tavi-cvs-commit — CVS commit changelogs should be forwarded to this list

Re: [Tavi-cvs-commit] CVS: tavi/parse macros.php,1.10,1.10.2.1 save.php,1.5,1.5.2.1 transforms.php,1.25,1.25.2.1

[Scott M. <sm...@an...>]

Even,

> > >    if($macros_index != -1)
> > > -    { $ParseEngine[$macros_index] = 'parse_noop'; }
> > > +    { $ParseEngine[$macros_index] = 'parse_clear'; }
> > 
> > I realize you've now changed this to parse_no_macros.  But realize that
> > either this or parse_no_macros will prevent category pages from showing
> > up as having links to pages in their category.  Consider what will happen
> > to andstuff.org/CategoryHomePage, for example.
> 
> I thought about that, and my(!) decision was to not have macros parsed
> for links. But I see that someone might argue with my reasoning. Do you
> know if this maybe should be an option to enable/disable links from
> macros?
> 
> Regarding andstuff.org/CategoryHomePage and the backlinks from that one.
> It would not be affected since I just changed the on save stuff, and not
> the find function. Am I mistaken in that?

I wasn't thinking about backlinks; as long as the backlink function uses
find and not the link database we're okay.  I was specifically thinking of
the link database.  Consider a page that is only referenced in a category
list.  Your change causes that page to appear on OrphanedPages, which is
not correct; the page is in fact not orphaned.  Essentially, I am concerned
that a category's contents are not associated with that category for all the
various link database purposes.

Just my opinion. :-)  Since y'all are now in charge I defer to your decision.


  -- Scott

-- 
http://scott.andstuff.org/

[Tavi-cvs-commit] CVS: tavi/parse save.php,1.7,1.8

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi/parse
In directory sc8-pr-cvs1:/tmp/cvs-serv11777

Modified Files:
	save.php 
Log Message:
Fixed whitespace stuff...

Index: save.php
===================================================================
RCS file: /cvsroot/tavi/tavi/parse/save.php,v
retrieving revision 1.7
retrieving revision 1.8
diff -u -d -r1.7 -r1.8
--- save.php	30 Jun 2003 10:47:00 -0000	1.7
+++ save.php	1 Jul 2003 19:26:59 -0000	1.8
@@ -93,8 +93,6 @@
     $pagestore->new_sisterwiki($page, $result[1], $result[2]);
   }
 
-
   return $text;
 }
-
 ?>

[Tavi-cvs-commit] CVS: tavi/parse transforms.php,1.27,1.28

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi/parse
In directory sc8-pr-cvs1:/tmp/cvs-serv11625

Modified Files:
	transforms.php 
Log Message:
Fixed whitespace stuff, and removed doubling of parse_no_macros()

Index: transforms.php
===================================================================
RCS file: /cvsroot/tavi/tavi/parse/transforms.php,v
retrieving revision 1.27
retrieving revision 1.28
diff -u -d -r1.27 -r1.28
--- transforms.php	30 Jun 2003 10:47:01 -0000	1.27
+++ transforms.php	1 Jul 2003 19:26:22 -0000	1.28
@@ -172,12 +172,6 @@
 
 }
 
-function parse_no_macros($text)
-{
-  return preg_replace('/\\[\\[([^] ]+( [^]]+)?)]]/e',
-                      "", $text, -1);
-}
-
 function macro_token($macro, $trail)
 {
   global $ViewMacroEngine;
@@ -634,5 +628,4 @@
 
   return $text;
 }
-
 ?>

[Tavi-cvs-commit] CVS: tavi/lib pagestore.php,1.6,1.7

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi/lib
In directory sc8-pr-cvs1:/tmp/cvs-serv11445

Modified Files:
	pagestore.php 
Log Message:
Fixed whitespace stuff...

Index: pagestore.php
===================================================================
RCS file: /cvsroot/tavi/tavi/lib/pagestore.php,v
retrieving revision 1.6
retrieving revision 1.7
diff -u -d -r1.6 -r1.7
--- pagestore.php	30 Jun 2003 10:47:00 -0000	1.6
+++ pagestore.php	1 Jul 2003 19:25:30 -0000	1.7
@@ -55,7 +55,7 @@
     $qid = $this->dbh->query("SELECT time, author, version, username, " .
                              "comment " .
                              "FROM $PgTbl WHERE title='$page' " .
-                 "ORDER BY version DESC");
+                             "ORDER BY version DESC");
 
     $list = array();
     while(($result = $this->dbh->result($qid)))

Re: [Tavi-cvs-commit] Re: tabs

[Kimmo S. <kim...@su...>]

The problem with that is that I'm more lazy... :-)

In particular, I use "set autoindent" with vi.  With that set, vi will
place me at the same indent as the line I hit enter on (in insert mode).
It will use the least number of characters to achieve that, combining
both tabs and spaces.  So while I can indent the first line with spaces,
the subsequent lines will have tabs.

And while I could indent with the spacebar, I generally don't, as ^T
and ^D do "the right thing" across projects, given the correct config.
NetBSD uses tabs only, I use "set shiftwidth=4" in my own code, and
'Tavi uses "set shiftwdith=2".  By simply adjusting the vi config I
won't have to consciously remember the project for which I am writing.

+ Kim


| From:    Scott Moonen <sm...@an...>
| Date:    Mon, 30 Jun 2003 17:20:37 -0700
|
| > How do you get rid of tabs?  I can't figure out how to do that with vi
| > off the top of my head (or a quick look at the manual).
|
| Hehe.  I did it the lazy way; I always use the spacebar and never the Tab
| key.  When I was the only one developing, I never had to worry about
| expanding. :-)
|
| >
| > Should we install a commit filter that runs "expand" before writing to
| > the CVS repository?
| >
| > + Kim
| >
| >
| > | From:    Scott Moonen <sm...@an...>
| > | Date:    Mon, 30 Jun 2003 13:55:53 -0700
| > |
| > | > I will look into whether my editor inserts or removes tabs. Do we have
| > | > any policy as to not using/using tabs within code? (Or should I just
| > | > look for an option to leave tabs as they are within my editor? )
| > |
| > | I have always tried to keep the source completely free of tabs, with
| > | indentation at two spaces.  I'd recommend that continue.
| > |
| > |
| > |   -- Scott
| > |
| > | --
| > | http://scott.andstuff.org/
| > |
| >
| >
| > -------------------------------------------------------
| > This SF.Net email sponsored by: Free pre-built ASP.NET sites including
| > Data Reports, E-commerce, Portals, and Forums are available now.
| > Download today and enter to win an XBOX or Visual Studio .NET.
| > http://aspnet.click-url.com/go/psa00100006ave/direct;at.asp_061203_01/01
| > _______________________________________________
| > Tavi-cvs-commit mailing list
| > Tav...@li...
| > https://lists.sourceforge.net/lists/listinfo/tavi-cvs-commit
| >
|
| --
| http://scott.andstuff.org/
|

Re: [Tavi-cvs-commit] Re: tabs

[Scott M. <sm...@an...>]

> How do you get rid of tabs?  I can't figure out how to do that with vi
> off the top of my head (or a quick look at the manual).

Hehe.  I did it the lazy way; I always use the spacebar and never the Tab
key.  When I was the only one developing, I never had to worry about
expanding. :-)

> 
> Should we install a commit filter that runs "expand" before writing to
> the CVS repository?
> 
> + Kim
> 
> 
> | From:    Scott Moonen <sm...@an...>
> | Date:    Mon, 30 Jun 2003 13:55:53 -0700
> |
> | > I will look into whether my editor inserts or removes tabs. Do we have
> | > any policy as to not using/using tabs within code? (Or should I just
> | > look for an option to leave tabs as they are within my editor? )
> |
> | I have always tried to keep the source completely free of tabs, with
> | indentation at two spaces.  I'd recommend that continue.
> |
> |
> |   -- Scott
> |
> | --
> | http://scott.andstuff.org/
> |
> 
> 
> -------------------------------------------------------
> This SF.Net email sponsored by: Free pre-built ASP.NET sites including
> Data Reports, E-commerce, Portals, and Forums are available now.
> Download today and enter to win an XBOX or Visual Studio .NET.
> http://aspnet.click-url.com/go/psa00100006ave/direct;at.asp_061203_01/01
> _______________________________________________
> Tavi-cvs-commit mailing list
> Tav...@li...
> https://lists.sourceforge.net/lists/listinfo/tavi-cvs-commit
> 

-- 
http://scott.andstuff.org/

[Tavi-cvs-commit] Re: alpha patterns

[Even H. <ev...@pv...>]

Hi Scott!

On Mon, Jun 30, 2003 at 02:12:46PM -0700, Scott Moonen wrote:
> I think you're right in \w being better.  Does it include underscores, though?
> Don't know if that matters in many of the places we use it.

I believe it not to include underscores, but haven't tested lately.
Should I look into changing this, or should we leave it as is since it
mostly works as expected? With the exception of the 'international'
characters?
 
> Perhaps chr(1) is best for FlgChr? :-)

Maybe... Ideally we should have two \w's, one for lowercase and one for
uppercase... 

Regards,
Even Holen
-- 
<><   Even Holen, ev...@pv..., http://www.pvv.ntnu.no/~evenh/   :-)

Re: [Tavi-cvs-commit] Re: alpha patterns

[Kimmo S. <kim...@su...>]

Or chr(127) (DEL), as a "regular" character that can be produced on
a keyboard, yet doesn't appear in "printable" text.

+ Kim


| From:    Scott Moonen <sm...@an...>
| Date:    Mon, 30 Jun 2003 14:12:46 -0700
|
| Even,
|
| I think you're right in \w being better.  Does it include underscores, though
| ?
| Don't know if that matters in many of the places we use it.
|
| Perhaps chr(1) is best for FlgChr? :-)
|
|
|   -- Scott
|
| On Mon, Jun 30, 2003 at 10:53:01PM +0200, Even Holen wrote:
| > On Mon, Jun 30, 2003 at 01:48:22PM -0700, Scott Moonen wrote:
| > > Even,
| > >
| >   From diff of tavi/lib/db.php
| > > > -$LowerPtn = "[a-z\xdf-\xff]";
| > > > -$AlphaPtn = "[A-Za-z\xc0-\xff]";
| > > > +$LowerPtn = "[a-z\xdf-\xfe]";
| > > > +$AlphaPtn = "[A-Za-z\xc0-\xfe]";
| > > >  $LinkPtn = $UpperPtn . $AlphaPtn . '*' . $LowerPtn . '+' .
| > > >             $UpperPtn . $AlphaPtn . '*(\\/' . $UpperPtn . $AlphaPtn . '
| *)?';
| > >
| > > Are you sure that \xff is not a valid lower-case character in some charac
| ter
| > > sets?  I prefer to fix this problem by setting FlgChr to chr(128), instea
| d of
| > > exempting \xff from the regex.
| >
| > I thought about that for a while, and I believe it's more likely that
| > chr(128) is another character than chr(255). If you could provide some
| > good proof otherwise, I will change. My main concern is to eliminate the
| > bug...
| >
| > On a side note: Is there a reason why [A-Za-z] is used in a lot of
| > regexp instead of \w ? Also in this one we could use the fact that \w is
| > supposed to catch any word character. The downside is of cource any
| > locale which wrongly set, but the upside is that we get a lot of
| > nationalisation issues out of the way _if_ we used \w instead of
| > [A-Za-z]...
| >
| > Regards,
| > Even Holen
| > --
| > <><   Even Holen, ev...@pv..., http://www.pvv.ntnu.no/~evenh/   :-)
| >
|
| --
| http://scott.andstuff.org/
|
|
| -------------------------------------------------------
| This SF.Net email sponsored by: Free pre-built ASP.NET sites including
| Data Reports, E-commerce, Portals, and Forums are available now.
| Download today and enter to win an XBOX or Visual Studio .NET.
| http://aspnet.click-url.com/go/psa00100006ave/direct;at.asp_061203_01/01
| _______________________________________________
| Tavi-cvs-commit mailing list
| Tav...@li...
| https://lists.sourceforge.net/lists/listinfo/tavi-cvs-commit
|

