A little while back in the CB, someone (forgive me, I have forgotten who), mentioned it would be interesting to see the age distribution of those monks who have reported their ages for the PerlMonks Stats Pages. So I decided to take a stab at generating such stats.

The script currently pulls up the upcoming birthdays page and scrapes the age information out, then outputs the requested file format (HTML or PNG). A future version might pull XML data from the server instead when it is available.

Examples: HTML Output, Summary Graph, Detail Graph.

Update: T'was dada that proposed the age stats in the CB. Thanks for the reminder ;)
Changed PDF to PNG in the comments.

Update: Ack! Looks like ar0n beat me to this almost a whole year ago: hup - home node updater. Thanks for pointing this out ybiC, and don't blame me - blame ar0n! :D

Three Scripts for the Acolytes under the sky,
seven for the Friar-lords in their halls of stone,
nine for Mortal Bishops doomed to die,
one for the Dark Saint on his dark throne
in the Land of Monkwhore where the Black Stats lie.
One Mod to rule them all,
one Mod to find them,
one Mod to bring them all
and in the darkness bind them
in the Land of Monkwhore where the Black Stats lie.

A module that encapsulates behavior and code from jcwren's statswhore.pl and xstatswhore.pl, plus aspects from luke_repwalker.pl, and xluke_repwalker.pl. Also included is the histogram code by larryl.

I started playing with the scripts when I noticed that in the early versions jcwren had experienced problems using my HTML::TableExtract module. The later versions of each script instead use the XML ticker and XML::Twig module, an understandable choice given the circumstances. Nevertheless I started investigating what the original problems were with HTML::TableExtract. As it turns out, these problems have since gone away: perlmonks has a different query interface and results format for writeups via HTML, plus there are later versions of HTML::TableExtract that greatly assist in extracting HTML from the tables in question (as opposed to the traditional role of extracting tables from the HTML in question). These developments negate the need for hoop gymnastics and inheritance hell that jcwren had been forced into.

The XML interface is far quicker, but apparently not all people wish to install the associated XML modules and libraries.

This module incorporates both query methods, XML as well as HTML. The default is XML, but this behavior can be changed via the mode parameter.

If you have XML parsing problems with non UTF-8 encodings in your titles, then simply set the mode to HTML.

Behind a proxy server? Set your http_proxy environment variable to the URL of your proxy server.

See the POD for more information. Thanks to jcwren and larryl for the original code.

Basic usage:

#!/usr/bin/perl

use PerlMonks::StatsWhore;
$sw = PerlMonks::StatsWhore->new(
                                 user     => 'username',
                                 password => 'password',
                                );
print $sw->summary_as_string;
[download]

Check out the code if you're interested in inheritance and polymorphism.

Matt

Updates thus far:

• POD fix per Biker's suggestion
• Added $VERSION info. For gawd's sake we might as well attempt to approach the book keeping abilities of CPAN.
• Added reminder of motivation for XML barfing on non "standard" characters (forget Latin-2) in node titles thanks to another nudge from the ever-testing Biker.

A Perl extension to check PM experience points before a given PerlMonk reaches 'level vroom'.   Handy for monks of the Saint(10)ly persuasion.   Works fine as is, but substantial changes are in the works.

You can peruse all the supporting files initially generated by h2xs here, plus there's a tarball of the whole shebang.

(as this is my first serious-to-completion-of-working-code module writing effort, I'd particularly welcome any and all criticism, complaints, rotton fruit, or even constructive suggestions)

A simple little program that will keep a log of any changes to the title, id, or rep of your nodes, including the addition or deletion of said nodes.

2002-03-31 Edit by the node bunny : Changed login and password.

Please read the warnings about possible dangers. Do not simply execute any code/node you have reason to distrust, but that of course goes for manual downloads as well. This does work well for just downloading too though, if you just want to poke around in a script. :)

Update: The provided patch by Amoe has been applied. My way worked for me on Win2k, cygwin and FreeBSD, so I lazied out on this one. :) Thanks Amoe!

