(meta,csg) Added a simple meta.awk file and updated csg script

2 files changed, 55 insertions, 234 deletions

diff --git a/csg b/csg
index de8b6bd..7c144d2 100755
--- a/csg
+++ b/csg
@@ -1,4 +1,4 @@
-#!/usr/bin/env sh
+#/usr/bin/env sh
 
 ### Preprocessing 1
 ### Help message
@@ -24,15 +24,10 @@ Arguments:
                        generated webpages.
 
 Variables (you can predefine):
-  CSG_TIMESTAMP_FILE    File path for git-based timestamps (when git exists)
-                        Default: "./timestamps.txt"
-                        Disable: "none"
   CSG_TMP_DIR           Directory to put temporary files
                         Default: "/tmp"
   CSG_AWK               awk executable and flags
                         Default: searches for gawk
-  CSG_MD_TO_HTML        Program to convert Markdown to HTML and it'\''s flags
-                        Default: searches for lowdown
 
 Dependencies (required):
   bash, printf, find, cpio
@@ -65,36 +60,9 @@ URL="$4"
 [ -z "$URL" ] && echo "[Error] <base_url> cannot be an empty string!" \
 	&& csg_help 3
 
-LIB_SCRIPT="$5"
-[ -n "$LIB_SCRIPT" -a ! -d "$LIB_SCRIPT" ] && echo "[Error] $LIB_SCRIPT (<scripts_dir>) isn't a directory or doesn't exist!" \
-	&& csg_help 1
-
-PREV_ALL="$6"
-[ -n "$PREV_ALL" -a ! -d "$PREV_ALL" ] && echo "[Error] $PREV_ALL (<scripts_dir>) isn't a directory or doesn't exist!" \
-	&& csg_help 1
-
-PREV_MD="$7"
-[ -n "$PREV_MD" -a ! -d "$PREV_MD" ] && echo "[Error] $PREV_MD (<scripts_dir>) isn't a directory or doesn't exist!" \
-	&& csg_help 1
-
-ON_MD="$8"
-[ -n "$ON_MD" -a ! -d "$ON_MD" ] && echo "[Error] $ON_MD (<scripts_dir>) isn't a directory or doesn't exist!" \
-	&& csg_help 1
-
-ON_HTML="$9"
-[ -n "$ON_HTML" -a ! -d "$ON_HTML" ] && echo "[Error] $ON_HTML (<scripts_dir>) isn't a directory or doesn't exist!" \
-	&& csg_help 1
-
-ON_OTHER="${10}"
-[ -n "$ON_OTHER" -a ! -d "$ON_OTHER" ] && echo "[Error] $ON_OTHER (<scripts_dir>) isn't a directory or doesn't exist!" \
-	&& csg_help 1
-
 ### Preprocessing 3
 ### "Global" variables
 
-: ${CSG_TIMESTAMP_FILE:="$(realpath ./timestamps.txt)"}
-[ "$CSG_TIMESTAMP_FILE" = 'none' ] && CSG_TIMESTAMP_FILE=""
-
 : ${CSG_TMP_DIR:='/tmp'}
 [ ! -d "$CSG_TMP_DIR" ] && echo "[Error] $CSG_TMP_DIR (\$CSG_TMP_DIR) doesn't exist!" \
 	&& csg_help 1
@@ -109,74 +77,14 @@ if [ -n "$CSG_AWK" ]
 then :
 elif isCommand 'gawk'
 then
-	CSG_AWK='gawk
-		--posix'
+	CSG_AWK='gawk --posix'
 else
 	echo '[Error] Couldn'\''t find an awk implementation!'
 	csg_help 4
 fi
 
-if [ -n "$CSG_MD_TO_HTML" ]
-then :
-elif isCommand 'lowdown'
-then
-	CSG_MD_TO_HTML="lowdown
-		--html-no-escapehtml
-		--html-no-skiphtml
-		--parse-no-metadata
-		--parse-no-autolink"
-else
-	echo '[Error] Couldn'\''t find a Markdown to HTML convertion program!'
-	csg_help 4
-fi
-
-### Preprocessing 4
-### Timer/status function definitions
-
-overwriteLine() {
-	printf "\033[1A\033[K$1\n"
-}
-
-_dur=0
-startTimer() {
-	_dur="$(date +%s%N)"
-	printf "$1\n"
-}
-endTimer() {
-	_dur="$(echo "scale=4 ; ($(date +%s%N) - $_dur) / 1000000000" | bc -l)"
-	printf "\033[1A\033[K$1    %.3fs\n" $_dur
-}
-
-### Stage 0
-### Create timestamps.txt file, if necessary
-
-if [ -n "$CSG_TIMESTAMP_FILE" ] && isCommand 'git' && [ -d '.git' ]
-then
-	timestamps_writeDates() {
-		while read -r file
-		do
-			git log --follow --pretty=format:%ad --date="format:%d %B %Y" -- "$file" | $CSG_AWK -v dest="$DESTINATION" -v file="$file" '
-				BEGIN { file = dest substr(file, 2, length(file) - 3) "html" }
-				NR == 1 { later = $0 }
-				END { print file "\t" $0 "\t" later }' >> "$CSG_TIMESTAMP_FILE"
-		done << EOF
-$(find . -type f -name '*.md')
-EOF
-	}
-
-	headc="$(git rev-parse HEAD)"
-	if [ ! -f "$CSG_TIMESTAMP_FILE" ] || [ "$headc" != "$(head -1 "$CSG_TIMESTAMP_FILE")" ]
-	then
-	startTimer "[Stage 0] Creating timestamps file"
-		echo "$headc" > "$CSG_TIMESTAMP_FILE"
-		(cd "$SOURCE" && timestamps_writeDates)
-	endTimer "[Stage 0] Done"
-	fi
-fi
-
 ### Stage 1
 ### Copy everything from SOURCE to DESTINATION
-startTimer "[Stage 1] Moving files to destination"
 