[Tavi-cvs-commit] Re: tabs

[Kimmo S. <kim...@su...>]

How do you get rid of tabs?  I can't figure out how to do that with vi
off the top of my head (or a quick look at the manual).

Should we install a commit filter that runs "expand" before writing to
the CVS repository?

+ Kim


| From:    Scott Moonen <sm...@an...>
| Date:    Mon, 30 Jun 2003 13:55:53 -0700
|
| > I will look into whether my editor inserts or removes tabs. Do we have
| > any policy as to not using/using tabs within code? (Or should I just
| > look for an option to leave tabs as they are within my editor? )
|
| I have always tried to keep the source completely free of tabs, with
| indentation at two spaces.  I'd recommend that continue.
|
|
|   -- Scott
|
| --
| http://scott.andstuff.org/
|

[Tavi-cvs-commit] Re: alpha patterns

[Scott M. <sm...@an...>]

Even,

I think you're right in \w being better.  Does it include underscores, though?
Don't know if that matters in many of the places we use it.

Perhaps chr(1) is best for FlgChr? :-)


  -- Scott

On Mon, Jun 30, 2003 at 10:53:01PM +0200, Even Holen wrote:
> On Mon, Jun 30, 2003 at 01:48:22PM -0700, Scott Moonen wrote:
> > Even,
> > 
>   From diff of tavi/lib/db.php
> > > -$LowerPtn = "[a-z\xdf-\xff]";
> > > -$AlphaPtn = "[A-Za-z\xc0-\xff]";
> > > +$LowerPtn = "[a-z\xdf-\xfe]";
> > > +$AlphaPtn = "[A-Za-z\xc0-\xfe]";
> > >  $LinkPtn = $UpperPtn . $AlphaPtn . '*' . $LowerPtn . '+' .
> > >             $UpperPtn . $AlphaPtn . '*(\\/' . $UpperPtn . $AlphaPtn . '*)?';
> > 
> > Are you sure that \xff is not a valid lower-case character in some character
> > sets?  I prefer to fix this problem by setting FlgChr to chr(128), instead of
> > exempting \xff from the regex.
> 
> I thought about that for a while, and I believe it's more likely that
> chr(128) is another character than chr(255). If you could provide some
> good proof otherwise, I will change. My main concern is to eliminate the
> bug...
> 
> On a side note: Is there a reason why [A-Za-z] is used in a lot of
> regexp instead of \w ? Also in this one we could use the fact that \w is 
> supposed to catch any word character. The downside is of cource any
> locale which wrongly set, but the upside is that we get a lot of
> nationalisation issues out of the way _if_ we used \w instead of
> [A-Za-z]...
> 
> Regards,
> Even Holen
> -- 
> <><   Even Holen, ev...@pv..., http://www.pvv.ntnu.no/~evenh/   :-)
> 

-- 
http://scott.andstuff.org/

Re: [Tavi-cvs-commit] CVS: tavi/action save.php,1.8,1.9

[Kimmo S. <kim...@su...>]

I'm a huge fan of "cvs diff -u", as well as "cvs annotate" (the infamous
"cvs blame" command).  :-)

Regards,
+ Kim


| From:    Even Holen <ev...@pv...>
| Date:    Mon, 30 Jun 2003 22:47:31 +0200
|
| On Mon, Jun 30, 2003 at 04:19:16PM -0400, Kimmo Suominen wrote:
| > Here's another "odd" change -- why the added newline?  I prefer
| > not to have extra whitespace at the end of functions (or files).
|
| Sorry about that! I agree with you, no extra whitespace at the end of
| functions (or files).
|
| It's been some time since I've been using cvs/rcs on files so I just
| need a little period to adjust back into using it. I.e. to check diff's
| before I commit changes... Slap on my wrist! ;-)
|
| Regards,
| Even Holen
| --
| <><   Even Holen, ev...@pv..., http://www.pvv.ntnu.no/~evenh/   :-)
|

[smoonen@andstuff.org: Re: [Tavi-cvs-commit] CVS: tavi/parse macros.php,1.10,1.10.2.1 save.php,1.5,1.5.2.1 transforms.php,1.25,1.25.2.1]

[Scott M. <sm...@an...>]

Forgot to copy list.  -- Scott

----- Forwarded message from Scott Moonen <sm...@an...> -----
Date: Mon, 30 Jun 2003 14:07:22 -0700
From: Scott Moonen <sm...@an...>
To: Even Holen <ho...@us...>
Subject: Re: [Tavi-cvs-commit] CVS: tavi/parse macros.php,1.10,1.10.2.1 save.php,1.5,1.5.2.1 transforms.php,1.25,1.25.2.1

Even,

>    $q1 = $pagestore->dbh->query("SELECT l.link, SUM(l.count) AS ct, p.title " .
>                                 "FROM $LkTbl AS l LEFT JOIN $PgTbl AS p " .
>                                 "ON l.link = p.title " .
> +                               "WHERE p.title IS NULL " .
>                                 "GROUP BY l.link " .
> -                               "HAVING p.title IS NULL " .
>                                 "ORDER BY ct DESC, l.link");

Be sure to test this well under both old and new versions of MySQL.
When I converted the PHP loop to a join, I recall there being some
trickiness (can't remember specifics now, unfortunately).  I'm just
concerned that there was some particular reason I chose to use
"HAVING" instead of "WHERE".

