initial commit

3 files changed, 145 insertions, 0 deletions

diff --git a/parse_cgit b/parse_cgit
new file mode 100755
index 0000000..34029fe
--- /dev/null
+++ b/parse_cgit
@@ -0,0 +1,59 @@
+#!/bin/sh
+
+#
+# Parse a cgit-style repo index provided on stdin
+#
+
+if [ $# -lt 2 ]; then 
+	printf "Usage: %s <urlbase> <destdir>\n" $0
+	exit 1
+fi
+
+FIN="/dev/stdin"
+URLBASE="$1"
+DEST="$2"
+
+SUBPATH="/plain"
+
+CURL="torify curl -Ls "
+
+PROTO=${URLBASE%%:\/\/*}
+DIRBASE="$PROTO/${URLBASE##[a-z]*:\/\/}"
+echo "proto: $PROTO"
+echo "dirbase: $DIRBASE"
+
+READMES="README README.txt README.md readme readme.txt readme.md"
+
+
+repos=$(${CURL} "$URLBASE" | xml2tsv | grep -Ei "/html/body/div/div/table/tr/td/a[[:blank:]]+title=" \
+		| awk -F'\t' '{print $3,$4}' \
+		| sed -E 's/href=//g;s/ /\|/'\
+		)
+
+
+for r in $repos; do
+	printf "Retrieving repo %s...\n" $repo
+	link=$(echo "$r" | cut -d "|" -f 1 )
+	name=$(echo "$r" | cut -d "|" -f 2 )
+	baselink=$(printf "%s%s" $URLBASE $link)
+	REPODIR="$DEST/$DIRBASE/$link/"
+	mkdir -p $REPODIR
+	for f in $READMES; do
+		printf "    trying file %s..." $baselink/$SUBPATH/$f
+		torify curl -Ls "$baselink/$SUBPATH/$f" > $REPODIR/$f
+		failure=$(xml2tsv < $REPODIR/$f 2>/dev/null | \
+			grep -Eaic "/html/body/div/div/div[[:blank:]]+class=error[[:blank:]]+Not found")
+		echo $failure
+		if [ "$failure" != 0 ]; then 
+			rm -f $REPODIR/$f
+			printf "[FAILED]\n"
+		else
+			printf "[OK]\n"
+		fi
+		#sleep 1
+	done
+done
+
+### the readme file is at REPLINK/plain/README
+
+### if not found, look for "/html/body/div/div/div  class=error     Not found"
diff --git a/parse_stagit b/parse_stagit
new file mode 100755
index 0000000..39feda2
--- /dev/null
+++ b/parse_stagit
@@ -0,0 +1,62 @@
+#!/bin/sh
+
+#
+# Parse stagit-style repos
+#
+
+if [ $# -lt 2 ]; then 
+	printf "Usage: %s <urlbase> <destdir>\n" $0
+	exit 1
+fi
+
+FIN="/dev/stdin"
+URLBASE="$1"
+DEST="$2"
+
+SUBPATH="/file"
+
+CURL="torify curl -Ls "
+
+PROTO=${URLBASE%%:\/\/*}
+DIRBASE="$PROTO/${URLBASE##[a-z]*:\/\/}"
+echo "proto: $PROTO"
+echo "dirbase: $DIRBASE"
+
+READMES="README.html README.txt.html README.md.html readme.html readme.txt.html readme.md.html"
+
+
+repos=$($CURL "${URLBASE}" | xml2tsv | grep "/html/body/div/table/tbody/tr/td/a" \
+		| awk '{print $(NF-1), $NF}' \
+		| sed -E 's/href=//g;s/ /\|/'\
+		)
+
+
+for r in $repos; do
+	name=$(echo "$r" | cut -d "|" -f 2 )
+	link=$(echo "$r" | cut -d "|" -f 1 )
+	link="${link%%log.html}"
+	baselink=$(printf "%s/%s" $URLBASE $link)
+	printf "link: %s\nbaselink: %s\n" $link $baselink 1>&2 
+	REPODIR="$DEST/$DIRBASE/$link/"
+	mkdir -p $REPODIR
+	for f in $READMES; do
+		printf "    trying file %s..." $baselink/$SUBPATH/$f
+		$CURL "$baselink/$SUBPATH/$f" > $REPODIR/$f.tmp
+		failure=$(xml2tsv < $REPODIR/$f.tmp 2>/dev/null | \
+			grep -Eaic "^/html/head/title[[:blank:]]+404 Not Found")
+		echo $failure
+		if [ "$failure" = 1 ]; then 
+			printf "[FAILED]\n"
+		else
+			xml2tsv < $REPODIR/$f.tmp 2>/dev/null | \
+			  grep -Eai "/html/body/div/pre/a[[:blank:]]+href=#.*[[:blank:]]+class=line" | \
+			  cut -f 6- | \
+			  sed -E 's/\\n//g;s/\\t/\t/g;s/\\\\/\\/g' > $REPODIR/$f
+			  printf "[OK]\n"
+		fi
+		rm -f $REPODIR/$f.tmp
+		sleep 1
+	done
+done
+
+
diff --git a/search_repo b/search_repo
new file mode 100755
index 0000000..65dd9c5
--- /dev/null
+++ b/search_repo
@@ -0,0 +1,24 @@
+#!/bin/sh
+
+#
+# Search a set of repos for specific words and return a list of URLS
+# to the matching repos
+#
+
+if [ $# -lt 2 ]; then
+	printf "Usage: %s <dir> <word> [<word>...]\n" $0
+	exit 1
+fi
+
+FOLDER="$1"
+shift
+WORDS="$@"
+
+query=$(echo "$WORDS" | sed -E 's/\ /\|/g')
+
+cd "$FOLDER"
+
+res=$(grep -Eric "$query" | grep -v ":0$" | sort -t ':' -rnk2 | \
+	sed -E 's/([a-z]+)\//\1:\/\//1;s/\/[^\/]*$//' )
+
+echo "$res" | grep -Ei "^[a-z]+://"