Hi all!
On Sun, 6 May 2001, Robin wrote:
>I'm going to start a conversion over into using XHTML & add the DTD tags
>myself.
Great :)
>Our offical DTD header block is now
><?xml version="1.0" encoding="UTF-8"?>
Ouch, UTF-8 ? Many browsers won't like this charset. Actually I'm afraid
only
the major ones (IE and Moz. -then maybe NN6-) knows this charset.
Moreover I've faced some problems with this instruction inside php files:
the php
parser try... to parse <?xml.... How do you avoid this?
>The idea I had, there exists a common CSS file, that does NOT comply for
>the NN4 parent bug. The PHP file reads it in, and spits out a new one that
>is complete, rather like the CSS validator by W3. As a result, we only
>have to maintain the common one and one per language, and the PHP file
>spits out a combined one, that could even be customized per browser.
OK.
>Text direction and font face are a function of the character set.
Have you tried this with Moz or NN? As far as I remenber it won't work
with NN4 at least. Not a big problem nevertheless: NN4 does not support
any directive (even HTML/XHTML) for text direction :(
Regards,
______________________________________________________________________________
ifrance.com, l'email gratuit le plus complet de l'Internet !
vos emails depuis un navigateur, en POP3, sur Minitel, sur le WAP...
http://www.ifrance.com/_reloc/email.emailif
Sorry for missing the "{"
I modify it in one directory and test that it work.
Then I check out as another directory, edit it
3 lines to be the same but miss the "{".
, then check in..
I will be more careful (and test with the new
directory)
next time.
Now it work, but not in the same way that Olivier
want.
It only add new row, but not continue for adding more
and more rows. I agree that the change won't be very
complicated but now I am working with my in-place edit
feature. So I will work on cloning later unless
someone
does.
Korakot
__________________________________________________
Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices
http://auctions.yahoo.com/
Hi All!
This is a reply to a message from Korakot but maybe this
thread could be interesting for all developpers.
>I agree with you about 3 more variable
>maybe
>$strCharSet
>$strFontFace
>$strTextDirection
OK. For the $strCharSet we should just specify that it
must be member of a family like 'sans-serif'.
And what about $strFontSize (with a value such as 'x-
medium')?
>We could accept these values from Translation Patch.
>(and we could use them from phpMyChat, right?)
Yes: except the Font face, we could easilly set ourself the
other settings (already got them in phpMyChat).
>>- headers that should be handled by php rather than
>>'meta' tags. These headers are cache instruction and
>>charset definition;
>Please explain into more detail, so that I could
>understand what you mean.
Well, inside the 'header.inc.php3' instead of the exsting
meta tags we should use:
$now = gmdate('D, d M Y H:i:s') . ' GMT';
header('Expires: ' . $now);
header('Last-Modified: ' . $now);
header('Cache-Control: no-store, no-cache, must-revalidate'); // HTTP/1.1
header('Cache-Control: pre-check=0, post-check=0, max-age=0'); // HTTP/1.1
header('Pragma: no-cache'); // HTTP/1.0
// Defines the charset to be used
header('Content-Type: text/html; charset=' . $strCharSet);
This way you don't need to:
1) send headers to the browser...
2) ... that will have to send these values back to the
server...
3) ...so the later may decide what to do.
An other problem is that phpMyAdmin do not send any
DTD at the beginning of generadted HTML pages. This
do not fit the W3C standards and may be the reason of
many problems (at least with IE5 for Mac).
>>Lastly we should use an external stylesheet that is
>>called by each of the scripts (it would be easier to
>>customize phpMyAdmin this way). This stylesheet may
>>be a php file so we can pass et least one parameter to it
>>by url: the font face to use.
>I agree that we should use external style sheet.
>But not sure if making it a php file itself a good
>idea.
How can the take into account the font face else? The
only alternative I can image is... to have one external
stylesheet per language! Not sure it's a very good
solution ;)
>Let's consult the dev-list.
Here we are
>Lastly I'd like to discuss about select_lang.inc.php3
>drop this line.
>if (!function_exists('pmaLangDetect')) {
I've added it because under certain circumstances (can't
remenber when, unfortunatly) the 'select_lang.inc.php3'
script is called twice during the parsing related to one
action.
>rename pmaLangDetect to lang_detect
>drop $notFound and use break to go out of while loop
>use variable $http_lang or something short to store
>intermediate value... make it easier to read code.