>    if($macros_index != -1)
> -    { $ParseEngine[$macros_index] = 'parse_noop'; }
> +    { $ParseEngine[$macros_index] = 'parse_clear'; }

I realize you've now changed this to parse_no_macros.  But realize that
either this or parse_no_macros will prevent category pages from showing
up as having links to pages in their category.  Consider what will happen
to andstuff.org/CategoryHomePage, for example.


  -- Scott

-- 
http://scott.andstuff.org/

Re: [Tavi-cvs-commit] CVS: tavi/lib db.php,1.3,1.4 defaults.php,1.5,1.6 pagestore.php,1.5,1.6

[Kimmo S. <kim...@su...>]

| From:    Even Holen <ev...@pv...>
| Date:    Mon, 30 Jun 2003 22:44:04 +0200
|
| On Mon, Jun 30, 2003 at 04:18:31PM -0400, Kimmo Suominen wrote:
| > Even,
| >
| > A couple of questions below...
| >
| > |      if(!($qid = mysql_query($text, $this->handle)))
| > | -      { die($ErrorDatabaseQuery); }
| > | +      { die("<strong>$ErrorDatabaseQuery</strong><p>$text</p>"); }
| > |      return $qid;
| >
| > Have you made sure that passwords do not get exposed by this?  I didn't
| > have time to verify that from the code myself yet.
|
| I admit that I haven't thought about that. But I don't believe that's
| very likely in this setting. Do you have a test setting which would
| trigger such an error?

I would test with purposely misconfiguring the password on the test Wiki
and look for the error.  I can't think of other cases where the password
could be included in the error message.

I guess that would come from mysql_connect anyway, so maybe there is no
problem here at all.

| > | -			     "ORDER BY version DESC");
| > | +                 "ORDER BY version DESC");
| >
| > Why this change?
|
| That is a change made in my editor by changing tabs to spaces. I was not
| aware of that change. (I know it said it changed some tabs to spaces,
| but I didn't notice where)
|
| I will look into whether my editor inserts or removes tabs. Do we have
| any policy as to not using/using tabs within code? (Or should I just
| look for an option to leave tabs as they are within my editor? )
|
| Regards,
| Even Holen

You should get a behaviour from vi that matches the current code if you
use the following settings:

    set shiftwidth=2 tabstop=8

In other words, tabstops are 8 characters (since this is what all hardware
and terminal emulators assume generally, so it makes no sense to change
that) and indents are 2 characters.  Hard tabs will be used where possible.
(Indenting with vi is done in insert mode using CTRL-T, and undenting is
done in insert mode using CTRL-D.)

[An alternative would be to always use hard tabs, and everyone could then
just tell their editor to display these at a width that looks pleasing to
them -- e.g. with vi "set tabstop=2".  However, that would mean indenting
all the code again, and I'd rather not do that.]

Note that I'm only using vi as an example here since I'm familiar with it.

Regards,
+ Kim

Re: [Tavi-cvs-commit] CVS: tavi/lib db.php,1.3,1.3.4.1 defaults.php,1.5,1.5.2.1 diff.php,1.5,1.5.4.1 pagestore.php,1.5,1.5.2.1

[Even H. <eve...@pv...>]

On Mon, Jun 30, 2003 at 01:48:22PM -0700, Scott Moonen wrote:
> Even,
> 
  From diff of tavi/lib/db.php
> > -$LowerPtn = "[a-z\xdf-\xff]";
> > -$AlphaPtn = "[A-Za-z\xc0-\xff]";
> > +$LowerPtn = "[a-z\xdf-\xfe]";
> > +$AlphaPtn = "[A-Za-z\xc0-\xfe]";
> >  $LinkPtn = $UpperPtn . $AlphaPtn . '*' . $LowerPtn . '+' .
> >             $UpperPtn . $AlphaPtn . '*(\\/' . $UpperPtn . $AlphaPtn . '*)?';
> 
> Are you sure that \xff is not a valid lower-case character in some character
> sets?  I prefer to fix this problem by setting FlgChr to chr(128), instead of
> exempting \xff from the regex.

I thought about that for a while, and I believe it's more likely that
chr(128) is another character than chr(255). If you could provide some
good proof otherwise, I will change. My main concern is to eliminate the
bug...

On a side note: Is there a reason why [A-Za-z] is used in a lot of
regexp instead of \w ? Also in this one we could use the fact that \w is 
supposed to catch any word character. The downside is of cource any
locale which wrongly set, but the upside is that we get a lot of
nationalisation issues out of the way _if_ we used \w instead of
[A-Za-z]...

Regards,
Even Holen
-- 
<><   Even Holen, ev...@pv..., http://www.pvv.ntnu.no/~evenh/   :-)

[Tavi-cvs-commit] Re: tabs

[Scott M. <sm...@an...>]

> I will look into whether my editor inserts or removes tabs. Do we have
> any policy as to not using/using tabs within code? (Or should I just
> look for an option to leave tabs as they are within my editor? )

I have always tried to keep the source completely free of tabs, with
indentation at two spaces.  I'd recommend that continue.


