plog // perl xml/rss blogger

vim integration for xml-editing

published on 14.11.2005 by xir
a little vim-plugin for better editing xml-files.

i found a little xml plugin. it closes xml-tags automatically and does some other useful stuff on xml editing. actually i wanted a realy xml-editor like mutt's vi integration. there is nothing around. i try to do such a thingy with perls ncurses interface. but this will take a little while and could be the next step of this software.

so for first you find under /plog/plugins/vim/ftplugin/xml.vim a plugin for your personal vim. you need to copy the vim/ directory to user userhome and rename it as .vim. vim is searching for this directory and initialises the plugin automatically. you find help for this stuff on www.vim.org.

plog needs an other name

published on 14.11.2005 by xir
plog needs an other name. there is allready a system under this label

uups, i found a thingy called plog on sf.net. it's a blog-system too using php and mysql as backend. hmmm, i think plog should get an other name. if you have any ideas feel free to tell me. my first proposal was something like plogit but it seems a little boring. how ever, let me know if you find the right name for inappropriate software.

plog creates plain html-pages in the filesystem

published on 14.11.2005 by xir
horrible performance problems ...

plog does not create pages on the fly anymore. it was not possible on the server-engines i wrote it for. daemon.nethack.at needed around 30 seconds to create the pages. the current version creates html-content using the template engine in the memory and writes them directly to the filesystem. the plog engine itself resides under plog/engine/plog.pl now, the xml-content under plog/content/. the system writes the html-files to /plog/ where your webserver should point to.

performance problems

published on 13.11.2005 by xir
because of performance reasons it's not realy useful to create the pages on the fly. so the next big step is a caching system for plain-html files.

i did a cupple of tests with plog on daemon.nethack.at. it's still realy to much for this box. the new idea is to parse the xml-content files and create/write html-files directly in the filesystems. the whole perl engine is then just needed for recreating the pages. it will be called from shell, and rewrite or create all html-files. webvisitors then just call html-files without any logical part.

plog is born // a xml-based perl blogger

published on 13.11.2005 by xir
this is plog version 0.1. it's designed for older server-engines. it run's without any database-backend or webscripting. it just needs perl and a cupple of modules for perl. all contents are written directly in xml/rss files. the files can be used for the weboutput itself and/or as rss-feeds.

needed perl modules

      XML::RSS;             # perl rss-parser
      XML::Parser;          # perl xml-parser
      HTML::Template;       # perl html-templates
      CGI qw(:all);         # use as cgi for weboutput
      POSIX;                # posix setting
      

ok, i need to say, that it is totaly useless to create this software. there are allready tons of blogging systems. most of them use databases to store contents. it's designed especially for older but very lovely servers like daemon.nethack.at and because the author loves to play around ...

since daemon.nethack.at is a shell-server, and most of the people are able to use vi or at least mc for editing contents, i didn't build any interface for creating and editing xml-content. please take care to write well-formed xml. on any parse errors plog will cryout and warn you. contents are not displayed in this case.

all html stuff is created with perls template-engine. in this way we have a strictly segregation of content, logic and layout. you get help for all template-commands using "perldoc HTML::Template" from shell.