Changes [Sep 25, 2007]
CategoriesWubWiki
Low hanging fruit
Wild ideas
Not-so-wild ideas
Download wiki datab...
Revision history
More Changes...
Changes [Sep 25, 2007]
CategoriesMini.net is going to be decommissioned in the near future, so these will need to be identified and changed.
Also, many images used in wiki pages still live on mini.net - so a new home will have to be found for them
jdc I ran this script on a local copy of the wiki database to get a list of mini.net references:
package require Mk4tcl
mk::file open db /tmp/wiki/data/wikit.tkd
mk::view layout db.pages
foreach pn [mk::select db.pages -glob page "*mini.net*"] {
set nm [mk::get db.pages!$pn name]
set pg [mk::get db.pages!$pn page]
foreach i [split $pg] {
if { [string match "*mini.net*" $i] } {
if { [info exists ml($i)] } {
lappend ml($i) $pn
} else {
set ml($i) [list $pn]
}
if { [info exists pl($pn)] } {
lappend pl($pn) $i
} else {
set pl($pn) [list $i]
}
}
}
}
mk::file close db
set f [open mini_net_img_page.dat w]
foreach i [lsort -dictionary [array names ml]] {
puts $f "$i : $ml($i)"
}
close $f
set f [open mini_net_page_img.dat w]
foreach p [lsort -dictionary [array names pl]] {
set first 1
foreach i $pl($p) {
if { $first } {
puts $f "$p:"
set first 0
}
puts $f " $i"
}
}
close $f
exit
This script generates 2 files. mini_net_img_page.dat contains the mini.net links with a list of the pages in which they were found. mini_net_page_img.dat contains the pages with a list of mini.net references.
jdc I downloaded all png/jpg/gif files from mini.net I could find in wikit.tkd (local copy). They are now available from http://tclerswiki.googlepages.com. To access an image, add the image name this url: http://tclerswiki.googlepages.com. mini.net had different subdirectories where images were stored. This is not possible on http://tclerswiki.googlepages.com.
The following script will:
package require Mk4tcl
mk::file open db /tmp/wiki/data/wikit.tkd
while { 1 } {
set cnt 0
foreach pn [mk::select db.pages -glob page "*mini.net*"] {
set pg [mk::get db.pages!$pn page]
set rpg {}
foreach i [split $pg "\n"] {
if { [regexp {\[http://mini.net/(files|sdarchive|pub)/([^.\]]+\.(png|jpg|gif))\]} $i m f1 fnm] } {
puts "$pn: $i : $fnm"
incr cnt
regsub {\[http://mini.net/(files|sdarchive|pub)/([^.\]]+\.(png|jpg|gif))\]} $i "\[http://tclerswiki.googlepages.com/$fnm\]" i
}
if { [regexp {http://mini.net/tcl/([0123456789]+)} $i m pnr] } {
puts "$pn: $i : $pnr"
incr cnt
regsub {http://mini.net/tcl/[0123456789]+} $i "http://wiki.tcl.tk/$pnr" i
}
if { [regexp {http://mini.net/tcl/rss.xml} $i m] } {
puts "$pn: $i"
incr cnt
regsub {http://mini.net/tcl/rss.xml} $i "http://wiki.tcl.tk/rss.xml" i
}
lappend rpg $i
}
set pg [mk::set db.pages!$pn page [join $rpg "\n"]]
}
puts $cnt
if { $cnt == 0 } {
break
}
}
mk::file commit db
mk::file close db
exit
jdc After running these scripts, searching for mini.net references results in the list found here. Can any of those be done automatically?
mjanssen mentioned the following possible convertions in the Tcler's chat:
regsub -all {^http://mini.net/tcl/(.*)\*$} http://mini.net/tcl/see* {http://wiki.tcl.tk/_search?S=\1*}
regsub -all {http://mini.net/tcl/(\S+)\*} {dsakjhd adkh http://mini.net/tcl/see* daskhdasjhd} {http://wiki.tcl.tk/_search?S=\1*}
http://mini.net/sdarchive/ no need to change AFAIKS
http://mini.net/tcl/nstcl -> http://wiki.tcl.tk/nstcl
Changing an url to contain _search will break WubWiki / wikit compatibility because _search is not recognised by wikit. Any opinions?
jdc One possible problem with googlepages is the unknow download limit. When hitting this limit, images will no longer be available for some time.