  -- Scott

-- 
http://scott.andstuff.org/

Re: [Tavi-cvs-commit] CVS: tavi/action save.php,1.8,1.9

[Even H. <ev...@pv...>]

On Mon, Jun 30, 2003 at 04:19:16PM -0400, Kimmo Suominen wrote:
> Here's another "odd" change -- why the added newline?  I prefer
> not to have extra whitespace at the end of functions (or files).

Sorry about that! I agree with you, no extra whitespace at the end of
functions (or files). 

It's been some time since I've been using cvs/rcs on files so I just
need a little period to adjust back into using it. I.e. to check diff's
before I commit changes... Slap on my wrist! ;-)

Regards,
Even Holen
-- 
<><   Even Holen, ev...@pv..., http://www.pvv.ntnu.no/~evenh/   :-)

Re: [Tavi-cvs-commit] CVS: tavi/lib db.php,1.3,1.4 defaults.php,1.5,1.6 pagestore.php,1.5,1.6

[Even H. <ev...@pv...>]

On Mon, Jun 30, 2003 at 04:18:31PM -0400, Kimmo Suominen wrote:
> Even,
> 
> A couple of questions below...
> 
> |      if(!($qid = mysql_query($text, $this->handle)))
> | -      { die($ErrorDatabaseQuery); }
> | +      { die("<strong>$ErrorDatabaseQuery</strong><p>$text</p>"); }
> |      return $qid;
> 
> Have you made sure that passwords do not get exposed by this?  I didn't
> have time to verify that from the code myself yet.

I admit that I haven't thought about that. But I don't believe that's
very likely in this setting. Do you have a test setting which would
trigger such an error?

> | -			     "ORDER BY version DESC");
> | +                 "ORDER BY version DESC");
> 
> Why this change?

That is a change made in my editor by changing tabs to spaces. I was not
aware of that change. (I know it said it changed some tabs to spaces,
but I didn't notice where)

I will look into whether my editor inserts or removes tabs. Do we have
any policy as to not using/using tabs within code? (Or should I just
look for an option to leave tabs as they are within my editor? )

Regards,
Even Holen
-- 
<><   Even Holen, ev...@pv..., http://www.pvv.ntnu.no/~evenh/   :-)

Re: [Tavi-cvs-commit] CVS: tavi/lib db.php,1.3,1.3.4.1 defaults.php,1.5,1.5.2.1 diff.php,1.5,1.5.4.1 pagestore.php,1.5,1.5.2.1

[Scott M. <sm...@an...>]

Even,

On Sun, Jun 29, 2003 at 11:34:15AM -0700, Even Holen wrote:
> Index: defaults.php
> ===================================================================
> RCS file: /cvsroot/tavi/tavi/lib/defaults.php,v
> retrieving revision 1.5
> retrieving revision 1.5.2.1
> diff -u -d -r1.5 -r1.5.2.1
> --- defaults.php	22 Feb 2002 14:46:08 -0000	1.5
> +++ defaults.php	29 Jun 2003 18:34:13 -0000	1.5.2.1
> @@ -18,8 +18,8 @@
>  //   Note that changing this might require a change to parse/transforms.php so
>  //   that parse_wikiname knows how many parentheses are included in $LinkPtn.
>  $UpperPtn = "[A-Z\xc0-\xde]";
> -$LowerPtn = "[a-z\xdf-\xff]";
> -$AlphaPtn = "[A-Za-z\xc0-\xff]";
> +$LowerPtn = "[a-z\xdf-\xfe]";
> +$AlphaPtn = "[A-Za-z\xc0-\xfe]";
>  $LinkPtn = $UpperPtn . $AlphaPtn . '*' . $LowerPtn . '+' .
>             $UpperPtn . $AlphaPtn . '*(\\/' . $UpperPtn . $AlphaPtn . '*)?';

Are you sure that \xff is not a valid lower-case character in some character
sets?  I prefer to fix this problem by setting FlgChr to chr(128), instead of
exempting \xff from the regex.


  -- Scott


-- 
http://scott.andstuff.org/

Re: [Tavi-cvs-commit] CVS: tavi/lib db.php,1.3,1.4 defaults.php,1.5,1.6 pagestore.php,1.5,1.6

[Kimmo S. <kim...@su...>]

Even,

A couple of questions below...

| From:    Even Holen <ho...@us...>
| Date:    Mon, 30 Jun 2003 03:47:03 -0700
|
| Update of /cvsroot/tavi/tavi/lib
| In directory sc8-pr-cvs1:/tmp/cvs-serv2396/lib
|
| Modified Files:
| 	db.php defaults.php pagestore.php
| Log Message:
| Merged in release-v0_22-bugfix-branch-1
|
| Index: db.php
| ===================================================================
| RCS file: /cvsroot/tavi/tavi/lib/db.php,v
| retrieving revision 1.3
| retrieving revision 1.4
| diff -u -d -r1.3 -r1.4
| --- db.php	30 Nov 2001 22:10:15 -0000	1.3
| +++ db.php	30 Jun 2003 10:47:00 -0000	1.4
| @@ -28,7 +28,7 @@
|      global $ErrorDatabaseQuery;
|
|      if(!($qid = mysql_query($text, $this->handle)))
| -      { die($ErrorDatabaseQuery); }
| +      { die("<strong>$ErrorDatabaseQuery</strong><p>$text</p>"); }
|      return $qid;
|    }

