...>]

> >2) You are using array_slice, array_map functions which are implemente=
d
> >only in PHP4 (array_map PHP 4 >=3D 4.0.6). So far we use only PHP3
> >functions in AA. Maybe is the time to change, but we have at least ask
> >apc-aa-general.
>=20
> I'll ask.
>=20
> If necessary, maybe there are equivalents of any PHP4 functions we=20
> are using, for example an implementation of array_map written in=20
> PHP3, that we could conditionally include?

That's right for array_map and array_slice, but when we will start using
other PHP4 construct (like foreach()), then the PHP3 compatibility is
lost - there is no possibility, how to write it in PHP3.

> >On the other hand the 'x', 'y' and 'z' in
> >related_sel.php3 is not exactly the same as tagged ids - it is never
> >writen to the database. We add 'x', ... just as information to
> >StoreItem() function (see insert_fnc_ids() in include/itemfunc.php3 -
> >there  ut is used). In this function we remove 'x', ... and to the
> >database we store untagged item ids!

> This is what I keep asking (unanswered) questions about. In=20
> related_sel.php3 you have three buttons "Add" "Add Mutual" "Add=20
> Backward" which add "x", "y" or "z", if you add the result as an id=20
> then its is stripped, but if you add it as text it is stored, but=20
> doesn't seem to be used anywhere, and didn't work with related views!=20
> So why were the three buttons put in if the code doesn't use them, or=20
> was this some project that never got finished.

No, all is finished. We use 'x', 'y' and 'z' right if you add result as
an id. The 'x', 'y' and 'z' are used as 'flags' in insert_fnc_ids(). You
are right the 'x', 'y' and 'z' is stripped, but before it is stripped we
read this 'flag' and store the stripped id differently (based on the
'flag' value).

If it is 'x' (Add), we store the id to the current item.
If it is 'y' (Add Mutual), we store id to the current item AND the id of
current item we store in related item (so both items are then in
relation - mutualy)
If it is 'z' (Add Backward), we store only id of current item to related
item (the item with stripped id).

See insert_fnc_ids()!

If someone use related item window and stores the id as text, it is
wrong ussage of the field and it can't work. So far we never store taged
item into database.


> >
> >5) In zids.php3 you wrote:
> >
> >function id_t2l(&$str) {
> >#TODO: Look online for quicker way to substr last 32 chars - mitra
> >         if (ereg("(.*)([0-9a-f]{32})",$str,$parts))
> >                 return $parts[2];
> >
> >The quicker way is: substr($str, -32), but see 3)
>=20
> Ok - I'll make this 24..32, and I guess we can't use tags that end in=20
> a hex character [0-9a-f)

Or you can use some delimeter (tag:unpacked_id). As I write before, we
never use tags so far, so the syntax is on your choice.