Update 2: Suggestion 2 by belg4mit should be working now: Using the -d option for deparsing instead of executing should now work. Note that in-memory-deparsing uses a new feauture (v5.6.1?) in B::Deparse that won't work with older perls. Use together with -M or -F instead.

pm_username|pm_passwd|email_addy

It's that simple (at least I hope it is).

my example is below... please feel free to add your own.

update: Tue Apr 2 10:37:58 2002 GMT ~ changed off trigger due to Petruchio

update: Fri May 24 11:19:30 2002 GMT ~ " as added, so gots to take it out (and also added a \n remove, jeez, i'm gonna make this obsolete soon ~ and & ~ i hope the goods see pmdev wiki soon)

update: Fri Jan 17 05:31:13 2003 GMT -- more changes, more fixes.

A cool feature about your posts is the fact that you can put HTML comments in the title of your post. Armed with this feature, it is possible to put 'category markers' in your node titles and use a script to find them and make a nice list of your favorite nodes.

The following code will do just that, the demo is on my home node. All you need to do is mark your favorite nodes by hand with something like <!--code--> or <!--obfu-->. Well that's how i did it before i saw jcwren's Bulk Node Title Editor.

The side effect is that you have to link directly to a node's id instead of it's name when referring to it in other posts, chatbox, etc. But that's okay, you use id tags, don't you? Hey Rocko! Ask 'em if they use id tags!

Feel free to send comments and suggestions to crazyinsomniac! :D

UPDATE: - after gleaming good security info from jcwren, i changed this script to accept an interactive password similiar to how mysql asks for passwords: you can specify the password on the command line with the -p option (dangerous for mulit-user boxes), or you can just specify -p by itself and be prompted for the password. I use Term::ReadKey to keep the password 'invisible'.

Implements automatic updating of home node image, after so many hits on the home node. Uses a "bug" script (included) to detect hits, and this piece to automatically uploaded a new image after so many hits.

Please read the section on SECURITY and the rest of the documentation before using. Requires some responsibility on the part of the user.

Quickly check changes in your Perl Monks nodes' reputations. It will list all of your nodes, sorted by reputation in descending order, with any changes since the last time you ran Quickrep.

hup is a home node updater. You can use it on the command line, or from a crontab, to automate updating your homenode. As always, suggestions welcome.

Updates

• 2001-11-14 -- Added scratchpad support, courtesy of thatguy.
• 2002-01-19 -- Miscellaneous fixes. No longer warns when certain fields are empty; prompts for username and password

xml_pimp.pl - You heard of statswhore? You heard (x)luke_repwalker? Well dis da pimp!

Like always, no comments are welcome(now that's what I call sarcasm ;-))

updates:
July 20, 2001 - minor cosmetic changes;
July 21, 2001 - updated some pod (a blah... got by me)
July 27, 2001 - changed sort keys to sort {$a<=>$b> keys(print_HISTORY for -h1)
September 26, 2001 - added &ticker=yes (details: New ticker login for PM XML clients)
Sun Jan 13 21:11:13 2002 GMT - fixed serious bug nobody told me about, which I was unaware of due to my testing environment. I guess the good thing is i'm the only one using this ;)(or I lost the original posted version which worked)

I frequently update my pmchat script and have found that updating my node for it is quite a pain in the *ss. To simplify the experience, I wrote the following script that is called from my Makefile (when my script changes) and automatically updates its node under Code Catecombs.

This script WILL NOT work for you without modification. I hestitated to post it here in its form, but decided that it was probably worth the educational experience.

Note: the script draws from the cookie file that is created by my pmchat script for authentication.

Also, check out Personal Nodelet Restorer.

Comments and improvements always welcome. :)

Fixes:
thanks to crazyinsomniac PodMaster for pointing out my lack of Win32 support in the synopsis

UPDATE: - after gleaming good security info from jcwren, i changed this script to accept an interactive password similiar to how mysql asks for passwords: you can specify the password on the command line with the -p option (dangerous for mulit-user boxes), or you can just specify -p by itself and be prompted for the password. I use Term::ReadKey to keep the password 'invisible'. Because of the new nature, be sure and specify -p last, after -u and -s if used.

UPDATE (Mar 24,2002): - somebody moved 'lastnode' before 'node' so regex has been updated

Main features:

• Colorized output via ANSI escape sequences
• Compatible with both *nix and Win32 (mostly :)
• Displays most recent New Nodes
• Can launch a browser to view a node
• Check for and autoinstall updates
• An "Away" mode for not being in "Other Users"
• Uses the various XML Generators of PerlMonks
• Tracks XP status
• Tracks Reputation of your nodes (and displays WHICH nodes have changed -- and by how much)
• Requires NO configuration, download and run!
• Displays Other Users
• User configurable options (inside Pmchat) for
  • Time-stamping each message (for long-term running)
  • Turn off/on colorization
  • Number of New Nodes listed
  • Which browser to use for node viewing
  • Update on Launch
  • Seconds between polls for new messages (n/a on Windows)
  • Debugging mode. No output is sent to Perlmonks
  • Log file. Keeps a log of all messages. Set "logfile" to the desired filename. To stop, set "logfile" to "0" or "none".

Because pmchat is a text-mode client, it has the following shortcomings

• Doesn't render HTML tags
• Doesn't render special HTML entities (eg: &amp;)

By default, the file is written to perlmonks.html. If that file already exists, a backup will be created at Operlmonks.html before the file is overwritten.

I guess you need to have the WebFetch module installed to run this.

Special thanks to :
- OeufMayo for creating the XML::Parser tutorial that helped me create this.
- mirod for helping me with the XML::Parser problem.

Bugfixes:
- xml_char () was altered to read in entire string as mirod suggested
- the tests to return () were moved to xml_end () from xml_start () to keep from reading too many strings into one field

Suggestions please ...

The code is too large to post it here, so I'll point you to the web page instead.

There are a number of clients that use these modules. Check out my home node for the full list.

Enjoy!

Note: Threaded mode can get dramatically slower as the number of nodes increases, so be prepared to wait when threading more than a few days worth (1000+). Be sure the #! and 3 config variables are correct first.

Also: For some reason xNN is now very very slow checking anything more than 1 day of nodes. Nothing changed in the script which suggests that some change in the XML ticker (or something) is causing it. I'm kind of busy now so the fix will have to wait (unless you figure it out and send me a patch :-)

Code updated: 5.23.2002

Because it's larger than 50k reputer has two nodes on Perlmonks.
This is the source code node in code catacombs.
Please make replies to the reply node.

• Sort nodes by rep, title or date in HTML tables.
  html screenshot
• Three ways to graph number of nodes by reputation.
  html screenshot
• Display changed, deleted and new node info and rep change.
  html screenshot
• Cookies save info used for lwp transfers, and display options.
  html screenshot
• Can write up to 8 data files in temp dir, so give it write permission.
  html screenshot

Also includes proxy support, restricted public access mode,
integration with command line programs and compressed file export.

• Import/export functions disabled unless Compress::Zlib is installed
  
• Requires XML::Simple
  

Update: Fixed possible problem creating data files when first running script.

This is an update of luke_repwalker.pl that uses mirods XML::Twig module. It's faster because it generates few hits to the server. People like tilly should appreciate this... And ultimately it's more reliable, since it doesn't try to parse the tables of article listings.

I posted this as new code so that the old one would still be available. Some people have had trouble installing the various XML modules, or don't wish to.

xstatswhore.pl is an update of statswhore.pl that uses mirods XML::Twig module, and includes larryls histogram code (if you like the histogram part, be sure to thank him. ++Re: statswhore.pl). This version has the advantage of being faster by generating fewer hits to the server, and fixing the problem of having exactly a multiple of 50 nodes. And the histogram part is really cool..

I posted this as new code so that the old one would still be available. Some people have had trouble installing the various XML modules, or don't wish to.

To use just fill in the configuration variables (first!) and call it from your local web server.

functions include:

- emulates perlmonks chatterbox linking standards
- integrated message inbox with cc to self option
- new nodes ticker
- rep change ticker with history
- extract and display $NORM
- textarea input for editing extended messages
- large messages chunked for multiple send
- super search form with perldoc, cpan, google and babelfish
- launchpad for accessing your favorite search engines

Run it and select help for more info.

WARNING: Client authentication makes framechat more secure but exercise caution. Do not allow public access to your installation of framechat!

New in 2.0

• New nodes ticker!
• Rep change ticker with history!
• Client login & quit for better security.
• Create & edit optional external config file.
• Tools control panel to see & set parameters.
• /ignore and /unignore and see who's being ignored.
• /logout and /login implemented.
  • Not in the userlist but still see chat.
  • All links to perlmonks become logged out.
  • History still functions if enabled.
  • Logout disables node title lookup.

  Improved:
  

• Faster thanks to less print statements.
• More robust error handling (esp. xml parsing).
• XP nodelet now shows xp gain and loss messages (Ack!)
• inbox/launchpad/newnode frame can remember state.
• Notes added to update notification.

  Fixed in 2.0:
  

• Bug in link parser that caused italic history after an unclosed i tag.
• Last \$norm should now appear when code updated.
• Code is less redundant and more readable.

  Fixed in 2.01:
  

• A new node with 0 rep showed +0 change (a printf feature).
• New nodes and check update didn't work if proxy in use.
• New nodes loadtime was 1 second off.

  Fixed in 2.02:
  

• Empty document error if history file for present day didn't exist.
• IIS breakage addressed with a new config var: $broken.
• Change to talk frame enables submit with Tab+Enter sequence (thanks grinder).
• <i> tags would italicize cb text, oops.
• Rep related data files now 50% smaller.
• Added config file syntax status feedback and warnings.
• Improved the help and tools sections.
• NEW: Node rep history graph!

  Fixed in 2.03:
  

• control chrs in chat xml not being stripped due to missing /d modifier on tr in sub fixxml.
• client authentication erroneously succeeded if username and password not defined.

  Fixed in 2.04:
  

• Repaired the chat frame and norm parser to conform with changes at perlmonks.
• Added support for the latest shortcut tags: pad, kobe, jargon, perldoc. If you use an external config file these need to be added to the @tags array so they appear in the pulldown menu in the input frame, as well as a few items added to the %launch_urls hash.

To run framechat under mod_perl use OeufMayo's mod_perl patch.

d/l the code, and save the two pieces marked with ##filename=foo.html## as what the name is supposed to be. Be sure cookies and JavaScript are on. Then point your browser at perlmonks.html, and try it out.

With this, this morning I was able to walk through every post for several days worth of PerlMonks in just a few hours. Most of that time was spent writing up responses. With this I can follow the whole site with less energy than it would take to hit a reasonable portion of it through using Newest Nodes.

Update
Clearly my cookie handling in JavaScript sucks. Oh well, I don't like either cookies or JavaScript, so I guess that was my internal bias showing. :-) Now fixed.

Hear monks talk! Listen to your favorite monks battle it out in the chatterbox, as streaming audio! Make your co-workers think you're a basket case!

This script delivers the chatterbox as MP3 encoded streaming audio. It uses the festival/MBROLA text-to-speech (TTS) system to create the audio on the fly.

Since it's based on ZZamboni's Chatterbox Client you will need to download his module as well.

This program gets your personal messages from PerlMonks and displays or mails them to you. It can optionally check them off automatically. Typical use would be in a crontab entry like this:

0 * * * * $HOME/bin/pmpager.pl -m your@address.com
[download]

This program is also part of the PerlMonksChat distribution.

Update: I no longer post the graphs on my home node. The updated code given in the replies to this node is more modern. Thanks to everyone for the kind comments I received when I first wrote this.