CREATE ALGORITHM=UNDEFINED DEFINER=`root`@`%` SQL SECURITY DEFINER VIEW `entries` AS select `mtosblogs`.`mt_entry`.`entry_id` AS `id`,`mtosblogs`.`mt_entry`.`entry_allow_comments` AS `allow_comments`,`mtosblogs`.`mt_entry`.`entry_allow_pings` AS `allow_pings`,`mtosblogs`.`mt_entry`.`entry_atom_id` AS `atom_id`,`mtosblogs`.`mt_entry`.`entry_author_id` AS `author_id`,`mtosblogs`.`mt_entry`.`entry_authored_on` AS `authored_on`,`mtosblogs`.`mt_entry`.`entry_basename` AS `basename`,`mtosblogs`.`mt_entry`.`entry_blog_id` AS `blog_id`,`mtosblogs`.`mt_entry`.`entry_category_id` AS `category_id`,`mtosblogs`.`mt_entry`.`entry_class` AS `class`,`mtosblogs`.`mt_entry`.`entry_comment_count` AS `comment_count`,`mtosblogs`.`mt_entry`.`entry_convert_breaks` AS `convert_breaks`,`mtosblogs`.`mt_entry`.`entry_created_by` AS `created_by`,`mtosblogs`.`mt_entry`.`entry_created_on` AS `created_on`,`mtosblogs`.`mt_entry`.`entry_excerpt` AS `excerpt`,`mtosblogs`.`mt_entry`.`entry_keywords` AS `keywords`,`mtosblogs`.`mt_entry`.`entry_modified_by` AS `modified_by`,`mtosblogs`.`mt_entry`.`entry_modified_on` AS `modified_on`,`mtosblogs`.`mt_entry`.`entry_ping_count` AS `ping_count`,`mtosblogs`.`mt_entry`.`entry_pinged_urls` AS `pinged_urls`,`mtosblogs`.`mt_entry`.`entry_status` AS `status`,`mtosblogs`.`mt_entry`.`entry_tangent_cache` AS `tangent_cache`,`mtosblogs`.`mt_entry`.`entry_template_id` AS `template_id`,`mtosblogs`.`mt_entry`.`entry_text` AS `text`,`mtosblogs`.`mt_entry`.`entry_text_more` AS `text_more`,`mtosblogs`.`mt_entry`.`entry_title` AS `title`,`mtosblogs`.`mt_entry`.`entry_to_ping_urls` AS `to_ping_urls`,`mtosblogs`.`mt_entry`.`entry_week_number` AS `week_number`,`mtosblogs`.`mt_entry`.`entry_visit_count` AS `visit_count` from `mtosblogs`.`mt_entry`;
I’m hoping this will work.
id:integer allow_comments:boolean allow_pings:boolean atom_id:string author_id:integer authored_on:date basename:string blog_id:integer category_id:integer class:string comment_count:integer convert_breaks:string created_by:integer created_on:date excerpt:text keywords:string modified_by:integer modified_on:date ping_count:integer pinged_urls:integer status:integer tangent_cache:boolean template_id:integer text:text text_more:text title:string to_ping_urls:boolean week_number:integer visit_count:integer id:integer allow_anon_comments:boolean allow_comment_html:boolean allow_commenter_regist:boolean allow_comments_default:boolean allow_pings:boolean allow_pings_default:boolean allow_reg_comments:boolean allow_unreg_comments:boolean archive_path:string archive_tmpl_category:string archive_tmpl_daily:string archive_tmpl_individual:string archive_tmpl_monthly:string archive_tmpl_weekly:string archive_type:string archive_type_preferred:string archive_url:string autodiscover_links:boolean autolink_urls:boolean basename_limit:integer cc_license:string children_modified_on:date convert_paras:string convert_paras_comments:string created_by:string created_on:date custom_dynamic_templates:string days_on_index:index description:text email_new_comments:boolean email_new_pings:boolean entries_on_index:integer file_extension:strgin google_api_key:string internal_autodiscovery:boolean is_dynamic:boolean junk_folder_expiry:integer junk_score_threshold:float language:string manual_approve_commenters:boolean moderate_pings:boolean moderate_unreg_comments:boolean modified_by:date modified_on:date mt_update_key:string name:string old_style_archive_links:boolean ping_blogs:boolean ping_google:boolean ping_others:boolean ping_technorati:boolean ping_weblogs:boolean remote_auth_token:string require_comment_emails:boolean sanitize_spec:string server_offset:float site_path:string site_url:string sort_order_comments:string sort_order_posts:string status_default:string use_comment_confirmation:boolean welcome_msg:text words_in_excerpt:integer]]>
Drupal can manage several different common types of content right out of the box, and that includes blog entries.
To learn more about Drupal’s blogging capabilities, I’ve setup a new website over at Docunext to try out its features:
So far I’ve only scratched the surface, but I’m really looking forward to delving in deeper.
]]>Here are some of my test materials. The first is a PHP script, the other is a custom set of spamassassin scores.
<?php
$sender = "user@example.com";
$subject = "Comment";
$date = date("D, d M Y H:i:s O (T)");
$date_old = date("D, d M Y H:i:s O");
echo <<<EOF
Subject: $subject
Date: $date
From: www-data@example.com (www-data)
hi
hi
hi
hi
hi
EOF;
?>
Spamc settings appropriate for comments:
score MISSING_MID 0.0 score NO_RELAYS 0.0 score NO_RECEIVED 0.0]]>
For more details, see the Docunext wiki page about this:
]]>UPDATE mt_entry SET entry_text = REPLACE(entry_text, "<br/>" ,CHAR(10)),entry_text = REPLACE(entry_text, "<br />" ,CHAR(10)), entry_convert_breaks='markdown' WHERE entry_convert_breaks IS NULL;]]>
rm styles.css rm atom.xml rm sitemap.xml rm mt.js ln -s ../styles.css ./ ln -s ../atom.xml ./ ln -s ../sitemap.xml ./ ln -s ../mt.js ./]]>
While I would prefer learning about new advancements with the version of Movable Type I'm currently using, 4.34, I'm still pleased to interpret these news reports that Movable Type has a solid future.
Honestly, I'm very impressed with the foundation that Six Apart has laid for Movable Type and their future. The industry of publishing has a lot of growing pains to evolve through, but I think MTOS is a saving grace that will help the world transition to a new paradigm.
Big words, big meaning!
]]>But even for non-searches, mt-search.cgi is commonly used. More specifically, when users browse a site by clicking on tags, mt-search.cgi is the script which works the magic.
But why so much CPU utilization? Sure, my MTOS has over 200 blogs and almost ten thousand entries, but MySQL should be able to handle it, right? That's the hard part I'd imagine, so why it mt-search.cgi using so much CPU while MySQL is not? Then I started thinking, maybe its not the search, but the rendering of the template!
Looking at the default search_results.mtml template, I noticed right away that one template is used for raw searches and for tag searches. Not a big deal, that is simple logic. Then I remembered that tag searches display links to all the website tags.
That is something that definitely does not be rendered dynamically every time a search is made. AJAX to the rescue! I created a new index partial for the tag links, and now I'm going to load it dynamically using jQuery. I tested out a page load that didn't include all the tags and it was noticeably faster.
I did this already with the categories widget - and that's not even a dynamic page!
I would like to make the tag urls prettier, probably something like this:
/tagsearch/$blogid/$tag
After that, I'm going to review the SphinxSearch plugin, though I'd really like to use Xapian. I already use it with some Ruby applications I've build and its awesome.
]]>First off, he's got a yurt! I first learned of the word thanks to a random BBS named "The Frog's Yurt" - yes we're talking old school, before I could actually write software, I certainly used and configured a substantial amount of it!
Second of all, he's made some major contributions to Debian, my distribution of choice. In fact, I have a lot of respect, deserved or not, for Debian developers, as a default. I just can't help but think they do what they do for all the right reasons, even if they are ornery, rude, or just plain jerks (yes, he used IMNSHO when commenting about the use of PHP in embedded computers, even though it is used in several incredibly popular embedded platforms: FreeNAS, m0n0wall, and pfSense).
Anyway, over the past couple of years I've taken an interest in Joey's work, and as a start, began using ikiwiki for several wikis. Needless to say, it rocks! If you know me, you'll know that I'm of course using it in a manner much different than most people are, but that's what impresses me about his software - its incredibly flexible and capable.
I also take note of the posts he makes that find their way to Planet Debian. Today I read a post about Branchable - a hosting service for ikiwikis! Hence the creative title of this post: "Joe's Tree". The project is thanks to both Joey Hess and Lars Wirzenius, and I'm pleased to learn that they are down with the Franklin Street Statement, which even endorses the Affero GPL!
]]>I tried two solutions:
After realizing wikilinks didn't render as I'd expected with AlternateSyntaxParser, I decided to go with MarkdownSyntax - besides, all I really want is markdown; no need for Textile and/or friends.
I've previously converted a MediaWiki wiki to an ikiwiki and it was a pain to fix all the wacky MediaWiki-specific formatting.
UPDATE: One of the problems with MarkdownSyntax is that it renders every page! I patched it to use the same method as AlternateSyntaxParser:
if (preg_match('/^#MARKUP\s+([a-zA-Z0-9_]+)/', $text, $matches)) {
$text = substr($text, strlen($matches[0]));
]]>
Here's my setup:
This seems to be a much easier way to manage a complex custom template setup!
]]>Thanks to alternate templates, this task will become much easier in the future!!
I still have to edit CMS/Asset.pm though, and here are some more notes:
990c963,964
< $root_path = $blog->site_path;
---
> #$root_path = $blog->site_path;
> $root_path = $app->static_file_path;
1038c1012
< $asset_file = $q->param('site_path') ? '%r' : '%a';
---
> $asset_file = $q->param('site_path') ? '%s' : '%a';
1043c1017
< $asset_base_url = $app->param('site_path') ? '%r' : '%a';
---
> $asset_base_url = $app->param('site_path') ? '%s' : '%a';
I'm also going to try the same with the thumbnail image paths and URLs.
See Also:
]]>Specifically, I wanted to remove the keywords and excerpts fields from the entry forms. This should be able to be done via the preferences page, but I think because I entered some data into one of those fields into one of my posts, I was stuck with it.
Anyway, I found it easy to setup alternate templates for this purpose. I simple created a new directory, then specified it as the AltTemplatePath configuration parameter in mt-config.cgi.
After that, I created a cms sub-directory, then copied /usr/share/movabletype/tmpl/cms/edit_entry.mtml (that's the path on Debian, other setups may differ) into my alternate templates cms folder, and made the changes I wanted.
Voila! It works. :-)
]]>