Dr Fragen in the operating room


  • Snailjacked

    Well, I’ve found out that someone is hijacking my posts with no attribution and no permission. I’m not even sure I understand the logic as there isn’t even any advertising there — yet.
    If you’re seeing anything on the domain computersoftware.snailjack.com, it’s a rip off from somewhere else. Do me a favor and don’t click on any links, don’t support there advertising and just don’t go on their site.
    Other than restricting my RSS feed I’m not sure what I can do to stop it.
    Don’t support anything from the domain SNAILJACK.COM!

  • VoodooPad Pro Blogging

    Please note that VoodooPad Pro has been renamed to VoodooPad.
    It’s time to package up all my VoodooPad blogging scripts and assorted web export plugins so that I can

    1. Remember what the heck I was thinking.
    2. Remember what all these pieces were supposed to accomplish as a whole.
    3. Provide some sort of reference to others interested in this stuff.

    Concept: To use VoodooPad as a CMS for a web site, allowing for static sites and inclusion of blog-style concatenated page. The blog-style page was originally intended as a News page of a static business site. SEO optimization necessitating the addition of a page with re-newing content. Google likes it more.
    Disclaimer: I like Markdown and built these scripts around the fact that the VP document is essentially always formatted in plain text. Images need to be added using links.
    I will include a sample package that contains a VP document, a web export plugin and several script plugins. The site the sample is based upon is at.
    VoodooPad is a requirement as the meta data and triggers are necessary in the creation of the files. My workflow is as follows.

    1. Create new page/post.
    2. Run script plugin Blog > Meta Markup - HomepageMeta on new page.
    3. Run script plugin Blog > Make Homepage and Feed
    4. Run web export.
    5. FTP files up to site. I use Transmit. It’s also a great app.

    The script will move all the created files into a folder structure so that the file structure on the computer will match the web site structure.
    VoodooPad document
    Item meta data is contained in the HomepageMeta page of the document. This is where most of the personalized info on the site is contained.
    In the VP document, tags are used to create a pseudo-folder structure. This means that a page may only have one tag. Some pages are not exported but are used as reference, ie. links in the sidebar or the HomepageMeta page. Pages that are to be rendered at the root level of the site are tagged main.
    Triggers can be used to include certain information into specific pages or info on every page. I use this for things like Google Analytics which needs to be on every page or the analytics confirmation that is only on the index.html page.
    The Web Export Plugin
    Some customization of these files will be necessary for your specific site.

    Images are kept in the web export plugin. Files like robots.txt that need to be at the root level are kept in a root folder inside the web export plugin.

    • preflight.sh – creates the actual folder hierarchy, the folders are created in the preflight.sh script. The preflight.sh script also copies files like images to the export folder.

    • postflight.sh – can be used to rename pages that may be PHP based from .html to .php

    • deleteVPExtra.sh – deletes the extra pages that VoodooPad creates during a web export. No changes to this file are needed.

    • fixRelativePaths.sh – fixes the relative URLs that occur inside of nested pages. No changes to this file are needed.

    • parseMetaXml.rb – Parses the meta.xml file that VoodooPad creates and uses that infomation to move files into their correct folder structure on the disk and to create a valid sitemap.xml file for use with Google Sitemaps. It also creates and parses the vpDocMeta.xml file for variables.

    • Info.plist – This file was edited to call these scripts during the web export process.

    • vpDocMeta.xml – This file resides inside the root folder of the Web Export Plugin. It is created by the parseMetaXml.rb script. It contains the following data points.

    • root – default value is main. This is the name of the tag of rendered pages that belong at the root level of the folder structure.
    • baseUrl – This is the base URL for the site for creation of the sitemap.xml file.
    • isHomeIndex – default value is false. This is only used if the the page name in VP for the index page of the site is home. This is true if the main page is a blog-style homepage.
    • noIndex – default value is private. This is the name of a directory whose files you don’t want included in sitemap.xml.

    Script Plugins
    These are placed in the ~/Library/Application Support/VoodooPad/Script PlugIns/ folder. No changes should be required in these scripts. They will be viewable from the Plugin > Blog menu in VoodooPad.

    • meta_markup_HomepageMeta.lua – Takes the data from the page HomepageMeta in the VP doc and adds specific item meta data to the frontmost page.
    • meta_markup_update.lua – Propagates changes made to the HomepageMeta page to all the item meta data of all pages that require them. This is only just the pages used for the blog-style page.
    • blog_home_atom.lua – Creates the blog-style page now called home in the VP document and also creates a valid atom.xml feed in the output directory.

    Comments and questions are welcome. If some enterprising person wants to wrap this all up in the new plugin architecture that would be cool. I’m happy to help if I can.

  • WordPress and the iPhone

    So I’m trying out the latest WordPress for iPhone app and it’s pretty cool. What I’ve noticed though, because I use the [Markdown][md] plugin is that the preview shows the marked up content and not the rendered content. It would be great if somehow this could be fixed.
    [md]: http://daringfireball.net/projects/markdown/

  • WordPress SVN

    OK, since I’ve switched over to WordPress I have to admit it’s been pretty painless. I decided to go with the [SVN update process][1] since I know it’s much easier. But the monkey wrench in the process is having to switch to the new checkout for every stable release.
    So instead of
    $ cd ./blog
    $ svn up
    I have to do the following.
    $ cd ./blog
    $ svn sw http://svn.automattic.com/wordpress/tags/2.3.2/
    $ svn up
    Why isn’t there a symlink for the latest stable release? Then I could set up..
    $ svn sw http://svn.automattic.com/wordpress/tags/current
    and be done with it.
    [1]: http://codex.wordpress.org/Installing/Updating_WordPress_with_Subversion