doom9 archiving

[hypercube]

is it possible to publish .CHM archive of the forum ?

[Doom9]

we're talking several hundred MBs of SQL dump already and in HTML form it would probably get even bigger...

[hypercube]

so what ?
:D
I'm not scared by this, your forum is based on vBulletin,
it could be possible to build a converter to CHM format,
or a more compact format with a specific viewer, decompresing
HTML on the fly and showing in IE5 (in Visual Basic for example)

I never see vBulletin database, but all post should be in only
one table... with an odbc driver it could be easy to read it.

I don't have enought free time to do this but I can
start to work on it.

[hypercube]

I took a look on vBulletin. This is very simple:
you have to go in Admin pannel in "Import & Maintenance"
then select Back-up database
then save only "post" table.
it should be better to use PhpMyAdmin to zip the sql backup
of table "post"

"post" table is very simple to parse:

_postid_
_threadid_
_username_
_userid_
_title_
_dateline_
_attachmentid_
_pagetext_
_allowsmilie_
_showsignature_
_ipaddress_
_iconid_
_visible_
_edituserid_
_editdate_

with postid and threadid it is possible to partially reconstruct
all threads.

(user names are lost but this is not usefull)

[hypercube]

sorry, usernames are not lost, there are in "post" table
'thread' and 'forum' tables are also usefull to reconstruct
posts...

[Kaizen]

Hmm. I am also going to look into this.

It would be a hoorendous (sp) task to perform.

[hypercube]

It would be a hoorendous (sp) task to perform.
well, not so difficult... I'm working in EDM (Electronic Document Management) so I do this type of work every day... :D
For example, these days I worked on 20gig of jpg documents.
jpg => multipage PDFs => Oracle Database
This is very funny to do... :rolleyes:

(Most part of my work is to manage millions of documents in
Oracle Database via jdbc/odbc in VC++/Java/Visual Basic ;) )

[Kaizen]

Even so, we are unlikely to give you parts of our database.

It will be discussed and you will find out in due course.

[hypercube]

I can ty to build a win32 tool to convert sql dump of vbulletin in
HTML (CHM is another pb) with only 10 or 20 posts provided by
the team...
do you have some coders in the team ? :D

[hypercube]

I downloaded HTML Help Workshop, chm compiler is very simple to use.
index is build automatically. I have only to build the hierarchy
in toc.hhc ASCII file. This can be done with a few lines of code.

do you have the possiblity to launch a win32 app on the vbulletin base
with an odbc driver ?
if yes, I can start to code this app and send it to you. :D
if not, I can start to code an app without odbc acces, so working on
SQL dump (big disk space needed :( )

another solution is to build a bot to scan forum with HTTP requests
(may be the most simple idea, but the slowest)

in my opinion, archiving doom9 is very important for the community,
and can be benefict for the forum usability.

- using archives on CDR => forum bandwith remain free.
- no more post from admin to say "search before post !"
- more post from admin to say "download archive before post !" :D :D

[Kaizen]

I appreciate your enthusiasm and agree with you that its a good idea.

However, if it is to be done, it will be done 'in-house'.

[hypercube]

:-)

okay, good.

an idea for you:

you can produce HTML files with javascript witch
decompress content on the fly. Then put theses HTML
files in a single CHM and the final result could
be the smaller possible.