Have you made sure that passwords do not get exposed by this?  I didn't
have time to verify that from the code myself yet.

| Index: pagestore.php
| ===================================================================
| RCS file: /cvsroot/tavi/tavi/lib/pagestore.php,v
| retrieving revision 1.5
| retrieving revision 1.6
| diff -u -d -r1.5 -r1.6
| --- pagestore.php	22 Feb 2002 14:46:08 -0000	1.5
| +++ pagestore.php	30 Jun 2003 10:47:00 -0000	1.6
| @@ -55,7 +55,7 @@
|      $qid = $this->dbh->query("SELECT time, author, version, username, " .
|                               "comment " .
|                               "FROM $PgTbl WHERE title='$page' " .
| -			     "ORDER BY version DESC");
| +                 "ORDER BY version DESC");
|
|      $list = array();
|      while(($result = $this->dbh->result($qid)))

Why this change?

+ Kim

[Tavi-cvs-commit] CVS: tavi/action save.php,1.8,1.9

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi/action
In directory sc8-pr-cvs1:/tmp/cvs-serv2396/action

Modified Files:
	save.php 
Log Message:
Merged in release-v0_22-bugfix-branch-1

Index: save.php
===================================================================
RCS file: /cvsroot/tavi/tavi/action/save.php,v
retrieving revision 1.8
retrieving revision 1.9
diff -u -d -r1.8 -r1.9
--- save.php	22 Feb 2002 14:46:08 -0000	1.8
+++ save.php	30 Jun 2003 10:47:00 -0000	1.9
@@ -72,5 +72,6 @@
   parseText($document, $SaveMacroEngine, $page);
 
   $pagestore->unlock();                 // End "transaction".
+
 }
 ?>

[Tavi-cvs-commit] CVS: tavi/parse macros.php,1.10,1.11 save.php,1.6,1.7 transforms.php,1.26,1.27

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi/parse
In directory sc8-pr-cvs1:/tmp/cvs-serv2396/parse

Modified Files:
	macros.php save.php transforms.php 
Log Message:
Merged in release-v0_22-bugfix-branch-1

Index: macros.php
===================================================================
RCS file: /cvsroot/tavi/tavi/parse/macros.php,v
retrieving revision 1.10
retrieving revision 1.11
diff -u -d -r1.10 -r1.11
--- macros.php	22 Feb 2002 14:46:08 -0000	1.10
+++ macros.php	30 Jun 2003 10:47:00 -0000	1.11
@@ -142,6 +142,9 @@
         { $first = 0; }
 
       $text = $text . html_ref($page[1], $page[1]);
+      if ($page[4] == 0 ) {
+        $text .= " ** Empty, scheduled for deletion ** ";
+      }
     }
   }
 
@@ -159,9 +162,10 @@
   $q1 = $pagestore->dbh->query("SELECT l.link, SUM(l.count) AS ct, p.title " .
                                "FROM $LkTbl AS l LEFT JOIN $PgTbl AS p " .
                                "ON l.link = p.title " .
+                               "WHERE p.title IS NULL " .
                                "GROUP BY l.link " .
-                               "HAVING p.title IS NULL " .
                                "ORDER BY ct DESC, l.link");
+
 
   while(($result = $pagestore->dbh->result($q1)))
   {

Index: save.php
===================================================================
RCS file: /cvsroot/tavi/tavi/parse/save.php,v
retrieving revision 1.6
retrieving revision 1.7
diff -u -d -r1.6 -r1.7
--- save.php	30 Jun 2003 10:27:16 -0000	1.6
+++ save.php	30 Jun 2003 10:47:00 -0000	1.7
@@ -11,7 +11,6 @@
 {
   global $pagestore, $page, $ParseEngine, $Entity, $ParseObject;
   static $called = 0;
-
   $macros_index = -1;
   $transclude_index = -1;
   $elements_index = -1;
@@ -68,12 +67,11 @@
     $called = 1;
   }
 
-  if(preg_match('/^\\*InterWiki:\\s+([A-Z][A-Za-z]+)\s+(http:[^\\s]+)/',
+  if(preg_match('/^\*InterWiki:\s+\!?([A-Za-z]+)\s+(https?:[^\s]+)/',
                 $text, $result))
   {
     $pagestore->new_interwiki($page, $result[1], $result[2]);
   }
-
   return $text;
 }
 

Index: transforms.php
===================================================================
RCS file: /cvsroot/tavi/tavi/parse/transforms.php,v
retrieving revision 1.26
retrieving revision 1.27
diff -u -d -r1.26 -r1.27
--- transforms.php	30 Jun 2003 10:27:16 -0000	1.26
+++ transforms.php	30 Jun 2003 10:47:01 -0000	1.27
@@ -107,7 +107,7 @@
 {
   global $InterwikiPtn;
 
-  return preg_replace("/(^|[^A-Za-z])($InterwikiPtn)(\$|[^\\/=&~A-Za-z0-9])/e",
+  return preg_replace("/(^|[^A-Za-z])($InterwikiPtn)(?=\$|[^\\/=&~A-Za-z0-9])/e",
                       "q1('\\1').interwiki_token(q1('\\3'),q1('\\4')).q1('\\6')",
                       $text, -1);
 }
@@ -162,6 +162,14 @@
 {
   return preg_replace('/\\[\\[([^] ]+( [^]]+)?)]]/e',
                       "macro_token(q1('\\1'), q1('\\3'))", $text, -1);
+
+}
+
+function parse_no_macros($text)
+{
+  return preg_replace('/\\[\\[([^] ]+( [^]]+)?)]]/e',
+                      "", $text, -1);
+
 }
 
 function parse_no_macros($text)
@@ -289,7 +297,7 @@
 
   if($in_code)
   {
-    if($text == "</code>\n")
+    if(preg_match("|^\s*</code>\s*|", $text))
     {
       $Entity[count($Entity)] = array('code', $buffer);
       $buffer  = '';
@@ -302,7 +310,7 @@
   }
   else
   {
-    if($text == "<code>\n")
+    if(preg_match("|^\s*<code>\s*|", $text))
     {
       $in_code = 1;
       return '';