Ok, not a problem. We just need to clearly define the
coding standards we have to use before I made this
changes.
>In general, the old code is much easier to read and
>understand.
Right but the former have some annoying limits:
1) it was really difficult to choose between dialects. This
is not really a problem for 'fr' and 'fr-ch' for example (sorry Olivier ;)),
but it is one for 'zh and zh-tw' as I've been told, or for different spanish
dialects (same words
with different meanings, or words specific to one dialect
...);
2) the language detection I suggest in the current version
is the result of a digest of about 300 different HTTP_ACCEPT_LANGUAGE and
HTTP_USER_AGENT headers
sent by browsers. Some fits to the RFC specifications,
other not (or not exactly), but each of these 300 headers are taken into
account. I'm afraid the "old" version don't work with all
of them ;)
BTW, this reflect only my state of mind of course.
Regards,
______________________________________________________________________________
ifrance.com, l'email gratuit le plus complet de l'Internet !
vos emails depuis un navigateur, en POP3, sur Minitel, sur le WAP...
http://www.ifrance.com/_reloc/email.emailif
I've completed this feature now.
It should work ok, i've put it in both trees as it's a minor feature
really.
Robin Hugh Johnson
"Robbat2"
QTOD: "I used to be an idealist, but I got mugged by reality."
E-Mail : robbat2(a)orbis-terrarum.net
ICQ# : 30269588 or 41961639
Home Page : http://www.orbis-terrarum.net
Time Zone : Pacific Daylight (GMT - 8)
-----GEEK CODE-{---
Version: 3.12 Serial: 2001041400
GU/CS d- a--- C++++ L++++ U*+++ E----
e* h! tv-- X+ W+++ N+++ w--- M P+ R O-
D++ V-- PS+ PE-- Y+ PGP++ r !y t-- 5
s+:- b+++
---}-GEEK CODE-----
-----PGP INFO-{---
Key ID:0x7E20DFA1
FingerPrint:
5447C73A 30FB144C 89521B69 2D6A615E 7E20DFA1
---}-PGP INFO-----
I was wondering, are there any easy methods that we can use to keep our
two trees in sync? As it's a pain to write code for the stable tree and
then have to integrate it all over again with the devel tree.
Robin Hugh Johnson
"Robbat2"
QTOD: "I used to be an idealist, but I got mugged by reality."
E-Mail : robbat2(a)orbis-terrarum.net
ICQ# : 30269588 or 41961639
Home Page : http://www.orbis-terrarum.net
Time Zone : Pacific Daylight (GMT - 8)
-----GEEK CODE-{---
Version: 3.12 Serial: 2001041400
GU/CS d- a--- C++++ L++++ U*+++ E----
e* h! tv-- X+ W+++ N+++ w--- M P+ R O-
D++ V-- PS+ PE-- Y+ PGP++ r !y t-- 5
s+:- b+++
---}-GEEK CODE-----
-----PGP INFO-{---
Key ID:0x7E20DFA1
FingerPrint:
5447C73A 30FB144C 89521B69 2D6A615E 7E20DFA1
---}-PGP INFO-----
I've been looking a bit at all the languages files, and I see that the
existance of variables is horribly out of sync in some of them, yet others
have strings that are commented out, that I can't figure out why.
I'm going to try and put together a list of what variables there are,
which language files they exist in, and if they need to be translated or
not still.
Robin Hugh Johnson
"Robbat2"
QTOD: "I used to be an idealist, but I got mugged by reality."
E-Mail : robbat2(a)orbis-terrarum.net
ICQ# : 30269588 or 41961639
Home Page : http://www.orbis-terrarum.net
Time Zone : Pacific Daylight (GMT - 8)
-----GEEK CODE-{---
Version: 3.12 Serial: 2001041400
GU/CS d- a--- C++++ L++++ U*+++ E----
e* h! tv-- X+ W+++ N+++ w--- M P+ R O-
D++ V-- PS+ PE-- Y+ PGP++ r !y t-- 5
s+:- b+++
---}-GEEK CODE-----
-----PGP INFO-{---
Key ID:0x7E20DFA1
FingerPrint:
5447C73A 30FB144C 89521B69 2D6A615E 7E20DFA1
---}-PGP INFO-----
[about_me]
My name is Korakot Chaovavanich, age 24 (about average
:> ).
