Clean up the cleaning up

3 files changed, 30 insertions, 26 deletions

diff --git a/SuckItFos b/SuckItFos
index b59f6b0..11adaf9 100755
--- a/SuckItFos
+++ b/SuckItFos
@@ -10,7 +10,7 @@ filter="
  -name TWiki -prune -o \
 "
 
-pushd /opt/merged
+pushd /opt/mergedWork
 
 find /opt/Foswiki/data  ${filter} \
 -name "*.txt" -type f,l -printf "%P\n" | while read line
@@ -26,7 +26,6 @@ do
     # TODO - try curl, to see what is actually downloaded, and maybe not download unchanged pages.  curl to .HTM
     # Doesn't help with redownloads, coz natch a dynamic site isn't cached.  But I can at least comment out the curl command during testing to save time.
     curl --silent --no-progress-meter ${URL}/${base}/${file}?cover=print -o Foswiki/${base}/${file}.HTM
-
     cp Foswiki/${base}/${file}.HTM Foswiki/${base}/${file}.HTM_ORIGINAL
     csplit -ks Foswiki/${base}/${file}.HTM '%<div id="patternMainContents">%' '/<div class="foswikiAttachments foswikiFormStep" style="overflow:auto">/'
     if [ -f xx00 ]; then
@@ -58,18 +57,21 @@ do
     sed -i -E Foswiki/${base}/${file}.md \
         -e 's/\$/\$dlr\$/g' \
         -e 's/\{#.*\}//g' \
-        -e 's/\{\.pattern.*\}//g' \
-        -e 's/\{\.pattern.*//g' \
+        -e '/^:::/d' \
         -e '/^<!-- -->/d' \
-        -e '/^:::/d'
+#       -e 's/\{\.pattern.*\}//g' \
+#       -e 's/\{\.pattern.*//g' \
+
     echo -e "****\n[Original page](${URL}/${base}/${file}) where maybe you can edit it." >> Foswiki/${base}/${file}.md
 
 #    pandoc -t html -f commonmark_x --self-contained            Foswiki/${base}/${file}.md > Foswiki/${base}/${file}.htm
 #    cmark-gfm -t html -e footnotes -e table -e strikethrough   Foswiki/${base}/${file}.md > Foswiki/${base}/${file}.body
 #    ln -frs Foswiki/${base}/${file}.body combined/${base}/${file}.body
     ln -frs Foswiki/${base}/${file}.md combined/${base}/${file}.md
-done
 
-notYetAnotherWiki.lua
+    if [ -f xx01 ]; then
+        rm xx01
+    fi
+done
 
 popd
diff --git a/SuckItPm b/SuckItPm
index 156ee9f..a63eb08 100755
--- a/SuckItPm
+++ b/SuckItPm
@@ -10,20 +10,20 @@ filter="
  -not -name ".pageindex" -a \
 "
 
-pushd /opt/merged
+pushd /opt/mergedWork
 
 find /opt/pmwiki/wiki.d  ${filter} \
 -name "*.*" -type f,l -printf "%P\n" | while read line
 do
     base=`echo "${line}" | cut -d '.' -f 1`
     file=`echo "${line}" | cut -d '.' -f 2`
+#    page="?n=${line}"
     mkdir -p PmWiki/$base
     mkdir -p combined/$base
     echo "Converting ${URL}/?n=${base}.${file}?action=print -> PmWiki/${base}/${file}.md"
 #    pandoc -f html -t markdown --self-contained ${URL}/?n=${base}.${file} >PmWiki/${base}/${file}.md
     # TODO - try curl, to see what is actually downloaded, and maybe not download unchanged pages.  curl to .HTM
     # Doesn't help with redownloads, coz natch a dynamic site isn't cached.  But I can at least comment out the curl command during testing to save time.
-#    curl --no-progress-meter ${URL}/?n=${base}.${file} -o PmWiki/${base}/${file}.HTM
 #    curl --no-progress-meter ${URL}/?n=${base}.${file}?action=markdown -o PmWiki/${base}/${file}.MD
     curl --no-progress-meter ${URL}/?n=${base}.${file}?action=print -o PmWiki/${base}/${file}.HTM
     cp PmWiki/${base}/${file}.HTM PmWiki/${base}/${file}.HTM_ORIGINAL
@@ -44,7 +44,6 @@ do
         -e "s/class='vspace'//g" \
         -e "s/class='wikilink'//g" \
         -e "s/style='.*;'//g"
-#       -e "s/class='.*'//g" \
 #       -e "s/style='background-color: #.*;'//g" \
 #       -e "s/style='font-size: .*;'//g"
 
