Almost there

Now it just needs to figure out that it needs to synthesize directory
listings for dat/pools/files/*/*.1 files.

3 files changed, 14 insertions, 2 deletions

diff --git a/bin/cdxget b/bin/cdxget
index a54612d..46d56c4 100755
--- a/bin/cdxget
+++ b/bin/cdxget
@@ -4,4 +4,4 @@ url='http://web.archive.org/cdx/search/cdx?'
 for arg in "$@"; do
 	url+="$s${arg%%=*}=$(printf '%s' "${arg#*=}"|urlencode)&"
 done
-curl -s "$url"
+curl -sL "$url"
diff --git a/bin/fmt-metadata b/bin/fmt-metadata
index 7867d63..c92419b 100755
--- a/bin/fmt-metadata
+++ b/bin/fmt-metadata
@@ -19,5 +19,8 @@ $stdin.each_line do |line|
 		datetime = Time.parse("#{datetime} +00:00").utc.strftime('%Y-%m-%d %H:%M')
 	end
 
-	puts ("%-22s  %s  %3s%s" % [ name, datetime, size_numb, size_unit ])
+	# discard the size, I guess.  The number of digits precision was
+	# inconsistent over the years.
+	#puts ("%-22s  %s  %3s%s" % [ name, datetime, size_numb, size_unit ])
+	puts ("%-22s  %s" % [ name, datetime ])
 end
diff --git a/bin/poolify b/bin/poolify
index af8bf40..e33821b 100755
--- a/bin/poolify
+++ b/bin/poolify
@@ -13,6 +13,7 @@ main() {
 	set -euE -o pipefail
 	shopt -s nullglob
 
+	echo '# Pass 1'
 	while read -r snap name date time size; do
 		dirpart="${name%/*}"
 		filepart="${name##*/}"
@@ -22,6 +23,7 @@ main() {
 		ln -sr "$filedir/$filepart" "$snapdir"
 	done < "$1"
 
+	echo '# Pass 2'
 	while read -r time url; do
 		name="${url##*/Public/}"
 		dirpart="${name%/*}"
@@ -63,6 +65,13 @@ main() {
 			i+=1
 		done
 	done < "$2"
+
+	echo '# Pass 3'
+	while read -r missing; do
+		if [[ -f "${missing/.OLD}/${missing##*_}" ]]; then
+			ln -sr "${missing/.OLD}/${missing##*_}" "$missing"
+		fi
+	done < <(find dat/pools/files/*-PROGRAMS_CVTUTF.OLD_* -type d -empty)
 }
 
 main "$@"