I am working as an assistant researcher at a
government
organization (NECTEC- www.nectec.or.th) in the Open
Source
Project to promote the use/development of Open Source
Software in Thailand. (best suit for for-fun
programmer like me)
I have about 2-year experience in Perl, PHP/MySql,
4-year in
HTML and Javascript. When I first met phpMyAdmin, I
found it
to be the essensial tool I use almost all the time.
Now I wish it
easier to use than Access, even with it web-interface
limitation.
So comes my feature request (XXXXX), please check it
initial
implimentation (just upload).. I guess you all must
like it.
I can speak only in english and thai(th), so please
use only
english (7-bin ascii :>)... Now I don't know how to
spell or
pronounce names of our 2 project admins...
Oliver/Looc ?
My website ? not update for 3 years...
My resume? only in thai, sorry..
My picture... ok.. in 1998, still quite cute
http://geocities.com/korakot2/grad/stand.JPG
This is actually my first time as developer in an
international
team of OSS. But I have help submit patch for
sourceforce i18n
(language auto-detect and message translation). I also
convert
mysql help to MS htmlhelp format. (but they don't
accept it
to be distributed on their mysql website)
[mysql_dev]
Forgive me if I put too many issues in one single
email.
I suggest that we have an agreement on development
and set up a page for development guideline
(on phpmyadmin.sourceforge.net or on project doc page)
Please see a good example in SourceForge doc page.
Now there are too many channels to discuss the
development
discussion. Should a feature be discuss in Feature
Tracker
or Developer Mailinglist or Forum or personal(1-1)
email?
We could discuss it in mailing list first and post
some
summary in Feature Tracker.
My 2 examples are i18n and permission-checking.
Looc discuss i18n with me using 1-1 mail (not respond
yet)
and now he has commit it to CVS. I think we should
discuss first on dev-mail or forum before commit.
According to XP (extreme programming), we should
have 2 developers looking at the code togeter before
commit. (<looc> sorry for not promptly respone mail)
I have also suggested a guideline for interface design
in the Feature Request. (maybe a wrong place)
This guide (checking for action permission before
showing action link/button) could result in many
features (so I submit it there). But I could have
sent it here instead.
Other issues (guideline for development) that we must
decide are
- should we go for php4 or still 100% support php3?
- By permission checking, we could check the mysql
version and decide to provide some features or not
based on version number (eg. Full-text search on
3.23.3?+ on MYISAM table type)
- translation could be done with 3 approach, which one
should we use (define(), array, or gettext) ?
- browser support and javascript. (eg. my inplace-edit
feature using DOM scripting) We must support IE,
Moz,Konq. Should we try support NN4/Opera DHTML
as well?
Some question on currest source in CVS
- I see the config file has change the config vars of
user/password/onlydb. I am afraid this would affect
the use of new user who could once drop and run
phpmyadmin without editing anything (user=root,
no-password/no-onlydb).
- The language detection code is still not pretty.
The variable convention such as $notFound and
function pmaLangDetect is not consistent with
the rest of php. This is one reason I suggest we
have a coding guideline (like sourceforge's)
Finally, Thank you so much for bring the phpmyadmin
development to sourceforge. I plan to contribute more
feature request :> , ok.. more code as well.
And don't forget to check my in-place edit code in
feature request page.
Korakot Chaovavanich
__________________________________________________
Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices
http://auctions.yahoo.com/
Hi All!
>BTW do we have "official" translators for every
>language?
I don't think so.
>So a lag will develop, and it will be clean to see the new
>strings at the end of file.
Good idea :). An alternative would be to add a comment
at the end of the strings that needs to be translated (let's
say '// to translate' for example): this way alphabetical
order can be kept and it would easy for translators to find
what they have to work on.
Just my 2 cents...
BTW have you tried to upload some file to both the
current devel. versions. Maybe I'm wrong but it seems not
to work at all (even if a success message is displayed). I'll
try to investigate on this today.
Have a nice day/night,
Loïc
______________________________________________________________________________
ifrance.com, l'email gratuit le plus complet de l'Internet !
vos emails depuis un navigateur, en POP3, sur Minitel, sur le WAP...
http://www.ifrance.com/_reloc/email.emailif
Hi again :)
the $cfgShowBlob is in the config file, but sometimes we want to see the
blobs while browsing, sometimes not.
I think a button (on the showing records page) would be useful for
this. I would prefer to make this a session variable, but since we have
to support php3, this would become another get parameter?
Take the default from config file, but let them change it...
Marc (Lem9)