   Honza



On St, 2003-01-22 at 22:54, Mitra wrote:
> At 6:08 PM +0100 22/1/03, Honza Malik wrote:
> >Hi Mitra,
> >
> >    thans for the changes, it is step forward. A few comments:
> >
> >1) I fixed small bug in zids->shortids() - check CVS
>=20
> Thanks - I guess none of my test views are using short ids!
>=20
> >2) You are using array_slice, array_map functions which are implemente=
d
> >only in PHP4 (array_map PHP 4 >=3D 4.0.6). So far we use only PHP3
> >functions in AA. Maybe is the time to change, but we have at least ask
> >apc-aa-general.
>=20
> I'll ask.
>=20
> If necessary, maybe there are equivalents of any PHP4 functions we=20
> are using, for example an implementation of array_map written in=20
> PHP3, that we could conditionally include?
>=20
> >
> >3) Your assumption that packed ids are EXACTLY 16 characters long (and
> >unpacked are EXACTLY 32 characters long) is unfortunately wrong. I can=
't
> >say you why - we generate the ids by new_id() function, but in our
> >database are packed ids for item 16 (in 91907 items), 15 (366) and 14
> >(4) characters long.
>=20
> Strange - because new_id uses "md5" which has to return 128 bits.
>=20
> I wonder if these were ids which generated a leading ascii zero, the=20
> ratios look reasonable for this, i.e. 919007/366 =3D 251, which is=20
> close to a one in 256 chance, i.e. something that occurred when 8=20
> bits (2^8=3D256) all were zero.    This makes sense if we assume that=20
> md5 returns a number which the ereg in new_id casts as a string, and=20
> so ignores the leading zero.
>=20
> I've fixed zids.php3 to allow 12..16 character packed ids, and 24..32=20
> long, but this could break my tagged ids. I've also fixed=20
> util.php3.new_id to only return a string of exactly 16 chars.
>=20
>=20
> >
> >4) You wrote:
> >>  The main reason for doing this was to add tagged ids, i.e. being ab=
le
> >>  to have a field in a record that includes an extra piece of
> >>  information to type a link, for example the existing related_sel.ph=
p3
> >>  adds an 'x','y' or 'z' for the direction of the link (although this
> >>  doesn't seem to be used anywhere at the moment!).
> >
> >It is good idea of tagged ids.
>=20
> Thanks - I'm building a site now that uses them.
>=20
> >On the other hand the 'x', 'y' and 'z' in
> >related_sel.php3 is not exactly the same as tagged ids - it is never
> >writen to the database. We add 'x', ... just as information to
> >StoreItem() function (see insert_fnc_ids() in include/itemfunc.php3 -
> >there  ut is used). In this function we remove 'x', ... and to the
> >database we store untagged item ids!
>=20
>=20
> This is what I keep asking (unanswered) questions about. In=20
> related_sel.php3 you have three buttons "Add" "Add Mutual" "Add=20
> Backward" which add "x", "y" or "z", if you add the result as an id=20
> then its is stripped, but if you add it as text it is stored, but=20
> doesn't seem to be used anywhere, and didn't work with related views!=20
> So why were the three buttons put in if the code doesn't use them, or=20
> was this some project that never got finished.
>=20
> I don't care too much ... except that I've been trying very hard not=20
> to break code, that I couldn't figure out how it could ever be used!
>=20
> >
> >5) In zids.php3 you wrote:
> >
> >function id_t2l(&$str) {
> >#TODO: Look online for quicker way to substr last 32 chars - mitra
> >         if (ereg("(.*)([0-9a-f]{32})",$str,$parts))
> >                 return $parts[2];
> >
> >The quicker way is: substr($str, -32), but see 3)
>=20
> Ok - I'll make this 24..32, and I guess we can't use tags that end in=20
> a hex character [0-9a-f)
>=20
> - Mitra
>=20
>=20
> >On =DAt, 2003-01-21 at 07:48, Mitra wrote:
> >>  Some big changes  on the CVS version.
> >>
> >>  DO NOT DOWNLOAD ON LIVE SERVERS YET
> >  >
> >  > The big change is the id encapsulation discussed on here a couple =
of
> >  > weeks ago. I've added a class "zids" which encapsulates an array o=
f
> >  > ids.
> >  >
> >  > Basically the way it works is to store the ids in whatever format
> >>  they are supplied in, and then has methods to convert and return th=
em
> >>  in the format needed, so for example it can be initialized with
> >>  packed ids, and then return long ids.
> >>
> >>  The main reason for doing this was to add tagged ids, i.e. being ab=
le
> >  > to have a field in a record that includes an extra piece of
