From lstephan at w-seils.com Tue Oct 4 12:29:42 2016 From: lstephan at w-seils.com (=?iso-8859-1?Q?Lo=EFc_St=E9phan?=) Date: Tue, 4 Oct 2016 12:29:42 +0200 Subject: [TYPO3-Solr] Indexing of shortcuts Message-ID: Hello, I try to index pages which are shortcut but It doesn?t seem to work. I use allowedPageTypes = 1,7 to add shortcut doktype. Pages which are shortcuts, are added to the queue (when a page is edited, scheduler show 99,99% indexed instead of 100%) but the page is not indexed in Solr. The field ? indexed ? is filled in ? tx_solr_indexqueue_item ? with timestamp. Have you any idea ? Is it possible to index shortcut ? TYPO3 ? 6.2.27 solr 3.0.2 Thank you, Lo?c From olivier.dobberkau at dkd.de Tue Oct 4 18:59:31 2016 From: olivier.dobberkau at dkd.de (Olivier Dobberkau) Date: Tue, 4 Oct 2016 18:59:31 +0200 Subject: [TYPO3-Solr] Indexing of shortcuts In-Reply-To: References: Message-ID: On 04.10.16 12:29, Lo?c St?phan wrote: > Is it possible to index shortcut ? it would expect that we index the content of the page to which the short cut is pointing to. Olivier From lstephan at w-seils.com Wed Oct 5 09:33:09 2016 From: lstephan at w-seils.com (=?iso-8859-1?Q?Lo=EFc_St=E9phan?=) Date: Wed, 5 Oct 2016 09:33:09 +0200 Subject: [TYPO3-Solr] Indexing of shortcuts In-Reply-To: References: Message-ID: > Is it possible to index shortcut ? > > it would expect that we index the content of the page to which the short cut is pointing to. Yes, the content of the page which the shortcut is pointing to is indexed. This is a standard page. ? Lo?c -----Message d'origine----- De?: typo3-project-solr-bounces at lists.typo3.org [mailto:typo3-project-solr-bounces at lists.typo3.org] De la part de Olivier Dobberkau Envoy??: mardi 4 octobre 2016 19:00 ??: typo3-project-solr at lists.typo3.org Objet?: Re: [TYPO3-Solr] Indexing of shortcuts On 04.10.16 12:29, Lo?c St?phan wrote: > Is it possible to index shortcut ? it would expect that we index the content of the page to which the short cut is pointing to. Olivier _______________________________________________ TYPO3-project-solr mailing list TYPO3-project-solr at lists.typo3.org http://lists.typo3.org/cgi-bin/mailman/listinfo/typo3-project-solr From birte.weishaupt at uni-bremen.de Fri Oct 7 09:58:40 2016 From: birte.weishaupt at uni-bremen.de (Birte Weishaupt) Date: Fri, 7 Oct 2016 09:58:40 +0200 Subject: [TYPO3-Solr] HTML in indexing fields Message-ID: Hi, when indexing pages I have a field where I store the page path. This works fine, except for one thing. The breadcrumbs contain html tags for the links and for some reason the entities in the string get encoded. So instead of I end up with <a href="" >. I already tried several things to get rid of the entity encoding but without success. When I set this up it used to work but something must have changed in the meantime. Here is my TS: lib.solrBreadcrumb = HMENU lib.solrBreadcrumb { special = rootline special.range = 0|-1 includeNotInMenu = 1 1 = TMENU 1.noBlur = 1 1.CUR = 1 1.target = _self 1.NO { stdWrap.field = title linkWrap = ||*| / |*| } 1.CUR { stdWrap.field = title linkWrap = ||*| / |*| doNotLinkIt = 1 } } plugin.tx_solr { index { queue { pages { fields { breadcrumb_stringS < lib.solrBreadcrumb .. with typo3 6.2.27 solr 3.1.2 solrfal 2.1.3 tika 2.0.0 Does anyone have an idea? Thanks a lot :) Regards, Birte From jigal.van.hemert at typo3.org Mon Oct 10 23:08:56 2016 From: jigal.van.hemert at typo3.org (Jigal van Hemert) Date: Mon, 10 Oct 2016 23:08:56 +0200 Subject: [TYPO3-Solr] HTML in indexing fields In-Reply-To: References: Message-ID: Hi, On 07/10/2016 09:58, Birte Weishaupt wrote: > when indexing pages I have a field where I store the page path. This works > fine, except for one thing. The breadcrumbs contain html tags for the links > and for some reason the entities in the string get encoded. So instead of href=""> I end up with <a href="" >. I already tried several things to > get rid of the entity encoding but without success. When I set this up it > used to work but something must have changed in the meantime. https://github.com/TYPO3-Solr/ext-solr/releases/tag/3.1.0 HTML escaping after retrieval from solr The data ist now esapced right after the retrieval from Solr. In rare cases when you need to store HTML in Solr documents you can use the following configuration to mark these fields as trusted fields and skip the escaping: plugin.tx_solr.search.trustedFields = my_first_html_field,my_second_html_field -- Jigal van Hemert TYPO3 CMS Active Contributor TYPO3 .... inspiring people to share! Get involved: typo3.org From Tino.Riedel at gmx.de Fri Oct 14 16:20:57 2016 From: Tino.Riedel at gmx.de (Tino Riedel) Date: Fri, 14 Oct 2016 16:20:57 +0200 Subject: [TYPO3-Solr] Scheduler doesn't index new tx_news records Message-ID: Hi, im using tx_solr 3.1.2 and TYPO3 6.2.26. The scheduler indexes pages, tx_news and some custom stuff. A Task "Index Queue Worker" runs every night. So new, modified and deletet pages will indexed correctly. But no new tx_news records will indexed untill i force reindexing the site. I have no idea what I can still do. Best regards Tino From jigal.van.hemert at typo3.org Sat Oct 15 20:13:33 2016 From: jigal.van.hemert at typo3.org (Jigal van Hemert) Date: Sat, 15 Oct 2016 20:13:33 +0200 Subject: [TYPO3-Solr] Scheduler doesn't index new tx_news records In-Reply-To: References: Message-ID: Hi, On 14/10/2016 16:20, Tino Riedel wrote: > The scheduler indexes pages, tx_news and some custom stuff. > A Task "Index Queue Worker" runs every night. So new, modified and deletet pages will indexed correctly. > But no new tx_news records will indexed untill i force reindexing the site. If you add/modify new records in the TYPO3 backend they should be indexed the next run of the scheduler task. The backend uses the class DataHandler to insert/modify records and EXT:solr hooks into that class to detect new and changed records. If you use an Extbase plugin to add news records in the frontend this won't work as extbase persistence doesn't use DataHandler. If you use some extension to import new news records it depends on the way the extension sends data to the database. You can always check in the table tx_solr_indexqueue_item if there is a record for your new news record. If the scheduler task for solr indexing isn't at 100% and won't increase anymore on new runs there is a problem in the indexing queue. It could be that there are items with errors (investigate the error and decide if the record should be indexed at all) or there can be leftovers from removed index queue definitions. -- Jigal van Hemert TYPO3 CMS Active Contributor TYPO3 .... inspiring people to share! Get involved: typo3.org From tino.riedel at gmx.de Mon Oct 17 13:55:44 2016 From: tino.riedel at gmx.de (Tino Riedel) Date: Mon, 17 Oct 2016 13:55:44 +0200 Subject: [TYPO3-Solr] Scheduler doesn't index new tx_news records In-Reply-To: References: Message-ID: Hi, i have checked the tx_solr_indexqueue_item table and found records for the new tx_news entries. But the field "indexed" was 0. The process in the scheduler had never index the new items. The reason for that will be the id in the column root. My tree is like 5000-homepage (site root) |_5001-site1 |_5002-site2 |_5003-site3 |_... |_6000-blog.homepage (also a site-root) |_6001-site1 |_6002-site2 |_6003-site3 |_... |_6010-sysfolder for tx_news The task in the scheduler for indexing is configured for den root with id 5000. The hook save new tx_news items to the table tx_solr_indexqueue_item with id 6000. So the task does't hit the entries i think? My aim is that the task indexes all entries, also the new tx_news entries below page with id 6000. As mentioned in the wiki you have to add queue.[indexConfig].additionalPageIds if there is data outside the side-root (in my example 5000), is't it? I think queuing stops at the next site-root? Best regards Tino Am 15.10.2016 um 20:13 schrieb Jigal van Hemert: > Hi, > > On 14/10/2016 16:20, Tino Riedel wrote: >> The scheduler indexes pages, tx_news and some custom stuff. >> A Task "Index Queue Worker" runs every night. So new, modified and >> deletet pages will indexed correctly. >> But no new tx_news records will indexed untill i force reindexing the >> site. > > If you add/modify new records in the TYPO3 backend they should be > indexed the next run of the scheduler task. The backend uses the class > DataHandler to insert/modify records and EXT:solr hooks into that class > to detect new and changed records. > If you use an Extbase plugin to add news records in the frontend this > won't work as extbase persistence doesn't use DataHandler. If you use > some extension to import new news records it depends on the way the > extension sends data to the database. > > You can always check in the table tx_solr_indexqueue_item if there is a > record for your new news record. > If the scheduler task for solr indexing isn't at 100% and won't increase > anymore on new runs there is a problem in the indexing queue. It could > be that there are items with errors (investigate the error and decide if > the record should be indexed at all) or there can be leftovers from > removed index queue definitions. > From jigal.van.hemert at typo3.org Mon Oct 17 21:37:32 2016 From: jigal.van.hemert at typo3.org (Jigal van Hemert) Date: Mon, 17 Oct 2016 21:37:32 +0200 Subject: [TYPO3-Solr] Scheduler doesn't index new tx_news records In-Reply-To: References: Message-ID: Hi, On 17/10/2016 13:55, Tino Riedel wrote: > 5000-homepage (site root) > |_5001-site1 > |_5002-site2 > |_5003-site3 > |_... > |_6000-blog.homepage (also a site-root) > |_6001-site1 > |_6002-site2 > |_6003-site3 > |_... > |_6010-sysfolder for tx_news > > The task in the scheduler for indexing is configured for den root with > id 5000. The hook save new tx_news items to the table > tx_solr_indexqueue_item with id 6000. So the task does't hit the entries > i think? Nested site roots are never a good idea. At first things seem to work, but sooner or later you'll run into an extension or a construction in your own configuration that doesn't work with nested roots. There are solutions to make things work as if there were nested roots. You can use mount points or create shortcut page to make blog.homepage appear in the menu next to 5001-site1, 5002-site2, etc. > My aim is that the task indexes all entries, also the new tx_news > entries below page with id 6000. As mentioned in the wiki you have to > add queue.[indexConfig].additionalPageIds if there is data outside the > side-root (in my example 5000), is't it? Now you see the ambiguous situation that was created. Folder 6010 is inside both roots but it probably considers 6000 as the root because it's the first page it encounters in the rootline that is marked as root. AdditionalPageIds are probably necessary if you want to have the news records indexed with root page 5000. -- Jigal van Hemert TYPO3 CMS Active Contributor TYPO3 .... inspiring people to share! Get involved: typo3.org