[Tavi-cvs-commit] CVS: tavi/lib db.php,1.3,1.4 defaults.php,1.5,1.6 pagestore.php,1.5,1.6

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi/lib
In directory sc8-pr-cvs1:/tmp/cvs-serv2396/lib

Modified Files:
	db.php defaults.php pagestore.php 
Log Message:
Merged in release-v0_22-bugfix-branch-1

Index: db.php
===================================================================
RCS file: /cvsroot/tavi/tavi/lib/db.php,v
retrieving revision 1.3
retrieving revision 1.4
diff -u -d -r1.3 -r1.4
--- db.php	30 Nov 2001 22:10:15 -0000	1.3
+++ db.php	30 Jun 2003 10:47:00 -0000	1.4
@@ -28,7 +28,7 @@
     global $ErrorDatabaseQuery;
 
     if(!($qid = mysql_query($text, $this->handle)))
-      { die($ErrorDatabaseQuery); }
+      { die("<strong>$ErrorDatabaseQuery</strong><p>$text</p>"); }
     return $qid;
   }
 

Index: defaults.php
===================================================================
RCS file: /cvsroot/tavi/tavi/lib/defaults.php,v
retrieving revision 1.5
retrieving revision 1.6
diff -u -d -r1.5 -r1.6
--- defaults.php	22 Feb 2002 14:46:08 -0000	1.5
+++ defaults.php	30 Jun 2003 10:47:00 -0000	1.6
@@ -18,8 +18,8 @@
 //   Note that changing this might require a change to parse/transforms.php so
 //   that parse_wikiname knows how many parentheses are included in $LinkPtn.
 $UpperPtn = "[A-Z\xc0-\xde]";
-$LowerPtn = "[a-z\xdf-\xff]";
-$AlphaPtn = "[A-Za-z\xc0-\xff]";
+$LowerPtn = "[a-z\xdf-\xfe]";
+$AlphaPtn = "[A-Za-z\xc0-\xfe]";
 $LinkPtn = $UpperPtn . $AlphaPtn . '*' . $LowerPtn . '+' .
            $UpperPtn . $AlphaPtn . '*(\\/' . $UpperPtn . $AlphaPtn . '*)?';
 

Index: pagestore.php
===================================================================
RCS file: /cvsroot/tavi/tavi/lib/pagestore.php,v
retrieving revision 1.5
retrieving revision 1.6
diff -u -d -r1.5 -r1.6
--- pagestore.php	22 Feb 2002 14:46:08 -0000	1.5
+++ pagestore.php	30 Jun 2003 10:47:00 -0000	1.6
@@ -55,7 +55,7 @@
     $qid = $this->dbh->query("SELECT time, author, version, username, " .
                              "comment " .
                              "FROM $PgTbl WHERE title='$page' " .
-			     "ORDER BY version DESC");
+                 "ORDER BY version DESC");
 
     $list = array();
     while(($result = $this->dbh->result($qid)))
@@ -148,11 +148,14 @@
     {
       $this->dbh->query("UPDATE $IwTbl SET where_defined='$where_defined', " .
                         "url='$url' WHERE prefix='$prefix'");
+      echo "<pre>InterWikiUpd: $prefix, $where_defined, $url</pre>\n";
     }
     else
     {
       $this->dbh->query("INSERT INTO $IwTbl(prefix, where_defined, url) " .
                         "VALUES('$prefix', '$where_defined', '$url')");
+
+      echo "<pre>InterWiki: $prefix, $where_defined, $url</pre>\n";
     }
   }
 

[Tavi-cvs-commit] CVS: tavi ChangeLog.txt,1.7,1.8

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi
In directory sc8-pr-cvs1:/tmp/cvs-serv2396

Modified Files:
	ChangeLog.txt 
Log Message:
Merged in release-v0_22-bugfix-branch-1

Index: ChangeLog.txt
===================================================================
RCS file: /cvsroot/tavi/tavi/ChangeLog.txt,v
retrieving revision 1.7
retrieving revision 1.8
diff -u -d -r1.7 -r1.8
--- ChangeLog.txt	28 Jun 2003 12:14:47 -0000	1.7
+++ ChangeLog.txt	30 Jun 2003 10:47:00 -0000	1.8
@@ -2,12 +2,43 @@
 
 WikkiTikkiTavi - ChangeLog
 --------------------------
+2003-06-30  Even Holen <eve...@pv...>
+  * parse/transforms.php and parse/save.php: Removed fixed regarding
+    parse_clear which removed all text. Inserted parse_no_macros instead
+  * Fixed sisterwiki-definition like interwiki's below
+
+2003-06-28  Even Holen <eve...@pv...>
+  * parse/macros.php:
+    - In WantedPages changed sql-query (having => where) to make it work with
+      mysql 4.0.0alpha (and later on)
+    - Added warning of deletion on empty pages in OrphanedPages-macro
+  * lib/diff.php: Detect if we're running on Windows, and then use rand()
+    instead of posix_getpid().
+  * templates/common.php: Added a div#toolbar_top around the toolbar
+  * templates/wiki.css: Added a display:none for div#toolbar_top on
+    media:print to disable printing of menu. Also added a clear:both to
+    the hr in div#header so that one might right-justify the image in the
+    header without it being below the line.
+  * parse/transforms.php:
+    - Allow white space around <code>/</code>
+    - Added assertion to interwiki-regexp to allow interwiki-links to appear
+      next to each other
+    - Added parse_clear() in addition to parse_noop() to be able to remove
+      text in parse_define_links
+  * parse/save.php:
+    - Allowed escaping InterWiki-entries with !, allowing start with a
+      lowercase character, and allowing https:-links as InterWiki-entries.
+    - Changed parse_noop to parse_clear in parse_define_links to avoid
+      using macro names in the link tables (and thus from appearing on
+      WantedPages if not defined)
+  * lib/db.php: Include error message in error output in query(). NOTE!
+    This should be extended/enhanced for all the db-functions...
+  * lib/defaults: Changed alphapatterns to _not_ include $FlgChr, \xff
 
 2002-02-22  Scott Moonen  <sm...@an...>
    * **  Version 0.22 released **
 
-2002-02-19  Scott Moonen  <sm...@an....
-
+2002-02-19  Scott Moonen  <sm...@an...>
    * Added code to HTML template to add the page's name, with WikiWords
      split apart, to the list of meta keywords for a page.  This should
      aid in search engine categorization of pages.

[Tavi-cvs-commit] CVS: tavi/template common.php,1.8,1.9 wiki.css,1.8,1.9

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi/template
In directory sc8-pr-cvs1:/tmp/cvs-serv2396/template

Modified Files:
	common.php wiki.css 
Log Message:
Merged in release-v0_22-bugfix-branch-1

Index: common.php
===================================================================
RCS file: /cvsroot/tavi/tavi/template/common.php,v
retrieving revision 1.8
retrieving revision 1.9
diff -u -d -r1.8 -r1.9
--- common.php	22 Feb 2002 14:46:09 -0000	1.8
+++ common.php	30 Jun 2003 10:47:01 -0000	1.9
@@ -74,10 +74,13 @@
     print $args['headsufx'];
 ?>
   </h1>
+  <div id="toolbar_top">
 <?php
   if($args['toolbar'])
     { print html_toolbar_top(); }
 ?>
+
+</div>
 <hr />
 </div>
 <?php

Index: wiki.css
===================================================================
RCS file: /cvsroot/tavi/tavi/template/wiki.css,v
retrieving revision 1.8
retrieving revision 1.9
diff -u -d -r1.8 -r1.9
--- wiki.css	15 Jan 2002 16:02:43 -0000	1.8
+++ wiki.css	30 Jun 2003 10:47:01 -0000	1.9
@@ -48,3 +48,12 @@
 pre
   { background: transparent;
     color:      #383838; }
+
+div#header #logo
+  { float: right; }
+div#header hr
+  { clear: both; }
+
+@media print {
+ div#header #toolbar_top { display: none}
+}

[Tavi-cvs-commit] CVS: tavi/parse save.php,1.5.2.1,1.5.2.2 transforms.php,1.25.2.1,1.25.2.2

[Even H. <ho...@us...>]

Update of /cvsroot/tavi/tavi/parse
In directory sc8-pr-cvs1:/tmp/cvs-serv964

Modified Files:
      Tag: release-v0_22-bugfix-branch
	save.php transforms.php 
Log Message:
Something strange with sticky options...

Index: save.php
===================================================================
RCS file: /cvsroot/tavi/tavi/parse/save.php,v
retrieving revision 1.5.2.1
retrieving revision 1.5.2.2
diff -u -d -r1.5.2.1 -r1.5.2.2
--- save.php	29 Jun 2003 18:34:13 -0000	1.5.2.1
+++ save.php	30 Jun 2003 10:36:22 -0000	1.5.2.2
@@ -24,7 +24,7 @@
       { $elements_index = $i; }
   }
   if($macros_index != -1)
-    { $ParseEngine[$macros_index] = 'parse_clear'; }
+    { $ParseEngine[$macros_index] = 'parse_no_macros'; }
   if($transclude_index != -1)
     { $ParseEngine[$transclude_index] = 'parse_noop'; }
   if($elements_index != -1)

Index: transforms.php
===================================================================
RCS file: /cvsroot/tavi/tavi/parse/transforms.php,v
retrieving revision 1.25.2.1
retrieving revision 1.25.2.2
diff -u -d -r1.25.2.1 -r1.25.2.2
--- transforms.php	29 Jun 2003 18:34:13 -0000	1.25.2.1
+++ transforms.php	30 Jun 2003 10:36:22 -0000	1.25.2.2
@@ -8,20 +8,11 @@
 //   converted back into HTML  (or, in the future, perhaps some other
 //   representation such as XML).
 
-
-// The following functions is mainly used in the save process, and are mainly
-// used to avoid parsing of macros, transcludes, ... and erronoeous
-// identification of links within those elements.
 function parse_noop($text)
 {
   return $text;
 }
 
-function parse_clear($text)
-{
-  return "";
-}
-
 // The following function "corrects" for PHP's odd preg_replace behavior.
 // Back-references have backslashes inserted before certain quotes
 // (specifically, whichever quote was used around the backreference); this
@@ -171,6 +162,13 @@
 {
   return preg_replace('/\\[\\[([^] ]+( [^]]+)?)]]/e',
                       "macro_token(q1('\\1'), q1('\\3'))", $text, -1);
+
+}
+
+function parse_no_macros($text)
+{
+  return preg_replace('/\\[\\[([^] ]+( [^]]+)?)]]/e',
+                      "", $text, -1);
 
 }