From typo3 at rvt.dds.nl Sat Oct 1 02:00:37 2005 From: typo3 at rvt.dds.nl (Ries van Twisk) Date: Fri, 30 Sep 2005 18:00:37 -0600 Subject: [Typo3-dev] Wildcard match on globalString Message-ID: Dear Developers, am I correct that wildcardMatch is not completely supported as read from tsRef? When reading it can do this on loginUser [loginUser = *] [end] But when I try to do this: [globalString = TSFE:page|tx_rvtgbweathericaotag_icao = *] [end] The routine: function matchWild($haystack,$needle) {...} Doesn't do a wildcard match on just *, I would suspect (I might be wrong) that it would match any empty string there. Instead this case condition get's triggered and does a exact match. case 'beforeafter': if (strstr($haystack,substr($needle,1,-1))) return true; break; What can I do to do a wildcard on globalString (match a non empty string)? except writing a user function..... cheers, Ries van Twisk From stanislas.rolland at fructifor.ca Sat Oct 1 01:36:33 2005 From: stanislas.rolland at fructifor.ca (Stanislas Rolland) Date: Fri, 30 Sep 2005 19:36:33 -0400 Subject: [Typo3-dev] frontend editing In-Reply-To: References:

Message-ID: Ernesto Baschny [cron IT] a ?crit : > > Yeah, it looks very cool! The technology behind it is just the same as > we already have in the original RTE that comes with TYPO3 core. It uses > document.execCommand, which is IE-specific, which is why it won't work > on any other platform (just like "rte" in TYPO3). > execCommand is also supported by gecko browsers such as Mozilla, Firefox and Netscape 6+ as well as recent release of Safari (although incomplete and not yet usable in Safari), which makes htmlArea RTE and other cross-browser editors possible. > The only difference in TYPO3 RTE is that the edit area always has a > "backend" look (white background, 1px black border, always the same > width/height), while in editland the RTE is styled the same way as the > site (and without a border), so you get a feeling as if you are working > directly on the site. > You could skin htmlArea RTE to your liking. Regards, Stanislas From tobi at 54tec.com Sat Oct 1 10:24:02 2005 From: tobi at 54tec.com (tobias erlacher) Date: Sat, 01 Oct 2005 10:24:02 +0200 Subject: [Typo3-dev] Debugging Typo3/PHP 2 (mini review) References:

Message-ID: Hi also maguma studio and maguma workbench have xdebug support Am Thu, 29 Sep 2005 18:00:37 +0200 schrieb Andreas Otto : > Hi Stig, > > On Donnerstag, 29. September 2005 16:46, Stig N?rgaard F?rch wrote in > typo3.dev: > >> But what does it offer me that DBG doesn't? > > I just like it more ;-) Especially the profiling and stacktraces. > I have not compared these two debuggers. > >> I can't find any free editors that works with Xdebug. Are there any?` > > Any editor that supports the dbg / dgbp protocol should work AFAIK. > > Quanta+ in KDE 3.5 will have support for xdebug. > > > Cheers, > Andreas > -- Erstellt mit Operas revolution?rem E-Mail-Modul: http://www.opera.com/m2/ From kraftb at kraftb.at Sat Oct 1 15:17:12 2005 From: kraftb at kraftb.at (Bernhard Kraft) Date: Sat, 01 Oct 2005 15:17:12 +0200 Subject: [Typo3-dev] RFC (core patch): flexform field path Message-ID: Hallo, I run into a obvious "bug" or more a "non-feature" of T3 Flexforms. If you use an for generating the elements of an select field and those items shall depend on the value of another field in the flexform then you get the problem that the doesn't know which flexform-field it renders actually. Of course you have the name of the field - but if you use a flexform with sections (variable number of sub-elements) then you can't know in which section you are actually. I made two patches to correct this. The first one is simpler but you will have to parse the passed "formPrefix" string in your itemsProcFunc to find out the path of the field in the flexform. The second one keeps track of the actual path in the flexform as an array and uses this as argument to the itemsProcFunc. The pathes are here: 1: http://www.think-open.org/kraftb/01_flexField_path_patch.diff 2: http://www.think-open.org/kraftb/02_flexField_path_patch.diff Could somebody please take a look at those two and tell me which one is more safe to implement into the core concerning downwards compatiblility etc. greets, Bernhard From elmar.hinz at vcd-berlin.de Sat Oct 1 23:54:53 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Sat, 01 Oct 2005 23:54:53 +0200 Subject: [Typo3-dev] frontend editing In-Reply-To: References:

Message-ID: >> > execCommand is also supported by gecko browsers such as Mozilla, Firefox > and Netscape 6+ as well as recent release of Safari (although incomplete > and not yet usable in Safari), which makes htmlArea RTE and other > cross-browser editors possible. > >> > You could skin htmlArea RTE to your liking. > > Regards, > Stanislas I hope htmlArea is getting the default RTE of TYPO3 as soon as possible. It's allways annoying when the default configuration forces people to I.E. That should be history. Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From niederlag at ikd01.de Sun Oct 2 07:33:11 2005 From: niederlag at ikd01.de (Peter Niederlag) Date: Sun, 02 Oct 2005 07:33:11 +0200 Subject: [Typo3-dev] frontend editing In-Reply-To: References:

Message-ID: Hi, Elmar Hinz schrieb: [...] > I hope htmlArea is getting the default RTE of TYPO3 as soon as possible. > It's allways annoying when the default configuration forces people to > I.E. That should be history. Already on Roadmap "ZAP the Gremlins". ;) http://typo3.org/development/roadmap/ p. -- Peter Niederlag http://www.niekom.de * TYPO3 & EDV Dienstleistungen * http://www.typo3partner.net * professional services network * From steffen at mail.kommwiss.fu-berlin.de Sun Oct 2 15:21:12 2005 From: steffen at mail.kommwiss.fu-berlin.de (=?ISO-8859-1?Q?Steffen_M=FCller?=) Date: Sun, 02 Oct 2005 15:21:12 +0200 Subject: [Typo3-dev] frontend editing In-Reply-To: References:

Message-ID: Hi. On 01.10.2005 23:54 Elmar Hinz wrote: >>> > > > I hope htmlArea is getting the default RTE of TYPO3 as soon as possible. > It's allways annoying when the default configuration forces people to > I.E. That should be history. > In my eyes, it is also annoying that standard BE users can't choose their favorite RTE. Only the admin is able to choose a system wide RTE. If you have the same opinion, please vote for bug/feature 878: http://bugs.typo3.org/view.php?id=878 Or even better, help fixing it ;-) -- cheers Steffen From martin.kutschker-n0spam at no5pam-blackbox.net Sun Oct 2 19:00:27 2005 From: martin.kutschker-n0spam at no5pam-blackbox.net (Martin Kutschker) Date: Sun, 02 Oct 2005 19:00:27 +0200 Subject: [Typo3-dev] frontend editing In-Reply-To: References:

Message-ID: Steffen M?ller schrieb: > Hi. > > On 01.10.2005 23:54 Elmar Hinz wrote: > >> >>I hope htmlArea is getting the default RTE of TYPO3 as soon as possible. >>It's allways annoying when the default configuration forces people to >>I.E. That should be history. >> > > > In my eyes, it is also annoying that standard BE users can't choose > their favorite RTE. Only the admin is able to choose a system wide RTE. > > If you have the same opinion, please vote for bug/feature 878: > http://bugs.typo3.org/view.php?id=878 > > Or even better, help fixing it ;-) I don't want my users to choose the RTE. RTEs are hard to configure and don't have the same feature set. Masi From murphy at spreekonzept.de Mon Oct 3 06:58:54 2005 From: murphy at spreekonzept.de (Thomas Murphy) Date: Sun, 02 Oct 2005 21:58:54 -0700 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB Message-ID: Hello everyone, I'm using adodb in one of my extensions to access data from a foreign mySQL-DB. After I successfully connected to it, it seems the Connection to the TYPO3_DB is lost - it then tries to connect to the TYPO3_DB with the data (username/password) I used for the foreign mySQL-DB. (!) ("Access denied for user...") I made some dirty hotfix for this by manually reconnecting to the TYPO3_DB with the username/password for that DB after my plugin has finished. I checked php.ini settings for maximum persistent and normal connections, but they are all set to "-1" (unlimited). Any other ideas? bye, Thomas From elmar.hinz at vcd-berlin.de Sun Oct 2 22:29:55 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Sun, 02 Oct 2005 22:29:55 +0200 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB In-Reply-To: References: Message-ID: Thomas Murphy wrote: > Hello everyone, > > I'm using adodb in one of my extensions to access data from a foreign > mySQL-DB. After I successfully connected to it, it seems the Connection > to the TYPO3_DB is lost - it then tries to connect to the TYPO3_DB with > the data (username/password) I used for the foreign mySQL-DB. (!) > ("Access denied for user...") > > I made some dirty hotfix for this by manually reconnecting to the > TYPO3_DB with the username/password for that DB after my plugin has > finished. > I checked php.ini settings for maximum persistent and normal > connections, but they are all set to "-1" (unlimited). Hi Thomas, how are you? I don't have the feeling that it has anything to do with the count of possible connections. Rather with both libraries trying to use the same "default connection". I don't know if those libraries are using the function mysql_connect but I think http://php.speedbone.de/manual/en/function.mysql-connect.php could give you at least some clues of how connections fall back to defaults if not all parameters are given. Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From typo3 at rvt.dds.nl Mon Oct 3 01:42:28 2005 From: typo3 at rvt.dds.nl (Ries van Twisk) Date: Sun, 02 Oct 2005 17:42:28 -0600 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB In-Reply-To: References: Message-ID: Thomas Murphy wrote: >Hello everyone, > >I'm using adodb in one of my extensions to access data from a foreign >mySQL-DB. After I successfully connected to it, it seems the Connection >to the TYPO3_DB is lost - it then tries to connect to the TYPO3_DB with >the data (username/password) I used for the foreign mySQL-DB. (!) >("Access denied for user...") > >I made some dirty hotfix for this by manually reconnecting to the >TYPO3_DB with the username/password for that DB after my plugin has >finished. >I checked php.ini settings for maximum persistent and normal >connections, but they are all set to "-1" (unlimited). > >Any other ideas? > > > I made a little extention called rvt_adodb that you can use. At least on my site I heaven't seen such a problem. Ries From typo3 at fm-world.ru Mon Oct 3 07:37:17 2005 From: typo3 at fm-world.ru (Dmitry Dulepov) Date: Mon, 03 Oct 2005 08:37:17 +0300 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB In-Reply-To: References: Message-ID: Hi! Thomas Murphy wrote: > I'm using adodb in one of my extensions to access data from a foreign > mySQL-DB. After I successfully connected to it, it seems the Connection > to the TYPO3_DB is lost - it then tries to connect to the TYPO3_DB with > the data (username/password) I used for the foreign mySQL-DB. (!) > ("Access denied for user...") This is not related to ADODB but rather to usage of the obsolete PHP function named "mysql()" in Typo3 (probably, not in typo3 core but in extensions). This function will use the last opened connection. As soon as you connect to the external database, everything stops working... I fall into the same problem two moths ago (asked about it here too). The only workaround is to re-open connection to the typo3 database. Dmitry. From triphot69 at hotmail.com Mon Oct 3 10:23:17 2005 From: triphot69 at hotmail.com (Jean-Baptiste Rio) Date: Mon, 03 Oct 2005 10:23:17 +0200 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB In-Reply-To: References:

Message-ID: Ries van Twisk wrote: > I made a little extention called rvt_adodb that you can use. > At least on my site I heaven't seen such a problem. > > Ries Hi Ries, I've tried to use your extension. Seems to work fine, except that there is something strange. We had at the same time, for others reasons, a very low performance on our server and specially on the mysql process (nearly 80% of the CPU), so we've logged all the MySQL process running at a time. I've just installed your extension and configured the access to an external database but without using it (i was planning to use it for a new site). But your extension logged several mysql process at a time. I've de-installed your extension, and the process disappeared. As if each time we asked to the Typo3 database, Typo3 tried to connect to the external database... weird isn't it ? Just your information. Seems i'm the only one to have seen that. So... maybe an error from my installation (Running Typo3 3.7.0 and PHP5.0). Regards, Jean-Baptiste From triphot69 at hotmail.com Mon Oct 3 12:34:08 2005 From: triphot69 at hotmail.com (Jean-Baptiste Rio) Date: Mon, 03 Oct 2005 12:34:08 +0200 Subject: [Typo3-dev] excludeUidList : Exclude current page from a menu Message-ID: Before we have, maybe one day, a full stdWrap on this menu param, i suggest a little enhancement which allows to exclude the current page from a menu (typically when using special with update or directory). This enhancement accept "this" in the uid list. For example : lib.mymenu = HMENU lib.mymenu.entryLevel = -2 lib.mymenu.excludeUidList = this lib.mymenu.1 = TMENU ... This produces a menu with the "sisters and brothers" pages from a page (so except the page itself). I've changed these codelines in class.tslib_menu.php (at 2 positions : line 763 and line 2613) if (trim($this->conf['excludeUidList'])) { - $banUidArray = t3lib_div::intExplode(',', $this->conf['excludeUidList']); + $banUidList = str_replace('this', $GLOBALS['TSFE']->page['uid'], $this->conf['excludeUidList']); + $banUidArray = t3lib_div::intExplode(',', $banUidList); } I use Typo3 3.7.0 and PHP5. I sent a feature issue in the bugtracking to trace this enhancement. Regards, Jean-Baptiste From murphy at spreekonzept.de Tue Oct 4 00:01:57 2005 From: murphy at spreekonzept.de (Thomas Murphy) Date: Mon, 03 Oct 2005 15:01:57 -0700 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB In-Reply-To: References:

Message-ID: Dmitry Dulepov wrote: > This is not related to ADODB but rather to usage of the obsolete PHP > function named "mysql()" in Typo3 (probably, not in typo3 core but in > extensions). This function will use the last opened connection. As soon > as you connect to the external database, everything stops working... > > I fall into the same problem two moths ago (asked about it here too). > The only workaround is to re-open connection to the typo3 database. Thank you Dmitry. AFAI can recall the extensions failed have been sr_languagemenu and sr_static_info. And they seemed to be loaded after my extension. So I have to stick with reconnecting. bye, Thomas From murphy at spreekonzept.de Tue Oct 4 00:07:26 2005 From: murphy at spreekonzept.de (Thomas Murphy) Date: Mon, 03 Oct 2005 15:07:26 -0700 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB In-Reply-To: References:

Message-ID: Thomas Murphy wrote: > Dmitry Dulepov wrote: > >> I fall into the same problem two moths ago (asked about it here too). >> The only workaround is to re-open connection to the typo3 database. > > So I have to stick with reconnecting. btw.: Is there a "clean way" to reconnect? Right now I just hardcoded my username/password into the connection statement, because AFAIK the connection settings are not part of the TYPO3_CONF_VARS. Can I somehow access the username/password for the DB from my extension scope? bye, Thomas From typo3 at fm-world.ru Mon Oct 3 15:31:20 2005 From: typo3 at fm-world.ru (Dmitry Dulepov) Date: Mon, 03 Oct 2005 16:31:20 +0300 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB In-Reply-To: References:

Message-ID: Hi! Thomas Murphy wrote: > btw.: Is there a "clean way" to reconnect? Right now I just hardcoded my > username/password into the connection statement, because AFAIK the > connection settings are not part of the TYPO3_CONF_VARS. > Can I somehow access the username/password for the DB from my extension > scope? > There are variables in typo3conf/localconf.php: ===================== $typo_db_username = '...; // Modified or inserted by TYPO3 Install Tool. $typo_db_password = '...'; // Modified or inserted by TYPO3 Install Tool. $typo_db_host = 'localhost'; // Modified or inserted by TYPO3 Install Tool. $typo_db = '...'; // Modified or inserted by TYPO3 Install Tool. ===================== You can define they as 'global' or use $GLOBALS['...'] and reconnect. Dmitry. From typo3 at monosock.org Mon Oct 3 16:53:41 2005 From: typo3 at monosock.org (Michiel Roos) Date: Mon, 03 Oct 2005 16:53:41 +0200 Subject: [Typo3-dev] [Typo3-tt-news] category tree for kickstarter? References: Message-ID: On Wed, 28 Sep 2005 14:08:44 +0200, Elmar Hinz wrote: > * How big do you estimate the effort to adapt the class for an own > extension? 2 hours getting to grips with the concept and adjusting stuff in the class file. I adapted the tt_news class file for use with the 'photoblog' extension. It took me some time to get it right. I also stripped out a lot of tt_news specific code. Take a peak in the photoblog source and have a look at the file: class.tx_photoblog_treeview.php > * Do you share my opinion that a catagory tree selector should be > general available to use in BE-Forms? In example as an extension. Could be nice as an extension. > * Does such an extension already exist? I don't think so, I didn't search for it though. > * Else is it planned? > * Should we take further steps to reach this target? Implement ;-) Michiel Roos From elmar.hinz at vcd-berlin.de Mon Oct 3 17:13:13 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Mon, 03 Oct 2005 17:13:13 +0200 Subject: [Typo3-dev] Re: [Typo3-tt-news] category tree for kickstarter? In-Reply-To: References:

Message-ID: Hallo Michiel, thank you for your optimistic advice. I will give it a try so. >>* Do you share my opinion that a catagory tree selector should be >>general available to use in BE-Forms? In example as an extension. > > > Could be nice as an extension. > > [...] >>* Else is it planned? >>* Should we take further steps to reach this target? > > > Implement ;-) When I find myself adapting it for the third extension I will probalbly do so. Currently other things are still on top of my personal TYPO3 agenda. Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From news at monosock.org Mon Oct 3 17:35:19 2005 From: news at monosock.org (Michiel Roos) Date: Mon, 03 Oct 2005 15:35:19 +0000 Subject: [Typo3-dev] [Typo3-tt-news] category tree for kickstarter? References:

Message-ID: One more thing that took me a while to figure out. The treeview seems to have been written for the pages table. So the treeview needs a 'title' field. I could not find a way around this. The table structure I now use for the category table is: `uid` int(11) unsigned NOT NULL auto_increment, `pid` int(11) unsigned NOT NULL default '0', `tstamp` int(11) unsigned NOT NULL default '0', `crdate` int(11) unsigned NOT NULL default '0', `cruser_id` int(11) unsigned NOT NULL default '0', `deleted` tinyint(4) unsigned NOT NULL default '0', `hidden` tinyint(4) unsigned NOT NULL default '0', `title` varchar(60) NOT NULL default '', `parent_category` int(11) unsigned NOT NULL default '0', You can add anything you like, like 'description' or something else. Good luck. Michiel From elmar.hinz at vcd-berlin.de Mon Oct 3 17:56:54 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Mon, 03 Oct 2005 17:56:54 +0200 Subject: [Typo3-dev] [Typo3-tt-news] category tree for kickstarter? In-Reply-To: References:

Message-ID: Michiel Roos wrote: > One more thing that took me a while to figure out. The treeview seems to > have been written for the pages table. So the treeview needs a 'title' > field. I could not find a way around this. The table structure I now use > for the category table is: > Thank you. It is no problem to call the name of the category title. I guess that's the usual habit. Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From jerome at jerome.net Mon Oct 3 18:05:41 2005 From: jerome at jerome.net (Jerome) Date: Mon, 3 Oct 2005 18:05:41 +0200 Subject: [Typo3-dev] Page shortcuts & FE editing Message-ID: Hello I have a problem : when I FE edit my typo3 pages, the ones declared as Shortcuts can not be displayed, and then not be modified ( especially their headers ) Does anybody has a tip for me ? J?r?me Schneider Typo3 Ameos http://www.ameos.com typo3dev at ameos.com From typo3 at rvt.dds.nl Mon Oct 3 19:44:47 2005 From: typo3 at rvt.dds.nl (Ries van Twisk) Date: Mon, 03 Oct 2005 11:44:47 -0600 Subject: [Typo3-dev] adodb - losing connection to TYPO3_DB In-Reply-To: References:

Message-ID: Jean-Baptiste Rio wrote: >Ries van Twisk wrote: > > >>I made a little extention called rvt_adodb that you can use. >>At least on my site I heaven't seen such a problem. >> >>Ries >> >> > >Hi Ries, > > >I've tried to use your extension. Seems to work fine, except that there >is something strange. We had at the same time, for others reasons, a >very low performance on our server and specially on the mysql process >(nearly 80% of the CPU), so we've logged all the MySQL process running >at a time. I've just installed your extension and configured the access >to an external database but without using it (i was planning to use it >for a new site). But your extension logged several mysql process at a >time. I've de-installed your extension, and the process disappeared. As >if each time we asked to the Typo3 database, Typo3 tried to connect to >the external database... weird isn't it ? > >Just your information. Seems i'm the only one to have seen that. So... >maybe an error from my installation (Running Typo3 3.7.0 and PHP5.0). > >Regards, > >Jean-Baptiste >_______________________________________________ > > Jean, this is indeed really weird... I'll go and take a look at this properly this week and see what I come up with. Thanks you for informing about this. regards, Ries van twisk From news at monosock.org Mon Oct 3 18:45:35 2005 From: news at monosock.org (Michiel Roos) Date: Mon, 03 Oct 2005 16:45:35 +0000 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension Message-ID: I would like the title and descriptions of my photoblog extension to be indexed. But I can't find anything on the subject. Maybe I'm looking in the wrong places? Any pointers? Cheers, Michiel Roos From elmar.hinz at vcd-berlin.de Mon Oct 3 19:39:25 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Mon, 03 Oct 2005 19:39:25 +0200 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension In-Reply-To: References: Message-ID: Michiel Roos wrote: > I would like the title and descriptions of my photoblog extension to be > indexed. > > But I can't find anything on the subject. Maybe I'm looking in the wrong > places? > > Any pointers? > > Cheers, > > > Michiel Roos It's the good old place. Best information I found here: http://typo3.org/documentation/document-library/doc_core_tsref/PHP_SCRIPT_INT/ The said counts for USER_INT in the same manner. That means, if title and description are output of the plugin, they can't be cached if you don't user USER instead of USER_INT. If same parameters produce same output in any case you should use USER. You should read http://typo3.org/development/articles/the-mysteries-of-chash/ then, too. That article is quite good, but it has forgotten to explain how you make a link use cHash. See TSref for typolink, so! If your output is really dynamic, that means if the same parameters may produce different output under circumstances, then you have to use USER_INT and you can't cache anything of the output. Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From pilot at m-lan.ru Mon Oct 3 22:19:06 2005 From: pilot at m-lan.ru (Pilot) Date: Tue, 4 Oct 2005 00:19:06 +0400 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension References:

Message-ID: it's strange, that nobody wrote search spider for typo3's pages. it's not so hard, becouse there aer a lot of feaches, for example, such string in HTML: : "Elmar Hinz" ???????/???????? ? ???????? ?????????: news:mailman.1.1128361352.7858.typo3-dev at lists.netfielders.de... > Michiel Roos wrote: >> I would like the title and descriptions of my photoblog extension to be >> indexed. But I can't find anything on the subject. Maybe I'm looking in >> the wrong >> places? >> >> Any pointers? >> >> Cheers, >> >> >> Michiel Roos > > It's the good old place. Best information I found here: > > http://typo3.org/documentation/document-library/doc_core_tsref/PHP_SCRIPT_INT/ > > The said counts for USER_INT in the same manner. That means, if title and > description are output of the plugin, they can't be cached if you don't > user USER instead of USER_INT. > > If same parameters produce same output in any case you should use USER. > You should read > http://typo3.org/development/articles/the-mysteries-of-chash/ then, too. > That article is quite good, but it has forgotten to explain how you make a > link use cHash. See TSref for typolink, so! > > If your output is really dynamic, that means if the same parameters may > produce different output under circumstances, then you have to use > USER_INT and you can't cache anything of the output. > > Regards > > Elmar > > > > > > > > > > > > > > > > > -- > Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, > Austria, Swiss, France, ... > Production of CO2 is killing people. > Production of CO2 just for fun is killing people just for fun. From erik at linnearad.no Mon Oct 3 22:50:52 2005 From: erik at linnearad.no (Erik Svendsen) Date: Mon, 3 Oct 2005 20:50:52 +0000 (UTC) Subject: [Typo3-dev] Some questions, problems and point of views about xml_parser_create Message-ID: First of all. It looks like this is a problem only with PHP 5. And it looks like the problem has somthing to do with the PHP function xml_parser_create (but I don't know enough about TYPO3 and PHP to be exactly sure). I have the problem two with two different functions (extensions), impexp and templavoila. But as far as I know the problem could be a problem everywhere XML parsing is used. The problem both places is that characters as ? ? ? (norwegian), ? ? ? etc. make the parser to crash (an error). I have made an previous post about the problem news://news.netfielders.de/mailman.1.1128075811.24776.typo3-english at lists.netfielders.de. The problem is also described in http://bugs.typo3.org/view.php?id=497 --- Copy of my earlier post --- If I use non-US characters, ex. ?, ? ? ? ? etc, in title, mapping instructions or sample data (in the Building data Structure page) I got the following error message when trying to look at the XML: The input content failed XML parsing: Line xx: Invalid character. The reference to the DS is also broken if I save the DS. The simple solution is not using any non-US characters. I don't have the same problem on a installation with PHP 4.4 (TYPO3 3.8.0, TV 4.0) I suppose the problem isn't TYPO3, but the XML parsing in PHP. But I make a post, both to give a solution to others who get the same problem, and to get some responses from other that have had/has the same problem. And maybe some who are more skilled in TYPO3 and PHP than me, can find the real solutions (and if necessary, post a bug at PHP.net). But I also think this is a problem that should be solved. It's more userfriendly to have titles in native language. I make the post both in typo3.english and typo3.dev. Responses should be given in typo3.dev. My configurations: Server 1: Fedora Core 4 X64, Apache 2.0.54, PHP 5.0.4, MySQL 4.1.12, eaccelerator 0.9.3. Server 2: Windows XP SP2, Apache 2.0.54, PHP 5.0.5, MySQL 5.0.12-beta-nt, eaccelerator 0.9.3. Server 3: Debian, Apache 1.3.33, PHP 4.4.0, MySQL 4.1.14 (no problem with non-US characters) --- end of copy --- You also got a problem if you are using FCE with TV. The editors can't use these norwegian characters in plain textfield. And it's a problem when you are making Norwegian websites (and a lot of other European languages). The solution in bug 497 do not help. To learn a little more about the problem and the XML parsing i made a small parsing (xml_parser_create) script with help of the PHP documentation. What I found was that every special character in norwegian, german, swedish gave an error if no charset encoding, uttf-8 and utf-16 was used in the xml-file. If charset iso-8859-1 was used, the parsing went okay. What I learn here gave me an idea that my problem with impexp had the same reason (using TV). The exported xml-file had the encoding iso-8859-1, and went straight trough my parser-script. But I couldnt import it correct. I then search and replaced ? with aa, ? with oe and ? with ae in the exported xml-file. And the import went smooth. I think i maybe only neccessary to change the part of the file who is refering to TV. I shall test it. I don't know if this problem is impexp or TV related. This gave me the idea (probably wrong) that the parserscript in impexp don't use the encoding in the xml-file. Somebody who knows? I have the understanding that TV is using utf-8. Is there any way to change this? As I see it, there are workaround for most of these problems, except the TCE problem. The easiest workaround is not using these special characters, but it's not the best solutions for the users. Another solution is maybe to force impexp and TV to use a spesific charset if needed. Someone maybe have an solution to this problem. It was a long post, but I think this is a problem that should be solved somehow. WBR, Erik Svendsen www.linnearad.no From elmar.hinz at vcd-berlin.de Mon Oct 3 23:37:52 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Mon, 03 Oct 2005 23:37:52 +0200 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension In-Reply-To: References:

Message-ID: Pilot wrote: > it's strange, that nobody wrote search spider for typo3's pages. > > it's not so hard, becouse there aer a lot of feaches, for example, such > string in HTML: > > : > Interesting question. So, what does a spider do? Crawling websides and storing information about them into an quickly searchable database. The point is, that the information of a typo3 website already is in such a database. So probably nobody came to the idea of storing them a second time into the same database. But as that special point, where you have a really dynamic plugin output but with some special static parts in it, spidering could make sense. Example? Your extension outputs realtime prices of shares. You have to use USER_INT because the prices change dynamically. But it could be interesting to have the names of the shares searchable indexed. It wouldn't matter to much if that spider has indexed an older price, when he came to visit the page, if the indexed name leads to the actual page. The easier solution is to build the browselist with one plugin as USER, where the entries point to a second plugin, on the same or a second side, that contains USER_INT to output the details. Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From erik at linnearad.no Mon Oct 3 23:42:47 2005 From: erik at linnearad.no (Erik Svendsen) Date: Mon, 3 Oct 2005 21:42:47 +0000 (UTC) Subject: [Typo3-dev] Some questions, problems and point of views about xml_parser_create References: Message-ID: Hello List, I have to correct myself on some point. The parser in impexp seem to be alright. I've tried importing two XML-files, one without any TV-structure, and one with no ?, ? or ? in the TV structure. It works fine. The problem should be isolated to the parser in templavoila. WBR, Erik Svendsen www.linnearad.no From pilot at m-lan.ru Tue Oct 4 01:01:00 2005 From: pilot at m-lan.ru (Pilot) Date: Tue, 4 Oct 2005 03:01:00 +0400 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension References:

Message-ID: it's called: "right hand pull to left ear " why you say only about price? what about name, descriptions.. etc.. no... spider is necessary for typo3 "Elmar Hinz" ???????/???????? ? ???????? ?????????: news:mailman.1.1128375658.13797.typo3-dev at lists.netfielders.de... > Pilot wrote: >> it's strange, that nobody wrote search spider for typo3's pages. >> >> it's not so hard, becouse there aer a lot of feaches, for example, such >> string in HTML: >> >> : >> > > Interesting question. > > So, what does a spider do? Crawling websides and storing information about > them into an quickly searchable database. > > The point is, that the information of a typo3 website already is in such a > database. So probably nobody came to the idea of storing them a second > time into the same database. > > But as that special point, where you have a really dynamic plugin output > but with some special static parts in it, spidering could make sense. > > Example? > > Your extension outputs realtime prices of shares. You have to use USER_INT > because the prices change dynamically. But it could be interesting to have > the names of the shares searchable indexed. It wouldn't matter to much if > that spider has indexed an older price, when he came to visit the page, if > the indexed name leads to the actual page. > > The easier solution is to build the browselist with one plugin as USER, > where the entries point to a second plugin, on the same or a second side, > that contains USER_INT to output the details. > > Regards > > Elmar > > > -- > Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, > Austria, Swiss, France, ... > Production of CO2 is killing people. > Production of CO2 just for fun is killing people just for fun. From klausm at in.tum.de Tue Oct 4 03:10:47 2005 From: klausm at in.tum.de (Martin Klaus) Date: Tue, 4 Oct 2005 03:10:47 +0200 Subject: [Typo3-dev] New hook suggestion. References: Message-ID: Hi list, Is it possible to vote for hooks? Then i will do it for this one ;-) I'm a user of the roles extension and I've ran into the extdeveval problem too. Stig's roles extension provides an big advantage for Typo3's user rights management, but not being able to use it together with the extdeveval extension out of the box is a big disadvantage. So please add this backend hook, that allows tx_roles and other extensions to use the top-menu frame at the same time. regards and thanks in advance, Martin Klaus "Stig N?rgaard F?rch" schrieb im Newsbeitrag news:mailman.1.1127488893.12502.typo3-dev at lists.netfielders.de... > Hi guys > > Tell me if it's bad manner to post this here. Hope not. > http://bugs.typo3.org/view.php?id=1481 > > Currently, if you want to put something in the topmenu frame in the > backend, you will have to do this through Xclassing. The problem is that > many developers have installed the 'extdeveval'-extension which makes use > of Xclassing to display links in the topmenu frame. That means that other > extensions can not access this space. > In this patch I have implemented a hook in alt_topmenu_dummy.php which > makes i possible for multiple extensions to add content to the topmenu. > It's possible to set a placement priority(1-9) through the variable > 'priority' of the class which the hook connects to. > This hook is needed if you want to use both 'Roles'-extension and > 'extdeveval'-extension. > > On bugs.typo3.org you can also see an example-gif from the backend. > > Best Regards > Stig N. F?rch From typo3 at fm-world.ru Tue Oct 4 09:00:06 2005 From: typo3 at fm-world.ru (Dmitry Dulepov) Date: Tue, 04 Oct 2005 10:00:06 +0300 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension In-Reply-To: References:

Message-ID: Hi! Pilot wrote: > it's strange, that nobody wrote search spider for typo3's pages. There will be one. Extension name is "crawler" but I think it is still private. Dmitry. From typo3 at fm-world.ru Tue Oct 4 09:06:39 2005 From: typo3 at fm-world.ru (Dmitry Dulepov) Date: Tue, 04 Oct 2005 10:06:39 +0300 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension In-Reply-To: References:

Message-ID: Hi! Elmar Hinz wrote: > So, what does a spider do? Crawling websides and storing information > about them into an quickly searchable database. Right. > > The point is, that the information of a typo3 website already is in such > a database. So probably nobody came to the idea of storing them a second > time into the same database. Wrong. It is in the database but not in "quickly searchable database". > Your extension outputs realtime prices of shares. You have to use > USER_INT because the prices change dynamically. But it could be > interesting to have the names of the shares searchable indexed. It > wouldn't matter to much if that spider has indexed an older price, when > he came to visit the page, if the indexed name leads to the actual page. This is a special search. Indexing is good for texts but not for prices. Texts consist from words and word forms. The one who searches for phrase usually need almost exact match (i.e. if if he look for "typo" than "typo3" and "typo" are both matches good but "mambo" is a bad match :)) For prices, range is usually more interesting than exact match, which makes it very different from text (Noone will search for shares with price set to $34.07, they will search for shares with price form $30 to $40). Also, as you said, price changes quite often, so there is no reason to *index* price data at all. If extension works with prices, it should have its own search capabilities and search data directly. Dmitry. From elmar.hinz at vcd-berlin.de Tue Oct 4 09:25:57 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Tue, 04 Oct 2005 09:25:57 +0200 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension In-Reply-To: References:

Message-ID: > > For prices, range is usually more interesting than exact match, which > makes it very different from text (Noone will search for shares with > price set to $34.07, they will search for shares with price form $30 to > $40). Also, as you said, price changes quite often, so there is no > reason to *index* price data at all. If extension works with prices, it > should have its own search capabilities and search data directly. > > Dmitry. Good. But what follows of that? That you tell the imaginary spider never to index prices at all? Maybe there are some static prices in an extension. Knowing the exact price could bring people back to the product. Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From michael at typo3.org Tue Oct 4 10:30:59 2005 From: michael at typo3.org (Michael Stucki) Date: Tue, 04 Oct 2005 10:30:59 +0200 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension References:

Message-ID: Pilot wrote: ^^^^^ > it's strange, that nobody wrote search spider for typo3's pages. > > it's not so hard, becouse there aer a lot of feaches, for example, such > string in HTML: > > : The bug is that config.disablePrefixComment was ignored when a plugin is inserted. This is already fixed in CVS. - michael -- Use a newsreader! Check out http://typo3.org/community/mailing-lists/use-a-news-reader/ From typo3 at fm-world.ru Tue Oct 4 10:35:02 2005 From: typo3 at fm-world.ru (Dmitry Dulepov) Date: Tue, 04 Oct 2005 11:35:02 +0300 Subject: [Typo3-dev] How do I get indexed search to work on a USER_INT extension In-Reply-To: References:

Message-ID: Hi! Elmar Hinz wrote: > Good. But what follows of that? That you tell the imaginary spider never > to index prices at all? Maybe there are some static prices in an > extension. Knowing the exact price could bring people back to the product. There are and for this purpose in typo3 :) Dmitry. From robert at typo3.org Tue Oct 4 12:41:18 2005 From: robert at typo3.org (Robert Lemke) Date: Tue, 04 Oct 2005 12:41:18 +0200 Subject: [Typo3-dev] Some questions, problems and point of views about xml_parser_create References:

Message-ID: Hi Erik, Erik Svendsen wrote: > I have to correct myself on some point. The parser in impexp seem to be > alright. I've tried importing two XML-files, one without any TV-structure, > and one with no ?, ? or ? in the TV structure. It works fine. The problem > should be isolated to the parser in templavoila. just wanted to let you know that I read your post and will take care of it in beginning of November when I'm working on TemplaVoila. I'm sorry that I don't find the time to look at it earlier ... cheers, robert -- Robert Lemke TYPO3 Association - Research & Development Member of the board http://association.typo3.org From user at macromediaforum.com Tue Oct 4 15:28:49 2005 From: user at macromediaforum.com (Lannsjo) Date: Tue, 4 Oct 2005 15:28:49 +0200 Subject: [Typo3-dev] loginbox Message-ID: Hi, I have this problem that I need to solve... I want to save the timestamp for the last login... so that a user can see when he logged in the last time.. What I want is to save the feUser:lastlogin value in a new field. But I need to save it just before it sets to the new value (when a user logs in).. and I cant find where the new value is set in tx_newloginbox.pi1.php Can someone help me here please /Martin From robertjohn at bedrijvenweb.nl Tue Oct 4 15:56:04 2005 From: robertjohn at bedrijvenweb.nl (Robert John de Stigter) Date: Tue, 4 Oct 2005 15:56:04 +0200 Subject: [Typo3-dev] Extending a FCE form Message-ID: Hi, I got the following situation. Users within my typo3 backend can place articles on pages. The articles are built up with a Flexible Content Element. That way they hand me the content and I take care of how it is placed. My problem is that users can't choose to place a listing or table in an article. Only plain text with the 'Text area for bodytext' editing type. What I did was enable the editing type to a rich text editor. But that gave them to many possibilities causing me to loose control on the output. I've also set the 'Text are for bodytext' to 'Content Elements'. That way they could create content of other types and link them in the article. But I find that to extensive. What I would to do is an item added to 'Editing type' when you build up the datastructure and handle the proccessing of it with some php code. Is there a hook for this maby? I was able to add an item to the 'editing type' list in the file 'templavoila/cm1/index.php', but I couldn't find out where the input is beeing processed (based on the selected editing type). Or are there other (better) ways to work this out. Regards, RJ From stig at 8620.dk Tue Oct 4 16:10:07 2005 From: stig at 8620.dk (=?ISO-8859-1?Q?Stig_N=F8rgaard_F=E6rch?=) Date: Tue, 04 Oct 2005 16:10:07 +0200 Subject: [Typo3-dev] New hook suggestion. In-Reply-To: References:

Message-ID: Hi Martin Klaus, I'm glad for your support to my extension. I hope that the hook will be adopted soon and before 4.0. /Stig Martin Klaus wrote: > Hi list, > > Is it possible to vote for hooks? Then i will do it for this one ;-) > I'm a user of the roles extension and I've ran into the extdeveval problem > too. > > Stig's roles extension provides an big advantage for Typo3's > user rights management, but not being able to use it together with the > extdeveval > extension out of the box is a big disadvantage. > > So please add this backend hook, that allows tx_roles and other extensions > to > use the top-menu frame at the same time. > > regards and thanks in advance, > Martin Klaus > > > "Stig N?rgaard F?rch" schrieb im Newsbeitrag > news:mailman.1.1127488893.12502.typo3-dev at lists.netfielders.de... > >>Hi guys >> >>Tell me if it's bad manner to post this here. Hope not. >>http://bugs.typo3.org/view.php?id=1481 >> >>Currently, if you want to put something in the topmenu frame in the >>backend, you will have to do this through Xclassing. The problem is that >>many developers have installed the 'extdeveval'-extension which makes use >>of Xclassing to display links in the topmenu frame. That means that other >>extensions can not access this space. >>In this patch I have implemented a hook in alt_topmenu_dummy.php which >>makes i possible for multiple extensions to add content to the topmenu. >>It's possible to set a placement priority(1-9) through the variable >>'priority' of the class which the hook connects to. >>This hook is needed if you want to use both 'Roles'-extension and >>'extdeveval'-extension. >> >>On bugs.typo3.org you can also see an example-gif from the backend. >> >>Best Regards >>Stig N. F?rch > > > From wilhelm at icecrash.com Tue Oct 4 16:45:53 2005 From: wilhelm at icecrash.com (Sven Wilhelm) Date: Tue, 04 Oct 2005 16:45:53 +0200 Subject: [Typo3-dev] Better templating: PHP! - Little add-on In-Reply-To: References: Message-ID: http://phptal.motion-twin.com Kind regards Sven From jsegars at alumni.rice.edu Tue Oct 4 17:59:42 2005 From: jsegars at alumni.rice.edu (Jeff Segars) Date: Tue, 04 Oct 2005 10:59:42 -0500 Subject: [Typo3-dev] TCA and Custom Softrefs Message-ID: I'm currently working on an extension that builds on the Import/Export Wizard by using custom softrefs and a simpler import interface. I've written a custom softref parser, but my question is how to make it be called during an export from the Import/Export Wizard. Within tx_impexp->getRelations(), the TCA for the current field is loaded to determine which softref parsers to use. For a field like bodytext, the default configuration is... [softref]=typolink_tag,images,email[subst],url I know I can manually add the key for my softref parser to each field in order to have the parser called. Likewise, I can edit the code in tx_impexp->getRelations() to append my key to whatever the TCA already contains, but neither of these options seem like very clean solutions. Is there some better way that I can force my key to be added to the softref list for every TCA field? Thanks, Jeff Segars From news at monosock.org Tue Oct 4 20:55:29 2005 From: news at monosock.org (Michiel Roos) Date: Tue, 04 Oct 2005 18:55:29 +0000 Subject: [Typo3-dev] release of wizard_crpagetree version 0.2.0 Message-ID: Hi, I just put up a new version of wizard_crpagetree for testing. Anyone care to push it around for a bit? Manual and CSH included. Cheers, Michiel From news at monosock.org Tue Oct 4 21:00:15 2005 From: news at monosock.org (Michiel Roos) Date: Tue, 04 Oct 2005 19:00:15 +0000 Subject: [Typo3-dev] release of wizard_crpagetree version 0.2.0 References: Message-ID: It's still a shy one ;-) From erik at linnearad.no Tue Oct 4 22:48:25 2005 From: erik at linnearad.no (Erik Svendsen) Date: Tue, 4 Oct 2005 20:48:25 +0000 (UTC) Subject: [Typo3-dev] Some questions, problems and point of views about xml_parser_create References: Message-ID: Hello Robert, Thanks, thats would be a great help (for me and other). I know you have a lot to do (as most of us), so take the time you need. I posted it mainly to get tips to get around the problem in meantime, and maybe help others who have the same problem. I'm moving some sites made with TV from one server to another, and we don't want to downgrade php on the new server (beacuse of another app). So I drop the use of some of the FCE's on the new installations, and drop all "bad" characters form TV DS before I export from the old server. Have to do the move before 15. oct. OT! Thanks for interesting and instructive tutorial/speaks at TYCON3. WBR, Erik Svendsen www.linnearad.no > Hi Erik, > > Erik Svendsen wrote: > >> I have to correct myself on some point. The parser in impexp seem to >> be alright. I've tried importing two XML-files, one without any >> TV-structure, and one with no ?, ? or ? in the TV structure. It works >> fine. The problem should be isolated to the parser in templavoila. >> > just wanted to let you know that I read your post and will take care > of it in beginning of November when I'm working on TemplaVoila. I'm > sorry that I don't find the time to look at it earlier ... > > cheers, > robert From gaumondpatrick at hotmail.com Wed Oct 5 00:21:40 2005 From: gaumondpatrick at hotmail.com (Patrick Gaumond) Date: Tue, 04 Oct 2005 18:21:40 -0400 Subject: [Typo3-dev] release of wizard_crpagetree version 0.2.0 In-Reply-To: References: Message-ID: Michiel Roos wrote: > I just put up a new version of wizard_crpagetree for testing. Anyone care > to push it around for a bit? Tested on 3.8 under Windows and PHP 5. It just works! Tried to add alias while creating the page and it worked fine. I think that if you get the text file importing running, that would be a nice replacement of the old "cms_plaintext_import" wich was released in 2003. Patrick From typo3 at monosock.org Wed Oct 5 08:51:22 2005 From: typo3 at monosock.org (Michiel Roos) Date: Wed, 05 Oct 2005 08:51:22 +0200 Subject: [Typo3-dev] release of wizard_crpagetree version 0.2.0 References:

Message-ID: Hey Patrick, Thanks for testing. Adding content input would be nice. We need some sensible rules for: - section delimiting; - what fields of a section to allow; And we shouldn't change the interface much more, it might become too cluttered. Cheers, Michiel From erik at linnearad.no Wed Oct 5 10:32:45 2005 From: erik at linnearad.no (Erik Svendsen) Date: Wed, 5 Oct 2005 08:32:45 +0000 (UTC) Subject: [Typo3-dev] Extending a FCE form References: Message-ID: Hello Robert John de Stigter, > What I did was enable the editing type to a rich text editor. But that > gave them to many possibilities causing me to loose control on the > output. I've also set the 'Text are for bodytext' to 'Content > Elements'. That way they could create content of other types and link > them in the article. But I find that to extensive. You have the opportunity to restrict the possibilities aviable in the RTE. E.g with TSconfig on either user, group or page level. I did it on a installation on user level a year ago. I think the documentations for the RTE have something about it, and also it's a video about the problem. If you are using htmlarea as RTE, the documentation are really extensiv at these point. WBR, Erik Svendsen www.linnearad.no From Andreas.Foerthner at netlogix.de Wed Oct 5 12:14:42 2005 From: Andreas.Foerthner at netlogix.de (=?ISO-8859-15?Q?Andreas_F=F6rthner?=) Date: Wed, 05 Oct 2005 12:14:42 +0200 Subject: [Typo3-dev] configure Access in Page header via a plugin on this page Message-ID: Hi list, is it possible to configure/set the FE-Usergroup of a page in the page header via a Plugin on this page? The problem is: I have a backend-module, that creates data in a database that is connected to a specific FE-Usergroup, which is also created by the module. The plugin now should read this data an automatically should show the page it is on, only to users of the connected usergroup. How would you do that, so that the author, who inserts the plugin, doesn't have to know the name of the usergroup? Thank you for your help. Greetings Andreas From Martin.Kutschker at n0spam-blackbox.net Wed Oct 5 14:33:42 2005 From: Martin.Kutschker at n0spam-blackbox.net (Martin Kutschker) Date: Wed, 05 Oct 2005 14:33:42 +0200 Subject: [Typo3-dev] Making translation for Core modules Message-ID: Hi! The csh_* Files for German and Bulgarian lack translations of eg Web>Template. Can I translate them for a customer without any external tools? Or do I have to be an official translator? I checked the sources for the csh_* extensions, but hey, there are lots of xml files and I didnn't figure out (quickly) how I would add files for the missing core extensions. Masi From brikou at gmail.com Wed Oct 5 15:41:36 2005 From: brikou at gmail.com (Brice BERNARD) Date: Wed, 5 Oct 2005 15:41:36 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" Message-ID: Have you ever try this search : "Page is being generated. If this message does not disappear within 30 seconds, please reload"? You will reach every page google failed to crawl, it's a big problem because google indexes this page as it was the one expected. How is it possible to avoid this message to appear, maybe we could prevent robot from indexing page by adding the meta : ... From typo3 at rvt.dds.nl Wed Oct 5 16:54:13 2005 From: typo3 at rvt.dds.nl (Ries van Twisk) Date: Wed, 05 Oct 2005 08:54:13 -0600 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References: Message-ID: Brice BERNARD wrote: >Have you ever try this search : "Page is being generated. If this message >does not disappear within 30 seconds, please >reload"? >You will reach every page google failed to crawl, it's a big problem >because google indexes this page as it was the one expected. > >How is it possible to avoid this message to appear, maybe we could prevent >robot from indexing page by adding the meta : content="noindex" />... >_______________________________________________ > > That would be indeed a good idea. I have luckly just one page in google index like that but I try to check this almost every week! http://www.google.com.ec/search?hs=IiT&hl=es&client=firefox-a&rls=org.mozilla%3Aen-US%3Aofficial&q=site%3Awww.livetravelguides.com+30+seconds&btnG=B%C3%BAsqueda&meta= setting a noindex on such a page sis indeed a really good idea. What do core developers think about this? regards, Ries van Twisk From steffen at kommwiss.fu-berlin.de Wed Oct 5 15:53:23 2005 From: steffen at kommwiss.fu-berlin.de (=?ISO-8859-1?Q?Steffen_M=FCller?=) Date: Wed, 05 Oct 2005 15:53:23 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References: Message-ID: Hi. Brice BERNARD wrote: > Have you ever try this search : "Page is being generated. If this message > does not disappear within 30 seconds, please > reload"? > You will reach every page google failed to crawl, it's a big problem > because google indexes this page as it was the one expected. > > How is it possible to avoid this message to appear, maybe we could prevent > robot from indexing page by adding the meta : content="noindex" />... Maybe a change of the HTTP status to 503 "Service Unavailable" would be a workaround. 503 is meant to be temporary. What's the correct handling? Kick the "Page is beeing generated page" and risk a timeout? -- cheers, steffen From oliver at liquidlight.co.uk Wed Oct 5 17:34:56 2005 From: oliver at liquidlight.co.uk (Oliver) Date: Wed, 05 Oct 2005 16:34:56 +0100 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References: Message-ID: Hi Brice, You can disable that message by commenting out line number 1335 in tslib/class.tslib_fe.php ie change: $this->tempPageCacheContent(); to: //$this->tempPageCacheContent(); Hope this helps, Oliver Brice BERNARD wrote: > Have you ever try this search : "Page is being generated. If this message > does not disappear within 30 seconds, please > reload"? > You will reach every page google failed to crawl, it's a big problem > because google indexes this page as it was the one expected. > > How is it possible to avoid this message to appear, maybe we could prevent > robot from indexing page by adding the meta : content="noindex" />... From r.fritz at colorcube.de Wed Oct 5 18:11:34 2005 From: r.fritz at colorcube.de (=?iso-8859-1?q?Ren=E9_Fritz?=) Date: Wed, 5 Oct 2005 18:11:34 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References: Message-ID: I collected the suggestions: ... ... header('Status: 503 Service Unavailable'); header('Retry-After: 30'); // 30 seconds is hardcoded but is ok I think header('Expires: -1'); header('Cache-Control: post-check=0, pre-check=0'); header('Pragma: no-cache'); header('Last-Modified: '.gmdate('D, d M Y H:i:s').' GMT'); Is that right? I'm a bit unsure about the caching thing which has just changed a little for the "Page is being generated" page. I have to ask Kasper. Ren? -- COLORCUBE digital media lab www.colorcube.de From Martin.Kutschker at n0spam-blackbox.net Wed Oct 5 18:28:50 2005 From: Martin.Kutschker at n0spam-blackbox.net (Martin Kutschker) Date: Wed, 05 Oct 2005 18:28:50 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: Oliver schrieb: > Hi Brice, > > You can disable that message by commenting out line number 1335 in > tslib/class.tslib_fe.php > > ie change: > $this->tempPageCacheContent(); > to: > //$this->tempPageCacheContent(); > > Hope this helps, No, because Google will index an empty page. Both ideas, setting the HTTP header and the HTML meta tag are good ideas. Masi From brikou at gmail.com Wed Oct 5 18:54:01 2005 From: brikou at gmail.com (Brice BERNARD) Date: Wed, 5 Oct 2005 18:54:01 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: I think these feature would be really great if implemented on typo3 3.9 :) - Brice - 2005/10/5, Ren? Fritz : > > I collected the suggestions: > > ... > > > > > ... > > header('Status: 503 Service Unavailable'); > header('Retry-After: 30'); // 30 seconds is hardcoded but is ok I think > > header('Expires: -1'); > header('Cache-Control: post-check=0, pre-check=0'); > header('Pragma: no-cache'); > header('Last-Modified: '.gmdate('D, d M Y H:i:s').' GMT'); > > > Is that right? > > I'm a bit unsure about the caching thing which has just changed a little > for > the "Page is being generated" page. I have to ask Kasper. > > Ren? > > > -- > COLORCUBE > digital media lab > > www.colorcube.de > _______________________________________________ > Typo3-dev mailing list > Typo3-dev at lists.netfielders.de > http://lists.netfielders.de/cgi-bin/mailman/listinfo/typo3-dev > From elmar.hinz at vcd-berlin.de Wed Oct 5 18:56:11 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Wed, 05 Oct 2005 18:56:11 +0200 Subject: [Typo3-dev] useCacheHash in FE-plugins, typolink, pi_base Message-ID: Hello FE-developers, we find 2 articels in the developers corner describing FE-links and cHash philosophie: http://typo3.org/development/articles/the-mysteries-of-chash/ http://typo3.org/development/articles/using-links-in-frontend-plugins/ Both articels don't touch the question how the cHash is initiated/created at all. This led me to the assumption that the cHash is created automatically by typolink function if it is used inside a USER plugin. What speaks against this assumption is the presence of a useCacheHash property in TSref for the typolink function. TYPOLINK 1.) Is cHash created automatically inside USER objects by the typolink function or do you have to build or configure them? 2.) How do you build links that match the concept of cHash and USER plugins? PI_BASE 3.a) Do the pi_base link functions work with the concept of cHash at all? 3.b) Do they set it automatically or do you have to configure it for pi_base link functions? 3.c) If you have to configure it how do you do it correctly? Thank you Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From typo3 at ingo-renner.com Wed Oct 5 19:28:57 2005 From: typo3 at ingo-renner.com (Ingo Renner) Date: Wed, 5 Oct 2005 19:28:57 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" References:

Message-ID: Am Wed, 5 Oct 2005 18:11:34 +0200 schrieb Ren? Fritz: Hi, > header('Status: 503 Service Unavailable'); > header('Retry-After: 30'); // 30 seconds is hardcoded but is ok I think > > header('Expires: -1'); > header('Cache-Control: post-check=0, pre-check=0'); > header('Pragma: no-cache'); > header('Last-Modified: '.gmdate('D, d M Y H:i:s').' GMT'); I would go for the second option. Because when telling search engines not to index the given URL they might not come back to this URL. But actually we want them to visit us again and get the correct content. Thus a 503 would be better as it is temporary like someone mentioned before... Ingo -- Use a newsreader! Check out http://typo3.org/community/mailing-lists/use-a-news-reader/ From r.fritz at colorcube.de Wed Oct 5 19:38:42 2005 From: r.fritz at colorcube.de (=?iso-8859-1?q?Ren=E9_Fritz?=) Date: Wed, 5 Oct 2005 19:38:42 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: > I would go for the second option. Because when telling search engines not > to index the given URL they might not come back to this URL. But actually > we want them to visit us again and get the correct content. Thus a 503 > would be better as it is temporary like someone mentioned before... That means should be removed... Ren? -- COLORCUBE digital media lab www.colorcube.de From ernst at cron-it.de Wed Oct 5 19:37:29 2005 From: ernst at cron-it.de (Ernesto Baschny [cron IT]) Date: Wed, 05 Oct 2005 19:37:29 +0200 Subject: [Typo3-dev] useCacheHash in FE-plugins, typolink, pi_base In-Reply-To: References: Message-ID: Elmar Hinz schrieb am 05.10.2005 18:56: > (...) > Both articels don't touch the question how the cHash is > initiated/created at all. This led me to the assumption that the cHash > is created automatically by typolink function if it is used inside a > USER plugin. That is correct. > What speaks against this assumption is the presence of a useCacheHash > property in TSref for the typolink function. Not really. Your plugin has to decide if a link is "cacheable" or not and set useCacheHash appropriately for links that are worthwhile to be cached. > TYPOLINK > > 1.) Is cHash created automatically inside USER objects by the typolink > function or do you have to build or configure them? It's created automatically if you call the typolink function with the "useCacheHash" property set. Normally in plugins you will create links using the pi_linkTP*() family of functions. These will set the useCacheHash automatically for you if you set the $cache parameter to "1" (ok, if you have $this->pi_USER_INT_obj = true in your plugin, this will override the $cache parameter). To decide which link to $cache and which not, you have to think if the combination of parameters that the user is submitting is something worthwhile to cache or not. Only make cacheable links if you have a countable amount of parameter/value combinations. See tt_news as an example. Its an USER object, and thus cacheable, if the user set the "allowCaching" config. All links go through pi_linkTP* and if allowCaching was true, these will get an cHash calculated and appended (tt_news-developer doesn't have to do anything for this to happen). The link-destination will then see the cHash, the parameters, and see they match, thus it was a link created by TYPO3, and thus, will cache the resulting output. This way noone can "pollute" typo3 cache just by requesting the same page with different parameters. > 2.) How do you build links that match the concept of cHash and USER > plugins? See tt_news, for an up-to-date example. > PI_BASE > > 3.a) Do the pi_base link functions work with the concept of cHash at all? See above. Yes, they all call typolink internally, so it works the same way. > 3.b) Do they set it automatically or do you have to configure it for > pi_base link functions? Just set the $cache parameter and make sure you have not set $this->pi_USER_INT_obj to TRUE in your plugin. > 3.c) If you have to configure it how do you do it correctly? Not needed! Cheers, Ernesto From elmar.hinz at vcd-berlin.de Wed Oct 5 19:52:55 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Wed, 05 Oct 2005 19:52:55 +0200 Subject: [Typo3-dev] useCacheHash in FE-plugins, typolink, pi_base In-Reply-To: References:

Message-ID: Thank you for your detailed answer. With this I see clear enough now. Remember to set $pi_checkCHash for USER to avoid the "empty cHash" problems. var $pi_checkCHash = TRUE; See: http://typo3.org/development/articles/the-mysteries-of-chash/ Example for function pi_linkTP_keepPIvars_url($overrulePIvars=array(),$cache=0,$clearAnyway=0): $pars = array('one' => 1, 'two' => 2, 'three' => 3); $label = 'Read more ...'; $ankerLink = $this->pi_linkTP_keepPIvars($label,$pars,1,1); Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From Martin.Kutschker at n0spam-blackbox.net Thu Oct 6 09:48:25 2005 From: Martin.Kutschker at n0spam-blackbox.net (Martin Kutschker) Date: Thu, 06 Oct 2005 09:48:25 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: Ren? Fritz schrieb: >>I would go for the second option. Because when telling search engines not >>to index the given URL they might not come back to this URL. But actually >>we want them to visit us again and get the correct content. Thus a 503 >>would be better as it is temporary like someone mentioned before... > > > That means > > should be removed... Or and ? So even if the page will be in the index, it will not archived with the temporary content? Masi From elmar.hinz at vcd-berlin.de Thu Oct 6 11:01:40 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Thu, 06 Oct 2005 11:01:40 +0200 Subject: [Typo3-dev] Empty cHash problem Message-ID: Hello people, Kasper Sk?rh?j describes in his article "The mysteries of &cHash" the empty &cHash problem. http://typo3.org/development/articles/the-mysteries-of-chash/page/3/ I wonder if this is really an empty cHash or rather an empty cache problem. It happens according to Kaspers article: a) when the cache is cleared for that page (is empty) b) when the first call comes with paramters that request a special view of the page c) when no cHash is given Wouldn't it also happen when a) ... b) ... c) a cHash is given but a wrong one Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From r.fritz at colorcube.de Thu Oct 6 11:34:24 2005 From: r.fritz at colorcube.de (=?iso-8859-1?q?Ren=E9_Fritz?=) Date: Thu, 6 Oct 2005 11:34:24 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: > Or and CONTENT="NOSNIPPET">? > > So even if the page will be in the index, it will not archived with the > temporary content? What exactly does the tags mean? Ren? -- COLORCUBE digital media lab www.colorcube.de From oliver at liquidlight.co.uk Thu Oct 6 11:48:14 2005 From: oliver at liquidlight.co.uk (Oliver) Date: Thu, 06 Oct 2005 10:48:14 +0100 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: nosnippet - From Google help: "A snippet is a text excerpt from the returned result page that has all query terms bolded. The excerpt allows users to see the context in which search terms appear on a web page, before clicking on the result. Users are more likely to click on a search result if it has a corresponding snippet." This value of NOSNIPPET removes the text snippet. noarchive - From Google help: "Google keeps the text of the many documents it crawls available in a cache. This allows an archived, or "cached", version of a web page to be retrieved for your end users if the original page is ever unavailable (due to temporary failure of the page's web server). The cached page appears to users exactly as it looked when Google last crawled it. The cached page also includes a message (at the top of the page) to indicate that it's a cached version of the page." Though I've heard that 'nosnippet' is no longer supported by google. Oliver Ren? Fritz wrote: >>Or and >CONTENT="NOSNIPPET">? >> >>So even if the page will be in the index, it will not archived with the >>temporary content? > > > What exactly does the tags mean? > > Ren? > From niederlag at ikd01.de Thu Oct 6 13:27:16 2005 From: niederlag at ikd01.de (Peter Niederlag) Date: Thu, 06 Oct 2005 13:27:16 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations Message-ID: Hi, ----------------------------------------------------------------- I'd like to get your opinions on the default handling of M-M-relations in TYPO3 and extensions created by the Kickstarter. ----------------------------------------------------------------- Currently the Kickstarter uses 'commalist of values' to realize M-M-relations on database-tables by default. In my opinion 'commalist of values' might be easier to handle for novices in PHP- and SQL-development but tend to lead to somewhat bad application-design. ------------------------------------------------------------------ Explanations ------------------------------------------------------------------ 'commalist of values' makes some operations truely difficult or expensive. For example "give me all users that belong to usergroup 234" can only be done by some expensive and error-prone "LIKE ..." SQL-construct or PHP-operation on the Result-Set. Also M-M-relations as 'commalist of values' does prevent the usage of Object-Relational-Toolkits such as Propel (PHP-based, http://propel.phpdb.org/trac/) or Hibernate (Java-based) from being a choice for data-handling or at least make it very, very difficult. --------------------- My suggestions: --------------------- * Switch to a true M-M-relation by default in the Kickstarter as soon as possible, leaving 'commalist of values a user-choice with an explanatory "warning" * Use true M-M relations for any new stuff possibly introduced in the TYPO3-core * Cleaning up of any 'commalist of values' in TYPO3-core for Version 5.0 Plz let me know what you think. Thx and cheers, Peter -- Peter Niederlag http://www.niekom.de * TYPO3 & EDV Dienstleistungen * http://www.typo3partner.net * professional services network * From Martin.Kutschker at n0spam-blackbox.net Thu Oct 6 14:12:08 2005 From: Martin.Kutschker at n0spam-blackbox.net (Martin Kutschker) Date: Thu, 06 Oct 2005 14:12:08 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: Oliver schrieb: > > Though I've heard that 'nosnippet' is no longer supported by google. Somebody should email them to update their help docs :-) Masi From arnsholt at broadpark.no Thu Oct 6 14:43:59 2005 From: arnsholt at broadpark.no (Arne Skjaerholt) Date: Thu, 06 Oct 2005 14:43:59 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References: Message-ID: I've run into some problems myself with this when developing. As you said, I found a (reasonably) painless solution with LIKE. In my opinion the best solution would be your option number three and removing all the comma-separated stuff from the next version, but options one and two might be good additions while the next version is being developed. Arne :wq From schwarzkopf.no.spam at artplan21.de Thu Oct 6 14:46:12 2005 From: schwarzkopf.no.spam at artplan21.de (Andreas Schwarzkopf) Date: Thu, 06 Oct 2005 14:46:12 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References: Message-ID: Hi Peter, Peter Niederlag schrieb: > 'commalist of values' makes some operations truely difficult or > expensive. For example "give me all users that belong to usergroup 234" > can only be done by some expensive and error-prone "LIKE ..." > SQL-construct or PHP-operation on the Result-Set. FIND_IN_SET('needle',field) works pretty good though it is not standard compliant and can be not used with the DBAL - therefore also not a stable solution > > Also M-M-relations as 'commalist of values' does prevent the usage of > Object-Relational-Toolkits such as Propel (PHP-based, > http://propel.phpdb.org/trac/) or Hibernate (Java-based) from being a > choice for data-handling or at least make it very, very difficult. > > --------------------- > My suggestions: > --------------------- > > * Switch to a true M-M-relation by default in the Kickstarter as soon > as possible, leaving 'commalist of values a user-choice with an > explanatory "warning" +1 > * Use true M-M relations for any new stuff possibly introduced in the > TYPO3-core +1 > * Cleaning up of any 'commalist of values' in TYPO3-core for Version 5.0 +1 (IMHO not in earlier versions) grtx Andreas From kraftb at gmx.net Thu Oct 6 15:49:54 2005 From: kraftb at gmx.net (Kraft Bernhard) Date: Thu, 06 Oct 2005 15:49:54 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References: Message-ID: Peter Niederlag wrote: > Currently the Kickstarter uses 'commalist of values' to realize > M-M-relations on database-tables by default. Commalist-separated realtions are a mess in the view of every database designer :) > * Cleaning up of any 'commalist of values' in TYPO3-core for Version 5.0 +1 !! I think it should already get started in 4.0 and there should be an option in 4.0 to enable this feature (altough it probably is still unstable). This would enable administrators which set up a new installation to already use true M-M relations. Also a tool should get coded which converts already existing databases from the bad-relations to the good-ones :) greets, Bernhard -- Kraft Bernhard MOKKA Medienagentur T: +43 - 1 - 895 33 33 - 50 From ernst at cron-it.de Thu Oct 6 16:21:41 2005 From: ernst at cron-it.de (Ernesto Baschny [cron IT]) Date: Thu, 06 Oct 2005 16:21:41 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References: Message-ID: Elmar Hinz schrieb am 06.10.2005 11:01: > Kasper Sk?rh?j describes in his article "The mysteries of &cHash" the > empty &cHash problem. > > http://typo3.org/development/articles/the-mysteries-of-chash/page/3/ > > I wonder if this is really an empty cHash or rather an empty cache problem. Empty cHash problem. Empty cache is never a problem, as it can be filled. :) And if its empty, the empty cHash is not a problem, its just a problem if the cache already filled! The problem, to sum up: When generating the page that will be cached where we have our USER plugin. The plugin will output something based on the piVars (e.g. display the detail view of the current record). If we don't take any precaution, our output will always be cached too by TYPO3 (this is why we are an USER object!). So the next requests, regardless of the parameters, would display the same output as the first request, thus making our parameters useless! Only the first request after clearing the cache wins. We have to avoid that in the first call already, by only allowing TYPO3 to cache the output if we are sure that the parameter combination is something can be checked against a valid cHash. > It happens according to Kaspers article: > > a) when the cache is cleared for that page (is empty) > b) when the first call comes with paramters that request a special view > of the page > c) when no cHash is given > > Wouldn't it also happen when > > a) ... > b) ... > c) a cHash is given but a wrong one The article section is devoted to the case when cHash is EMPTY, so this case does not apply. Also in this case, TYPO3 will see that the cHash is not valid for the current parameter set and caching will be disabled altogether, which has the same effect as having an empty cHash. Cheers, Ernesto From kasper2005 at typo3.com Thu Oct 6 16:44:06 2005 From: kasper2005 at typo3.com (Kasper =?iso-8859-1?q?Sk=E5rh=F8j?=) Date: Thu, 6 Oct 2005 16:44:06 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: Fine with me. Make sure that other cache-related headers are not send if you send these. You might have to toggle that with the flag that exposes that temp-content is delivered because from a technical point of view the 30-second notice is just a page cached for 30 seconds... - kasper On Wednesday 05 October 2005 18:11, Ren? Fritz wrote: > I collected the suggestions: > > ... > > > > > ... > > header('Status: 503 Service Unavailable'); > header('Retry-After: 30'); // 30 seconds is hardcoded but is ok I think > > header('Expires: -1'); > header('Cache-Control: post-check=0, pre-check=0'); > header('Pragma: no-cache'); > header('Last-Modified: '.gmdate('D, d M Y H:i:s').' GMT'); > > > Is that right? > > I'm a bit unsure about the caching thing which has just changed a little > for the "Page is being generated" page. I have to ask Kasper. > > Ren? -- - kasper ----------------- Think future, not feature From kasper2005 at typo3.com Thu Oct 6 16:45:43 2005 From: kasper2005 at typo3.com (Kasper =?iso-8859-1?q?Sk=E5rh=F8j?=) Date: Thu, 6 Oct 2005 16:45:43 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References: Message-ID: > Wouldn't it also happen when > > a) ... > b) ... > c) a cHash is given but a wrong one Nope, it will not. The problem as described in that article is specifically dealing with an empty value of cHash. A value found will trigger a verification process that afaik works just fine. - kasper From elmar.hinz at vcd-berlin.de Thu Oct 6 17:03:58 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Thu, 06 Oct 2005 17:03:58 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References:

Message-ID: > Empty cHash problem. Empty cache is never a problem, as it can be > filled. :) And if its empty, the empty cHash is not a problem, its just > a problem if the cache already filled! ??? Either this is not precise enough or tells something else then the article. There isn't a problem when the cache is already filled in the article. The problem is filling the empty cache with the right stuff. > > The problem, to sum up: > > When generating the page that will be cached where we have our USER > plugin. The plugin will output something based on the piVars (e.g. > display the detail view of the current record). If we don't take any > precaution, our output will always be cached too by TYPO3 (this is why > we are an USER object!). So the next requests, regardless of the > parameters, would display the same output as the first request, thus > making our parameters useless! Only the first request after clearing the > cache wins. So far it is right. Here we still have only one cache for all views. We have to avoid that in the first call already, by only > allowing TYPO3 to cache the output if we are sure that the parameter > combination is something can be checked against a valid cHash. Here seems to be a gap in your explenation. You still haven't created different chaches for different combinations of parameters. By the way. Is there a special table for caching views with additional parameters or is it the same table as for the old id + type combination? Which tables are they? Wich fields in the tables? > > >>It happens according to Kaspers article: >> >>a) when the cache is cleared for that page (is empty) >>b) when the first call comes with paramters that request a special view >>of the page >>c) when no cHash is given >> >>Wouldn't it also happen when >> >>a) ... >>b) ... >>c) a cHash is given but a wrong one > > > The article section is devoted to the case when cHash is EMPTY, so this > case does not apply. That is a quite philisophical answer. > > Also in this case, TYPO3 will see that the cHash is not valid for the > current parameter set and caching will be disabled altogether, which has > the same effect as having an empty cHash. That is exactly my thesis. The effect would be the same. Even if there isn't an empty cHash but an invalid one. > > Cheers, > Ernesto -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From typo3 at rvt.dds.nl Thu Oct 6 18:22:49 2005 From: typo3 at rvt.dds.nl (Ries van Twisk) Date: Thu, 06 Oct 2005 10:22:49 -0600 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: Arne Skjaerholt wrote: >I've run into some problems myself with this when developing. As you >said, I found a (reasonably) painless solution with LIKE. In my opinion >the best solution would be your option number three and removing all the >comma-separated stuff from the next version, but options one and two >might be good additions while the next version is being developed. > >Arne >:wq > >_______________________________________________ > > I don't understand how I can do a like to look for a relation. For example I have this list in a record: 345,6,17,231 How would I do a like query on this one to look for number for UID 1? Also, a query like this: LIKE ('%somedata%') cannot be optimized at the DB level because the RDBM cannot do a index scan but rather have to do a tablescan, means has to scan ALL records! I would love to see MM relations, unfortunatly I don't think that's possible anymore with typo3.... I have seen however that Kasper did some speed measurements and found that on mysql there is hardly no speed differences and so decided not to go for a relational DB setup. Ries From elmar.hinz at vcd-berlin.de Thu Oct 6 17:31:29 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Thu, 06 Oct 2005 17:31:29 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References:

Message-ID: Kasper Sk?rh?j wrote: >>Wouldn't it also happen when >> >>a) ... >>b) ... >>c) a cHash is given but a wrong one > > > Nope, it will not. The problem as described in that article is specifically > dealing with an empty value of cHash. A value found will trigger a > verification process that afaik works just fine. > > - kasper Hello Kasper, Thank you for precise ansewer. I will reflect about this point. Let me kindley hint you in this occasion to a small vacancy in this exellent article, that is really an ey-opener upon caching. I found no hint how the cHash is added to the link in practice. In the first edition of the "TYPO3-Handbuch from Galileo Press" that led me to the assumtion that it would be created under 2 mayor conditions automatically: 1. the use of USER-Object 2. the use of typolink-function It is created automatically, but you need a third condition: 3. setting of useCacheHash-configuration for the typolink function or of cache=1 parameter for pi_link-functions respectively. That information will hopfully run already into the second print of the book. Best Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From kasper2005 at typo3.com Thu Oct 6 17:39:01 2005 From: kasper2005 at typo3.com (Kasper =?iso-8859-1?q?Sk=E5rh=F8j?=) Date: Thu, 6 Oct 2005 17:39:01 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: For scanning a list of uids, there is a function in t3lib_db class that puts together four LIKE clauses. Works well. Regarding the principle of MM vs. uid lists I see your points but it is not an overnight change. The perspective of version 5 is possible, not anytime before. - kasper Think future, not feature From steffen at mail.kommwiss.fu-berlin.de Thu Oct 6 18:03:37 2005 From: steffen at mail.kommwiss.fu-berlin.de (=?ISO-8859-1?Q?Steffen_M=FCller?=) Date: Thu, 06 Oct 2005 18:03:37 +0200 Subject: [Typo3-dev] Google problem : "page is being generated" In-Reply-To: References:

Message-ID: Hi. On 05.10.2005 18:11 Ren? Fritz wrote: > > header('Status: 503 Service Unavailable'); > header('Retry-After: 30'); // 30 seconds is hardcoded but is ok I think > This is syntactically right. My prefered way would be to simply keep the session alive until the page is generated and then deliver it. (Set a timeout of 30 seconds if you like to). No "please wait" page, no HTTP response until the page is generated. Maybe as an option in the install-tool? -- cheers, Steffen From ernst at cron-it.de Thu Oct 6 18:03:55 2005 From: ernst at cron-it.de (Ernesto Baschny [cron IT]) Date: Thu, 06 Oct 2005 18:03:55 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References:

Message-ID: Elmar Hinz schrieb am 06.10.2005 17:03: >> Empty cHash problem. Empty cache is never a problem, as it can be >> filled. :) And if its empty, the empty cHash is not a problem, its just >> a problem if the cache already filled! > Either this is not precise enough or tells something else then the > article. There isn't a problem when the cache is already filled in the > article. The problem is filling the empty cache with the right stuff. If the cache is empty, it will be filled by the first attempt to request the given page. The problem is if we fill the cache, but the content is not really meant to be cached for that combination of parameters. In the article, this results in: It is not only confusing but altogether impossible to see the archive list because its place in the cache is taken by the article until someone clears the cache! So clearly this is implying that the cache was filled improperly, which is what I tried to "sum up" below: >> The problem, to sum up: >> >> When generating the page that will be cached where we have our USER >> plugin. The plugin will output something based on the piVars (e.g. >> display the detail view of the current record). If we don't take any >> precaution, our output will always be cached too by TYPO3 (this is why >> we are an USER object!). So the next requests, regardless of the >> parameters, would display the same output as the first request, thus >> making our parameters useless! Only the first request after clearing the >> cache wins. > So far it is right. Here we still have only one cache for all views. >> We have to avoid that in the first call already, by only >> allowing TYPO3 to cache the output if we are sure that the parameter >> combination is something can be checked against a valid cHash. > Here seems to be a gap in your explenation. You still haven't created > different chaches for different combinations of parameters. That's exactly the problem: if we allow TYPO3 to cache our output and we are an USER object, we won't get a second chance to generate other output for different views if we don't tell TYPO3 that we have different views (by linking to our second view with a cHash parameter, see previous thread). So every call to the same page that contains our plugin will return the same content, regardless of the parameters. This is why we (as a plugin writer) need to make sure that the cHash is being checked (if present, TYPO3 will check it on its own, if not present or empty, TYPO3 needs the plugins to inform if a checking is needed, which is what this section of the article is about). > By the way. Is there a special table for caching views with additional > parameters or is it the same table as for the old id + type combination? > Which tables are they? Wich fields in the tables? Its the same table, as the cache already uses different combination of parameters for caching the same page (e.g for different types). The table is cache_pages, the content is in the HTML field. You will have a huge configuration array in cache_data, which unserialized will also tell you (amongst other) the piVar-parameters that were used to create this cache request. But what TYPO3 uses to see which record to get is the md5-hash of a series of information (fe_group, parameters, type, language, etc), which will be compared against the "hash" field. >> The article section is devoted to the case when cHash is EMPTY, so this >> case does not apply. > That is a quite philisophical answer. Hehe, not really, because if cHash is not empty, the plugin writer doesn't have to do anything, because TYPO3 will already take care of it. This is why it's not really part of this section of the article, which deals with stuff that a plugin writer needs to do, and why its so. > That is exactly my thesis. The effect would be the same. Even if there > isn't an empty cHash but an invalid one. Right, just that with invalid ones the plugin writer won't even be bothered with it, which is not the case if there is no, or an empty cHash value. Cheers, Ernesto PS: hey, first posting of Kasper after 6 months absense of typo3-dev! Welcome back, Kasper! :) From niederlag at ikd01.de Thu Oct 6 18:13:12 2005 From: niederlag at ikd01.de (Peter Niederlag) Date: Thu, 06 Oct 2005 18:13:12 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: Hi, Kasper Sk?rh?j schrieb: > For scanning a list of uids, there is a function in t3lib_db class that puts > together four LIKE clauses. Works well. > > Regarding the principle of MM vs. uid lists I see your points but it is not an > overnight change. Of course I understand that. But the longer the Kickstarter produces uid lists by default the more things to clean in the future. > The perspective of version 5 is possible, not anytime > before. Top prio would be imho to assure that any new tables, fields introduced are using true MM-relations. That is my main concern. Second prio should be a good analysis if there is a need or wish to change old structure to new design in Version 5. To strengthen the point again: Imho we should change Kickstarter's default-behaviour *now* (read: as soon as possible) and take any other measures any time later. Cheers and thx, Peter -- Peter Niederlag http://www.niekom.de * TYPO3 & EDV Dienstleistungen * http://www.typo3partner.net * professional services network * From kasper2005 at typo3.com Thu Oct 6 17:41:10 2005 From: kasper2005 at typo3.com (Kasper =?iso-8859-1?q?Sk=E5rh=F8j?=) Date: Thu, 6 Oct 2005 17:41:10 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References:

Message-ID: thanks Elmar, I'm happy that only this minor lack is found :-) I'm not sure I will find time to update the article myself though. - kasper On Thursday 06 October 2005 17:31, Elmar Hinz wrote: > Kasper Sk?rh?j wrote: > >>Wouldn't it also happen when > >> > >>a) ... > >>b) ... > >>c) a cHash is given but a wrong one > > > > Nope, it will not. The problem as described in that article is > > specifically dealing with an empty value of cHash. A value found will > > trigger a verification process that afaik works just fine. > > > > - kasper > > Hello Kasper, > > Thank you for precise ansewer. I will reflect about this point. > > Let me kindley hint you in this occasion to a small vacancy in this > exellent article, that is really an ey-opener upon caching. > > I found no hint how the cHash is added to the link in practice. In the > first edition of the "TYPO3-Handbuch from Galileo Press" that led me to > the assumtion that it would be created under 2 mayor conditions > automatically: > > 1. the use of USER-Object > 2. the use of typolink-function > > It is created automatically, but you need a third condition: > > 3. setting of useCacheHash-configuration for the typolink function or of > cache=1 parameter for pi_link-functions respectively. > > That information will hopfully run already into the second print of the > book. > > Best Regards > > Elmar -- - kasper ----------------- Think future, not feature From kasper2005 at typo3.com Thu Oct 6 18:32:24 2005 From: kasper2005 at typo3.com (Kasper =?iso-8859-1?q?Sk=E5rh=F8j?=) Date: Thu, 6 Oct 2005 18:32:24 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: The kickstarter is Daniel Br?n / Ingmar Schlechts work. I think changing the _default_ behaviour is a good idea (still have possibility of uid-lists of course). - kasper On Thursday 06 October 2005 18:13, Peter Niederlag wrote: > Hi, > > Kasper Sk?rh?j schrieb: > > For scanning a list of uids, there is a function in t3lib_db class that > > puts together four LIKE clauses. Works well. > > > > Regarding the principle of MM vs. uid lists I see your points but it is > > not an overnight change. > > Of course I understand that. But the longer the Kickstarter produces uid > lists by default the more things to clean in the future. > > > The perspective of version 5 is possible, not anytime > > before. > > Top prio would be imho to assure that any new tables, fields introduced > are using true MM-relations. That is my main concern. > > Second prio should be a good analysis if there is a need or wish to > change old structure to new design in Version 5. > > To strengthen the point again: > Imho we should change Kickstarter's default-behaviour *now* (read: as > soon as possible) and take any other measures any time later. > > Cheers and thx, > Peter -- - kasper ----------------- Think future, not feature From elmar.hinz at vcd-berlin.de Thu Oct 6 18:43:29 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Thu, 06 Oct 2005 18:43:29 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References:

Message-ID: Kasper Sk?rh?j wrote: > thanks Elmar, I'm happy that only this minor lack is found :-) > > I'm not sure I will find time to update the article myself though. > > - kasper > Maybe Robert findes the time to spread this information in http://typo3.org/development/articles/using-links-in-frontend-plugins/ Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From intsys at swissinfo.org Thu Oct 6 20:38:08 2005 From: intsys at swissinfo.org (Stefan Beylen) Date: Thu, 06 Oct 2005 20:38:08 +0200 Subject: [Typo3-dev] Use T3 Database records for external applications Message-ID: Hello list, I want to use some tables filled with records from my T3 installation in an external PHP application (non-t3) Problem that I have is that t3 uses special tags such as 4ex Hi all, I'm having the hardest time finding a solution for this. I have a lot of pages that are only viewable if you are a logged-in user; however, I want those pages to show up in my menus, regardless of whether a user is logged in or not. When the user hits the page, I want to redirect to my loginbox, and then redirect them back to the page they were on after they log in. However, I just can't figure out a way to display pages regardless of their access. I've tried TypoScript and also hacking some of the core engine code, to no avail. I feel like this is something I need to hack in the core, which is why I'm posting to the dev list. Any suggestions would be greatly appreciated. Thanks, Keith From keith_morrison at dfci.harvard.edu Fri Oct 7 00:06:33 2005 From: keith_morrison at dfci.harvard.edu (Keith Morrison) Date: Thu, 06 Oct 2005 18:06:33 -0400 Subject: [Typo3-dev] Display page in menu, regardless of login status. In-Reply-To: References: Message-ID: Keith Morrison wrote: > Hi all, > > I'm having the hardest time finding a solution for this. I have a lot > of pages that are only viewable if you are a logged-in user; however, I > want those pages to show up in my menus, regardless of whether a user is > logged in or not. When the user hits the page, I want to redirect to my > loginbox, and then redirect them back to the page they were on after > they log in. However, I just can't figure out a way to display pages > regardless of their access. I've tried TypoScript and also hacking some > of the core engine code, to no avail. I feel like this is something I > need to hack in the core, which is why I'm posting to the dev list. Any > suggestions would be greatly appreciated. > > Thanks, > Keith I apologize for responding to myself, but I did find a note in the CVS log from 2005-4-21 by Kasper that seems to describe the ability to do this, but I can't find any follow up information on it. The revision note is on this page: http://cvs.sourceforge.net/viewcvs.py/typo3/TYPO3core/ChangeLog?rev=1.258 Thanks again, Keith From roger at redgumsoaps.com.au Fri Oct 7 00:12:29 2005 From: roger at redgumsoaps.com.au (Roger Bunyan) Date: Fri, 07 Oct 2005 08:12:29 +1000 Subject: [Typo3-dev] Display page in menu, regardless of login status. In-Reply-To: References: Message-ID: Keith Morrison wrote: >Hi all, > >I'm having the hardest time finding a solution for this. I have a lot >of pages that are only viewable if you are a logged-in user; however, I >want those pages to show up in my menus, regardless of whether a user is >logged in or not. > This might sound simple but why don't you set the page access to everyone but the content elements to the logged in usergroup. Put a login box with the setting "hide at login". -- Roger Bunyan Redgum Soaps http://redgumsoaps.com.au Sustainability will be achieved by a cultural change, not by technology. From typo3 at rvt.dds.nl Fri Oct 7 01:24:30 2005 From: typo3 at rvt.dds.nl (Ries van Twisk) Date: Thu, 06 Oct 2005 17:24:30 -0600 Subject: [Typo3-dev] using sys_history Message-ID: Dear list, is the a class set or functions to use the sys_history tables? I need to record changes in my database that are done in my extentions. cheers, Ries From elmar.hinz at vcd-berlin.de Fri Oct 7 00:34:06 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Fri, 07 Oct 2005 00:34:06 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References:

Message-ID: Elmar Hinz wrote: > Kasper Sk?rh?j wrote: > >>> Wouldn't it also happen when >>> >>> a) ... >>> b) ... >>> c) a cHash is given but a wrong one >> >> >> >> Nope, it will not. The problem as described in that article is >> specifically dealing with an empty value of cHash. A value found will >> trigger a verification process that afaik works just fine. >> >> - kasper > > > Hello Kasper, > > Thank you for precise ansewer. I will reflect about this point. > O.k. I got the point now. If an invalid cHash is discoverd, caching is disabled. The content can still be delivered dynamically but it isn't cached. In case the enemy had changed ?&tx_myext[uid]=9876554321? in order to spam our cache table the resulting hash string generated will with all probability not match ?13b5d6efa7?: Page caching will be disabled immediately, but the page will still be generated trying to display news item 9876554321 (which probably does not exist, but so what? The cache table is not loaded with that...) Now for delivery (not for caching): What happens if: a) the additional parameters are sensible b) the according view is already cachecd c) but the cHash is empty or wrong (because of a defect external link) 1.) Is the view created dynamically? 2.) Is it deliverd from the cache? Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From elmar.hinz at vcd-berlin.de Fri Oct 7 01:09:18 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Fri, 07 Oct 2005 01:09:18 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References:

Message-ID: > >>Either this is not precise enough or tells something else then the >>article. There isn't a problem when the cache is already filled in the >>article. The problem is filling the empty cache with the right stuff. > > > If the cache is empty, it will be filled by the first attempt to request > the given page. The problem is if we fill the cache, but the content is > not really meant to be cached for that combination of parameters. In the > article, this results in: > There we met. That's what I said. > checked (if present, TYPO3 will check it on its own, if not present or > empty, TYPO3 needs the plugins to inform if a checking is needed, which > is what this section of the article is about). > > It is informed by setting the useCacheHash paramter in the plugin. Still I don't make friendship with this concept. Programmers could forget to set it. Why can't it be done automatically for USER-objects? First checking if there are more parameters then type and id. And if so requireing a valid cHash before any saving action takes part, to say it shortend. I know there could be some other parameters around like the L-Parameter for languages. But that information could be fetched from TypoScript Setup. > > >>That is exactly my thesis. The effect would be the same. Even if there >>isn't an empty cHash but an invalid one. > > > Right, just that with invalid ones the plugin writer won't even be > bothered with it, which is not the case if there is no, or an empty > cHash value. > And why? Because if a wrong cHash is discoverd, caching is disabled at all. That point I missed. I imagined parametes would be thrown away, so that it would behave like no parameters at all. Best regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From arnsholt at broadpark.no Fri Oct 7 03:07:17 2005 From: arnsholt at broadpark.no (Arne Skjaerholt) Date: Fri, 07 Oct 2005 03:07:17 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: > For example I have this list in a record: > 345,6,17,231 > How would I do a like query on this one to look for number for UID 1? Now that you mention it, I remember that I didn't use LIKE, LIKE couldn't do that particular task, so I used REGEXP instead (I think that's a MySQL extension tho'). I just looked at my code, and here's how I did it: SELECT field FROM table WHERE comma_field REGEXP '(^|,)$uid(,|$)' It's kinda ugly, but it does the job. > Also, a query like this: LIKE ('%somedata%') cannot be optimized at the > DB level > because the RDBM cannot do a index scan but rather have to do a > tablescan, means has to scan ALL records! > > I would love to see MM relations, > unfortunatly I don't think that's possible anymore with typo3.... > > I have seen however that Kasper did some speed measurements and found > that on mysql there is > hardly no speed differences and so decided not to go for a relational DB > setup. The speed difference being negligeable is good at least, but there are two reasons I don't really like the current solution: 1) It's poor DB design. In fact, it's the kind of DB design I'd expect from "home-brew" software, not professional grade software. 2) Certain operations that could be done with a JOIN in a proper MM-relation enviroment are simply impossible to do without first selecting the usergroups field of a user (to take the example I worked with), splitting the string into an array, and then executing a second query to get the data you were looking for. Arne :wq From mathias.schreiber at wmdb.de Fri Oct 7 08:38:26 2005 From: mathias.schreiber at wmdb.de (Mathias Schreiber [wmdb]) Date: Fri, 07 Oct 2005 08:38:26 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: Peter Niederlag wrote: > Top prio would be imho to assure that any new tables, fields introduced > are using true MM-relations. That is my main concern. > > Second prio should be a good analysis if there is a need or wish to > change old structure to new design in Version 5. > > To strengthen the point again: > Imho we should change Kickstarter's default-behaviour *now* (read: as > soon as possible) and take any other measures any time later. Do you plan a structure like this: pages be_users mm_pages_beusers for each relation or do you plan something like one big mm table for all relations? t3_mm ----------------------------------------------------- |mm_field_1|mm_field_2|mm_table_1|mm_table_2|sorting| ----------------------------------------------------- I have to admin that personally I like databases with less tables better. Indexing should work with the "single-table-config". You keep "sorting" in mind? peace Mathias -- if ($GLOBALS['TSFE']->fe_user->user['ahnung'] == 0) { $this->fresseHalten = 1; } From mathias.schreiber at wmdb.de Fri Oct 7 08:39:32 2005 From: mathias.schreiber at wmdb.de (Mathias Schreiber [wmdb]) Date: Fri, 07 Oct 2005 08:39:32 +0200 Subject: [Typo3-dev] Use T3 Database records for external applications In-Reply-To: References: Message-ID: Stefan Beylen wrote: > Hello list, > > I want to use some tables filled with records from my T3 installation in > an external PHP application (non-t3) > > Problem that I have is that t3 uses special tags such as 4ex instead of getPageLink From there follow the path to (I think) tslib_cobj->typolink and check out what Agent K did there. -- if ($GLOBALS['TSFE']->fe_user->user['ahnung'] == 0) { $this->fresseHalten = 1; } From jer at moccompany.com Fri Oct 7 09:08:24 2005 From: jer at moccompany.com (Jan-Erik Revsbech) Date: Fri, 7 Oct 2005 09:08:24 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations References:

Message-ID: >> How would I do a like query on this one to look for number for UID 1? > > Now that you mention it, I remember that I didn't use LIKE, LIKE > couldn't do that particular task, That is not true. Making af call to $GLOBALS['TYPO3_DB']->listQuery($field,$value,$table) will return you ($field LIKE "%$value,%" OR $field LIKE "%$,value%" OR $field =$value) In the above case ($field LIKE "%1,%" OR $field LIKE "%$,1%" OR $field =1) which will do the trick. And it is up to DBAL($GLBOALS['TYPO3_DB']) to make WHERE clause that will work on a particular database. Alternativley you could use the MySQL function FIND_IN_SET which alow works, but is not SQL standard. I do agree however, that MM tables are much better. In a specific example we had 22.000 BE usergroups nested into eachother (child-parent relation is a comma list in table be_groups), and it was very clear that all queries on the be_groups table were significantly slowing down the system. We changed to mm relation instead and now everything worked fine. I agree that we should try to avoid have to many comma list in the core, but I agree that is is a big task. /Jan-Erik From thorsten.kahler at dkd.de Fri Oct 7 10:04:02 2005 From: thorsten.kahler at dkd.de (Thorsten Kahler) Date: Fri, 07 Oct 2005 10:04:02 +0200 Subject: [Typo3-dev] Use T3 Database records for external applications In-Reply-To: References:

Message-ID: Hi Mattes, I think Agent K introduced tslib_cObj::parseFunc(), didn't he? Regards Thorsten Mathias Schreiber [wmdb] wrote: > Stefan Beylen wrote: >> Hello list, >> >> I want to use some tables filled with records from my T3 installation >> in an external PHP application (non-t3) >> >> Problem that I have is that t3 uses special tags such as 4ex > instead of

> take a look as tslib_pibase->getPageLink > From there follow the path to (I think) tslib_cobj->typolink and check > out what Agent K did there. > From intsys at swissinfo.org Fri Oct 7 10:12:59 2005 From: intsys at swissinfo.org (Stefan Beylen) Date: Fri, 07 Oct 2005 10:12:59 +0200 Subject: [Typo3-dev] Use T3 Database records for external applications In-Reply-To: References:

Message-ID: Thorsten Kahler wrote: > I think Agent K introduced tslib_cObj::parseFunc(), didn't he? ok, thanks...I know that parseFunc is used...anyway, its kinda overkill for what I need it for though...a few simple str_replaces/regexps also do the job... ;) Thx for your help though, I was hoping for a more simple solution :) ...not needed anymore as I am already almost finished... thx! From niederlag at ikd01.de Fri Oct 7 10:18:22 2005 From: niederlag at ikd01.de (Peter Niederlag) Date: Fri, 07 Oct 2005 10:18:22 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: Hi, Mathias Schreiber [wmdb] schrieb: > Peter Niederlag wrote: > >> Top prio would be imho to assure that any new tables, fields introduced >> are using true MM-relations. That is my main concern. >> >> Second prio should be a good analysis if there is a need or wish to >> change old structure to new design in Version 5. >> >> To strengthen the point again: >> Imho we should change Kickstarter's default-behaviour *now* (read: as >> soon as possible) and take any other measures any time later. > > > Do you plan a structure like this: > pages > be_users > mm_pages_beusers > > for each relation or do you plan something like one big mm table for all > relations? IMHO it might be too early now to pin down a solution. But this ^^ approach can be handled already pretty good by TYPO3-core imho. > t3_mm > ----------------------------------------------------- > |mm_field_1|mm_field_2|mm_table_1|mm_table_2|sorting| > ----------------------------------------------------- > > I have to admin that personally I like databases with less tables better. > Indexing should work with the "single-table-config". >From the feedback I think there are a bunch of people that like to see TYPO3 follow the guidelines of well known DB-Design matters. > You keep "sorting" in mind? As said, it might be to early now to talk about any changes that we want to see in Version 5. Right now I think we got some awareness for the problem, and also some good points, *why* it is quite important in the long-time. So, go for true M-M with anything newly introduced and change default of Kickstarter *now*. Everything else we will see when time is appropriate. My 2 cents, Peter -- Peter Niederlag http://www.niekom.de * TYPO3 & EDV Dienstleistungen * http://www.typo3partner.net * professional services network * From kasper2005 at typo3.com Fri Oct 7 10:32:46 2005 From: kasper2005 at typo3.com (Kasper =?iso-8859-1?q?Sk=E5rh=F8j?=) Date: Fri, 7 Oct 2005 10:32:46 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: > 1) It's poor DB design. In fact, it's the kind of DB design I'd expect > from "home-brew" software, not professional grade software. It's a legacy solution from a time when we didn't have tools to manage easily the tables in the database. However it has prevailed because in many cases it is actually a quite handy and pragmatic way of getting things done, beautiful or not. Also figure; We introduced MM-tables for exactly the reason that we could use both methods (but this was post-be/fe-users which is why THEY still carry the old way with them). - k From kasper2005 at typo3.com Fri Oct 7 10:40:54 2005 From: kasper2005 at typo3.com (Kasper =?utf-8?q?Sk=C3=A5rh=C3=B8j?=) Date: Fri, 7 Oct 2005 10:40:54 +0200 Subject: [Typo3-dev] Display page in menu, regardless of login status. In-Reply-To: References:

Message-ID: Hi Keith, If this is not already in TSref I apologize that I didn't publish it yet - will do so today maybe. For the menu objects (eg. TMENU/GMENU) you have this property to set in TypoScript, that is what I referred to 21/4 this year: showAccessRestrictedPages integer (page id) / keyword ???NONE??? If set, pages in the menu will include pages with frontend user group access enabled. However the page is of course not accessible and therefore the URL in the menu will be linked to the page with the ID of this value. On that page you could put a login form or other message. If the value is ???NONE??? the link will not be changed and the site will perform page-not-found handling when clicked (which can be used to capture the event and act accordingly of course). Properties: .addParam = Additional parameter for the URL, which can hold two markers; ###RETURN_URL### which will be substituted with the link the page would have had if it had been accessible and ###PAGE_ID### holding the page id of the page coming from (could be used to look up which fe_groups was required for access. Example: showAccessRestrictedPages = 22 showAccessRestrictedPages.addParams = &return_url=###RETURN_URL###&pageId=###PAGE_ID### The example will link access restricted menu items to page id 22 with the return URL in the GET var ???return_url??? and the page id in the GET var ???pageId???. k On Friday 07 October 2005 00:06, Keith Morrison wrote: > Keith Morrison wrote: > > Hi all, > > > > I'm having the hardest time finding a solution for this. I have a lot > > of pages that are only viewable if you are a logged-in user; however, I > > want those pages to show up in my menus, regardless of whether a user is > > logged in or not. When the user hits the page, I want to redirect to my > > loginbox, and then redirect them back to the page they were on after > > they log in. However, I just can't figure out a way to display pages > > regardless of their access. I've tried TypoScript and also hacking some > > of the core engine code, to no avail. I feel like this is something I > > need to hack in the core, which is why I'm posting to the dev list. Any > > suggestions would be greatly appreciated. > > > > Thanks, > > Keith > > I apologize for responding to myself, but I did find a note in the CVS > log from 2005-4-21 by Kasper that seems to describe the ability to do > this, but I can't find any follow up information on it. The revision > note is on this page: > http://cvs.sourceforge.net/viewcvs.py/typo3/TYPO3core/ChangeLog?rev=1.258 > > Thanks again, > Keith > _______________________________________________ > Typo3-dev mailing list > Typo3-dev at lists.netfielders.de > http://lists.netfielders.de/cgi-bin/mailman/listinfo/typo3-dev -- - kasper ----------------- Think future, not feature From intsys at swissinfo.org Fri Oct 7 10:56:54 2005 From: intsys at swissinfo.org (Stefan Beylen) Date: Fri, 07 Oct 2005 10:56:54 +0200 Subject: [Typo3-dev] Use T3 Database records for external applications In-Reply-To: References:

Message-ID: Stefan Beylen wrote: > ...not needed anymore as I am already almost finished... maybe it is useful for someone else... p.s.: I disallowed my editors all other options in the RTE, so in my case this is all thats needed ;) p.p.s: I am no Regexp-ert (as probably most can see) ---8<------ function RTEtransform($text){ if (substr($text,4)!="

") $text="

".$text; $text=str_replace("\n","

",$text); $text=str_replace("\r","

",$text); $text=str_replace("

",$text); $text=str_replace("","",$text); $text=str_replace("","",$text); $text=str_replace("","",$text); $text=str_replace("","",$text); //mailtos $text=preg_replace("/\/", "

", $text); //external target $text=preg_replace("/\/", "

", $text); //internal target $text=preg_replace("/\/", "

", $text); $text=str_replace("","

") $text=$text."

"; return $text; } ---8<------ From elmar.hinz at vcd-berlin.de Fri Oct 7 11:00:08 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Fri, 07 Oct 2005 11:00:08 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: > > > From the feedback I think there are a bunch of people that like to see > TYPO3 follow the guidelines of well known DB-Design matters. > > I am for clean patterns, too. Most variations from it, that seem usefull on first glance tend to produce problems at last. On the long run there should also be some refacturing on tt_content, I think, wich somehow contains several entities in one table. Maybe for Version 8.3.7. Regards Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From elmar.hinz at vcd-berlin.de Fri Oct 7 11:25:51 2005 From: elmar.hinz at vcd-berlin.de (Elmar Hinz) Date: Fri, 07 Oct 2005 11:25:51 +0200 Subject: [Typo3-dev] Empty cHash problem In-Reply-To: References:

Message-ID: > > >>By the way. Is there a special table for caching views with additional >>parameters or is it the same table as for the old id + type combination? >>Which tables are they? Wich fields in the tables? > > > Its the same table, as the cache already uses different combination of > parameters for caching the same page (e.g for different types). The > table is cache_pages, the content is in the HTML field. You will have a > huge configuration array in cache_data, which unserialized will also > tell you (amongst other) the piVar-parameters that were used to create > this cache request. But what TYPO3 uses to see which record to get is > the md5-hash of a series of information (fe_group, parameters, type, > language, etc), which will be compared against the "hash" field. > > Thank you very much. We find some other cache tables: cache_pages : Our cached content cache_hash : ??? cache_imagesizes: caches imagesizes, so they don't need to be remeasured for html output cache_md5params: ??? cache_pagesection: ??? Elmar -- Climate change 2005: New Orleans, Sahel, Bangladesh, Spain, Portugal, Austria, Swiss, France, ... Production of CO2 is killing people. Production of CO2 just for fun is killing people just for fun. From traveler_in_time at gmx.net Fri Oct 7 12:25:52 2005 From: traveler_in_time at gmx.net (S. Teuber) Date: Fri, 7 Oct 2005 10:25:52 +0000 (UTC) Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations References:

Message-ID: Hi Peter, just for feedback: I agree with you in all points in this matter (kinda "global +1" ;-). IMO the database structure needs even more improvements, for example keeping all those extension-added fields out of tt_content and putting them in an extra table (tt_content_ext), because tt_content gets kinda HUGE and difficult to survey if you have a large site which uses many extensions. Plus, you get all the overhead from the additional fields even on (most of the times the majority of) pages that don't use any extension, which affects performance. Of course, not only the database, but TCEForms as well needed to be adapted to handle the more complex table structures, so this is a long time task for version 7 or 8 at best... Frankly, I missed the ability to edit related tables in *one* TCEForm in quite a few projects, always ending up by making the database design worse to fit TYPO3's needs. I'm not alone with this need; we had some threads about this in Typo3-dev recently, as you might remember, i.e. the "'virtual tables' with tca?"-thread in May/June. Sven From Martin.Kutschker at n0spam-blackbox.net Fri Oct 7 13:37:37 2005 From: Martin.Kutschker at n0spam-blackbox.net (Martin Kutschker) Date: Fri, 07 Oct 2005 13:37:37 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: S. Teuber schrieb: > Hi Peter, > > just for feedback: I agree with you in all points in this matter (kinda > "global +1" ;-). > > IMO the database structure needs even more improvements, for example > keeping all those extension-added fields out of tt_content and putting them > in an extra table (tt_content_ext), because tt_content gets kinda HUGE and > difficult to survey if you have a large site which uses many extensions. You would either clutter up tt_content_ext or would get lots of additional tt_content_ext_* tables. The former would give you no performance gain, because the table would have to joined with tt_content for normal content rendering. The latter would mean an even greater perfomance loss as you would have to make a new SELECT for each content element. Use flexforms if you need more fields if you're worried about the number of fields in tt_content. Unfortunately flexforms are flexible but don't work well in a Typoscript context. And they cannot be configured by page TS. :-( Still, they are a useful and powerful option for plugins. We should IMHO rather think about improving them, than about addin new tables. Masi From niederlag at ikd01.de Fri Oct 7 14:02:58 2005 From: niederlag at ikd01.de (Peter Niederlag) Date: Fri, 07 Oct 2005 14:02:58 +0200 Subject: [Typo3-dev] Flexforms and Typoscript [was: Re: RFC: TYPO3, Kickstarter and M-M database relations] In-Reply-To: References:

Message-ID: Hi, Martin Kutschker schrieb: [...] > Unfortunately flexforms are flexible but don't work well in a Typoscript > context. And they cannot be configured by page TS. :-( I my flexform pi's I do "flatten" all flexform-fields and values and add them to $this->cObject->data. That way you can easily use the values in Typoscript in a pretty flexible way. More details or an example can be obtained by PM. Cheers, Peter -- Peter Niederlag http://www.niekom.de * TYPO3 & EDV Dienstleistungen * http://www.typo3partner.net * professional services network * From Martin.Kutschker at n0spam-blackbox.net Fri Oct 7 14:53:34 2005 From: Martin.Kutschker at n0spam-blackbox.net (Martin Kutschker) Date: Fri, 07 Oct 2005 14:53:34 +0200 Subject: [Typo3-dev] Flexforms and Typoscript [was: Re: RFC: TYPO3, Kickstarter and M-M database relations] In-Reply-To: References:

Message-ID: Peter Niederlag schrieb: > Hi, > > Martin Kutschker schrieb: > [...] > >>Unfortunately flexforms are flexible but don't work well in a Typoscript >>context. And they cannot be configured by page TS. :-( > > > I my flexform pi's I do "flatten" all flexform-fields and values and add > them to $this->cObject->data. Well, there's even an API for that (in pi_base). I know and am quite happy with it. > That way you can easily use the values in Typoscript in a pretty > flexible way. Sure. I said nothing about using them in plugins. There they are great and their only drawback is the performance issue due xml parsing. The problems are elsewhere: You cannot use their contents in TS, eg. getData worls only on real/record fields. And you cannot remove them from the UI in the BE via page TS. You cannot set defaults via TS etc. Masi From traveler_in_time at gmx.net Fri Oct 7 15:00:00 2005 From: traveler_in_time at gmx.net (S. Teuber) Date: Fri, 7 Oct 2005 13:00:00 +0000 (UTC) Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations References:

Message-ID: Hi Martin, > You would either clutter up tt_content_ext or would get lots of > additional tt_content_ext_* tables. > > The former would give you no performance gain, because the table would > have to joined with tt_content for normal content rendering. Not if there is some "logic" that determines if additional extension fields are needed at all. But then we'd propably have to distinguish extensions that add something to a regular content element (except "insert plugin") from those that add completely new, own elements... I see that the whole issue needs some more fundamental thinking, but I'd rather not drop it lightly. > Use flexforms if you need more fields if you're worried about the > number of fields in tt_content. It's not about what *I* do, since most of the frequently used extensions aren't maintained by me - schade eigentlich ;-) Besides, I dislike the fact that a flexform field needs to be parsed to get to the actual data. If I got, say, 4 "items" like "view", "sort by", "sort order", "categories" (yes, these are taken from tt_news ;), I'd rather store them in 4 fields than in one field using XML - for easy access without the need for an XML parser. After all, TYPO3 may not be the only application using the database structure. Gotta admit that flexforms are really cool in the TCEforms, though. ;) Sven From mathias.schreiber at wmdb.de Fri Oct 7 15:05:22 2005 From: mathias.schreiber at wmdb.de (Mathias Schreiber [wmdb]) Date: Fri, 07 Oct 2005 15:05:22 +0200 Subject: [Typo3-dev] RFC: TYPO3, Kickstarter and M-M database relations In-Reply-To: References:

Message-ID: S. Teuber wrote: > After all, TYPO3 may not be the only application > using the database structure. BLASPHEMIE!!! ;-) -- if ($GLOBALS['TSFE']->fe_user->user['ahnung'] == 0) { $this->fresseHalten = 1; } From ernst at cron-it.de Fri Oct 7 15:15:39 2005 From: ernst at cron-it.de (Ernesto Baschny [cron IT]) Date: Fri, 07 Oct 2005 15:15:39 +0200 Subject: [Typo3-dev] cache tables (was: Re: Empty cHash problem) In-Reply-To: