[Xitami] New wiki project for Xitami

[R O]

so every email is a unique, static html page?  I connected to the Index listing page which has all the months/years by thread/subject/author/date, is each sort option a unique and individual static webpage?  
 
Meaning the two following pages are in fact unique static webpages?  Or is there some kind of dynamic aliasing going on here that makes it appear to be seperate static web pages?
 
/pipermail/xitami/2007-December/thread.html
&
/pipermail/xitami/2007-December/subject.html 
 
Two things come to mind:
 
1: If the above is true, then the owner of the fileserver can easily write a script that parses the directories based on subject and then on date and time.  This will chronologically order all threads based on subject.  Process each file in order as a text file...grab the author, date, time, subject and content (new content...not older repeated content) and then store that info into a DB with assigned record IDs.  The first article of a grouping based on subject will the master recordID, all subsequent same subjected articles will use the maser recordID to associate with as a common thread.  When a new subject is processed then that means a new recordID must be assigned and repeat the process.  This wouldn't necessarily be the cleanest method but I am certain it would get about 90% of everything.
2: Someone writes a LWRP script to connect to the webserver, grab all the links by "subject" from the index listing (going all the back to May 2001) and then connects to every HTML page strip out the author, subject, date, etc and store it a DB....using some of the same principles and methods from above.
Personally I think #1 would be easier assuming that directory and file structure is static.  Let me know and I will try to cook up a quick perl script that does some of the basic stuff.  All you would need to do is create the tables and db structure based on the notes provided in the script.  I would need to see a directory structure listing and the format of the file each thread is stored in.
 
Russel P. Olinger (310) 466-3399 rolinger1 at hotmail.com> Date: Thu, 7 Feb 2008 19:01:03 +0100> From: mw at 9000.org> To: xitami at lists.xitami.org> Subject: Re: [Xitami] New wiki project for Xitami> > Russel Olinger wrote, On 2008-02-07 18:45:> > Not really certain how to do it because of they are stored. I am looking at > > the old emails now. How are they currently stored? It would seem to me > > they are already in a database b/c they are sortable by > > date/threat/subject/author - to me that implies a database must be needed. > > No, actually not -- these archivers work by making lots of static HTML > files (!). That is true (or at least, was) for the major mailing list > archiver software too, namely Mhonarc. Mailman's internal archiver > (Pipermail) seems to work that way too. I have been thinking of writing > my own parsing scripts for some other mailing lists, it's probably > workable to use in-reply-to and references headers to construct > threading and put it in a database, but I'm lazy... ;-)> > Actually a mail-to-database-converter would be like the holy grail for > me, making it possible to actually construct a functional forum/mailing > list combination, but I have never found any such scripts...> > -- > Magnus Wedberg> http://www.magnuswedberg.com/> mw AT 9000.org> Warning! Objects in viewfinder may be closer than they appear.> > > --> Xitami Users Mailing List -- For Xitami support> To unsubscribe: http://lists.xitami.org/mailman/listinfo/xitami
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.xitami.org/pipermail/xitami/attachments/20080207/0d53bb5f/attachment.htm