@@ -57,23 +56,26 @@ do
         -e 's/\$/\$dlr\$/g' \
         -e 's/\{#.*\}//g' \
         -e '/^:::/d' \
-        -e '/\[Site$/d' \
-        -e '/^Page last modified on /d' \
-        -e '/^\[\]/d' \
-        -e "s/\`<a id='trailstart'>\`\{=html\}\`<\/a>\`\{=html\}//g" \
-        -e "s/^\`<img /<img /g" \
-        -e "s/^\`\`\`\{=html\}//g" \
-        -e "s/^\`\`\`//g" \
-        -e "s/\`\{=html\}//g"
+#       -e '/\[Site$/d' \
+#       -e '/^Page last modified on /d' \
+#       -e '/^\[\]/d' \
+#       -e "s/\`<a id='trailstart'>\`\{=html\}\`<\/a>\`\{=html\}//g" \
+#       -e "s/^\`<img /<img /g" \
+#       -e "s/^\`\`\`\{=html\}//g" \
+#       -e "s/^\`\`\`//g" \
+#       -e "s/\`\{=html\}//g"
 
-    echo -e "****\n[Original page](${URL}/${base}/${file}) where maybe you can edit it." >> PmWiki/${base}/${file}.md
+    # Don't need this, the parts we are grabbing already include that link at the bottom.
+#    echo -e "****\n[Original page](${URL}/${base}/${page}) where maybe you can edit it." >> PmWiki/${base}/${file}.md
 
 #    pandoc -t html -f commonmark_x --self-contained            PmWiki/${base}/${file}.md > PmWiki/${base}/${file}.htm
 #    cmark-gfm -t html -e footnotes -e table -e strikethrough   PmWiki/${base}/${file}.md > PmWiki/${base}/${file}.body
 #    ln -frs PmWiki/${base}/${file}.body combined/${base}/${file}.body
     ln -frs PmWiki/${base}/${file}.md combined/${base}/${file}.md
-done
 
-notYetAnotherWiki.lua
+    if [ -f xx01 ]; then
+        rm xx01
+    fi
+done
 
 popd
diff --git a/TODO.md b/TODO.md
index 0a748d2..4157083 100644
--- a/TODO.md
+++ b/TODO.md
@@ -3,21 +3,21 @@
 ## Do these
 
 Bugs -
-- PmWiki in it's current config needs that ?n=foo.bar nonsense for the Original page link.
 - https://nyaw.wiki.devuan.org/Foswiki/Main/JensKorte/WebPreferences.HTML &nbsp;        "Main" trail not getting whichPage()
-- https://nyaw.wiki.devuan.org/PmWiki/Site/Site.HTML &nbsp;                             missing everything after the "?action=attr." bit.
-- https://nyaw.wiki.devuan.org/PmWiki/Site/EditForm.HTML
-- https://nyaw.wiki.devuan.org/PmWiki/Profiles/Debdog.HTML &nbsp;                       pandoc can't handle the background table cell colours in the "Background colours" table, which is kinda the point of it.
 - https://nyaw.wiki.devuan.org/Foswiki/Sandbox/WebHome.HTML &nbsp;                      has that twisty thing which looks not simple to remove.
 - https://nyaw.wiki.devuan.org/Foswiki/Sandbox/WebChanges.HTML
 - https://nyaw.wiki.devuan.org/Foswiki/Sandbox/WebLeftBarExample.HTML
 - https://nyaw.wiki.devuan.org/Foswiki/Sandbox/WebPreferences.HTML &nbsp;               has that twisty thing which looks not simple to remove.
 - https://nyaw.wiki.devuan.org/Foswiki/Sandbox/WebSearch.HTML &nbsp;                    has that twisty thing which looks not simple to remove.
 - https://nyaw.wiki.devuan.org/Foswiki/Main/AdminGroup.HTML
-- https://nyaw.wiki.devuan.org/Foswiki/Main/DevuanCluster.HTML &nbsp;                   A very lengthy and complex document, I'll likely miss something, but chip away at the obvious.
 - https://nyaw.wiki.devuan.org/Foswiki/Main/JensKorte/WebLeftBar.HTML
 - https://nyaw.wiki.devuan.org/Foswiki/Main/WikiGroups.HTML &nbsp;                      has that twisty thing which looks not simple to remove.
+- https://nyaw.wiki.devuan.org/PmWiki/Site/Site.HTML &nbsp;                             missing everything after the "?action=attr." bit.
+- https://nyaw.wiki.devuan.org/PmWiki/Site/EditForm.HTML
+- https://nyaw.wiki.devuan.org/users/dunno/DevuanCluster.HTML &nbsp;                    A very lengthy and complex document, I'll likely miss something, but chip away at the obvious.
+- https://nyaw.wiki.devuan.org/users/Debdog.HTML &nbsp;                                 pandoc can't handle the background table cell colours in the "Background colours" table, which is kinda the point of it.
 - {.underline} is the result of <strong>foo</strong> getting lost in translation.
+- PmWiki in it's current config needs that ?n=foo.bar nonsense for the Original page link.  Which I'm currently neatly sidestepping, the scraped page has a similar thing.
 
 Check the timestamps on the files, only update if source is newer than destination.  Meh, it's already 600 times faster than the pandoc version.
 - One quirk to watch for is if a URL path changes, the docs that have that URL need to be redone.