> >>  information to type a link, for example the existing related_sel.ph=
p3
> >>  adds an 'x','y' or 'z' for the direction of the link (although this
> >>  doesn't seem to be used anywhere at the moment!).
> >>
> >>  In case you are interested I need the tagged ids for a project wher=
e
> >>  I want to flag businesses based on criteria (organic, community
> >>  supporter etc)  but want to have a value (good, medium, bad) on eac=
h
> >>  link.
> >>
> >>  I've changed code to use this class in places where I need it, its
> >>  generally applicable so gradually I'll change code as I work in
> >>  different places.  Because of the way the zids class is written, th=
e
> >>  tagged ids won't effect any code not designed to handle them becaus=
e
> >>  they'll be converted to long or quoted etc ids in those places.
> >>
> >>  The code also includes other methods for working on ids or collecti=
on
> >>  of ids, for example there is a "sqlin()" method which returns a
> >>  string like 'item.id IN ("12345","6778")'
> >>
> >>  Because this code makes changes in lots of places, its possible I'v=
e
> >>  broken something that I haven't been able to test. Please email me
> >>  and I'll make rapid fixes.
> >>
> >>  One other change I've made is to add a global variable that works
> >>  like $debug, but just gives warnings on places that something stran=
ge
> >>  is noticed , for example trying to unpack an id which is already
> >>  unpacked.  I encourage the use of this flag to help track errors,
> >>  without generating warnings for users.
> >>
> >>  It would be great if people with non-active installations could tes=
t
> >>  this, i.e. people for whom finding a bug won't muck up a live serve=
r!
> >>
> >>  Note the CVS version also includes Jakub's system-wide  m_gettext
> >>  changes, so that is another reason to download and test on a non-li=
ve
> >>  installation.
> >>
> >>  I'll be happy to take bug reports and treat them as a priority to g=
et
> >>  them fixed.
> >>
> >>  - Mitra
> >>
> >>
> >>  01/21/03 - code rule changes
> >>            - added global flag $errcheck like $debug but generates
> >>  warnings on str
> >>  angeness
> >>            - added zids.php3 to encapsulate id management including:
> >>              * adds support for tagged ids in many places (not all y=
et)
> >>              ** If related item window field set to store text, it w=
ill
> >>  cleanly ed
> >>  it these
> >>              *** More changes to related item window coming
> >>              * duplicates pack_id and unpack_id to pack_id128 and
> >>  unpack_id128 whi
> >>  ch check type
> >>            - commented out require of  _lang.php3 files from config.=
php3
> >
> >
> >
> >-------------------------------------------------------
> >This SF.net email is sponsored by: Scholarships for Techies!
> >Can't afford IT training? All 2003 ictp students receive scholarships.
> >Get hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more.
> >www.ictp.com/training/sourceforge.asp
> >_______________________________________________
> >Apc-aa-coders mailing list
> >Apc...@li...
> >https://lists.sourceforge.net/lists/listinfo/apc-aa-coders
>=20

Re: [Apc-aa-coders] default_fnc_variable

[Mitra <mi...@mi...>]

This was for the PTS project,  noone is using that code now, so the 
function can be commented out.

I'm working in that file at the moment, so I'll fix it in my next CVS update

- Mitra


At 8:56 PM +0100 22/1/03, Jakub Adamek wrote:
>I found a function default_fnc_variable() in itemfunc.php3 for Default
>field value added by Ram Prasad on 05-March-2002. This is a bad
>function. It returns ANY global variable the name of which you give as
>parameter.
>
>Ram, what did you need this function for? Let's correct it to return
>only one of allowed variables or delete it.
>
>Jakub
>
>
>
>-------------------------------------------------------
>This SF.net email is sponsored by: Scholarships for Techies!
>Can't afford IT training? All 2003 ictp students receive scholarships.
>Get hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more.
>www.ictp.com/training/sourceforge.asp
>_______________________________________________
>Apc-aa-coders mailing list
>Apc...@li...
>https://lists.sourceforge.net/lists/listinfo/apc-aa-coders


-- 
Mitra Technology Consulting -  www.mitra.biz - mi...@mi...
02-6684-8096  or 0414-648-0722 

Life is a Mystery to be Lived, not a Problem to be Solved