On Wed, 29 May 2002, Phil Marlowe wrote:

> I'd like a quick and easy method of deriving the CHK key for a document, 
> other than reading the nodes' response to a successful insert.
> 
> My reasons are convoluted, but they have to do with an attempt to come up 
> with a utility to insert large sites.  Right now fcpputsite is very 
> perfectionist.  It  will only write a mapfile if every single page has 
> inserted successfully *in this run*.

Actually, if you grab the latest fcpputsite from CVS, the --record-file
option records the filename, CHK, modification time and size of every
file.  If you use a previous --record-file on a retry, it marks every
file that made it as 'Done'.  If you keep a copy of your full-site log
around, you can insert just the map/DBR/index.htmls if you want.
(This may not work on windows, since I don't know if the win32 stat()
returns the same fields as posix stat().  If not, send patches please!)

Large sites lose currently.  Over-large maps end up failing, when in
reality lots of the data can be direct CHK links.  My current putsite
project is to fix this, somehow.  Possibly by doing URL rewriting on 
inserted HTML pages (non-HTML content rewritten as direct CHK links and
not included in the mapfile)

Right now I'm just using scripts to generate the HTML, but that's
somewhat of a loss.

--Dan


_______________________________________________
devl mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to