 (cd "$SOURCE" && \
 	find . -type f -not -name "\.*" -not -name '_header.html' -not -name '_footer.html' | \
@@ -184,164 +92,41 @@ startTimer "[Stage 1] Moving files to destination"
 
 while read -r file
 do
-	mv "$file" "${file%md}html"
+	[ -n "$file" ] && mv "$file" "${file%md}html"
 done << EOF
 $(find "$DESTINATION" -type f -name "*index.md" -links +1)
 EOF
 
-overwriteLine "[Stage 1] Combining awk scripts"
-
-combine_scripts() {
-	rm "$CSG_TMP_DIR/$2" >/dev/null 2>&1
-
-	[ -z "$1" ] && echo "" && return
-
-	scripts=$(find "$1" -type f -name "*.awk" | sort)
-	[ -z "$scripts" ] && echo "" && return
-
-	cat $scripts > "$CSG_TMP_DIR/$2"
-	echo "-f	$CSG_TMP_DIR/$2"
-}
-
-lib_script="$(combine_scripts "$LIB_SCRIPT" "csg-lib_script.awk")"
-on_md="$(combine_scripts "$ON_MD" "csg-on_md.awk")"
-on_html="$(combine_scripts "$ON_HTML" "csg-on_html.awk")"
-on_other="$(combine_scripts "$ON_OTHER" "csg-on_other.awk")"
-prev_all="$(combine_scripts "$PREV_ALL" "csg-prev_all.awk")"
-prev_md="$(combine_scripts "$PREV_MD" "csg-prev_md.awk")"
-
-echo
-endTimer "[Stage 1] Done"
-
 ### Stage 2
-### Transforming Markdown
-startTimer "[Stage 2] Transforming Markdown to HTML"
-
-# prev_all
-cat << EOF > "$CSG_TMP_DIR/csg-export.awk"
-function export(name, value) {
-	if (name == "" || value == "") {
-		return
-	}
-	# if is string
-	if (value + 0 != value) {
-		gsub(/"/, "\\\\\\"", value)
-	}
+### Meta
 
-	printf "	-v	" name "=" ((value + 0 == value) ? value : ("\\"" value "\\"")) "	"
-}
+echo 'BEGIN {' > meta.begin.awk
+while read -r file
+do
+	$CSG_AWK -f meta.awk "$file" >> meta.begin.awk &
+done << EOF
+$(find "$DESTINATION" -type f -name "*.md")
 EOF
-exports=$($CSG_AWK \
-          	-f "$CSG_TMP_DIR/csg-export.awk" \
-          	$lib_script $prev_all \
-          	$(find "$DESTINATION" -type f -name "*.md"))
+wait
+echo '}' >> meta.begin.awk
 
-echo '$0 != 0 {print}' > "$CSG_TMP_DIR/csg-print.awk"
-echo 'BEGIN { FILENAME = truefilename } NR == 1 { FILENAME = truefilename }' > "$CSG_TMP_DIR/csg-filename.awk"
+### Stage 3
+### Convertion
 
-transform_file() {
+convert_file() {
 	local file="$1"
 	local outf="${file%md}html"
 
-	local created edited
-	if [ -f "$CSG_TIMESTAMP_FILE" ]
-	then
-		{ read -r created; read -r edited; } << EOF
-$($CSG_AWK -v file="$outf" '
-	BEGIN { FS = "\t" }
-	$1 == file {
-		print $2 ORS $3
-		exit
-	}' "$CSG_TIMESTAMP_FILE")
-EOF
-	fi
-
-	# prev_md
-	local_exports=$($CSG_AWK\
-					-f	"$CSG_TMP_DIR/csg-export.awk"\
-					$lib_script	$prev_md\
-					"$file")
-
-	# on_md and on_html
 	$CSG_AWK \
-		-v created="$created" \
-		-v edited="$edited" \
-		$exports $local_exports \
-		$lib_script $on_md -f "$CSG_TMP_DIR/csg-print.awk" \
-		"$file" \
-	| $CSG_MD_TO_HTML \
-	| cat "$SOURCE/_header.html" - "$SOURCE/_footer.html" \
-	| $CSG_AWK \
-		-v created="$created" \
-		-v edited="$edited" \
-		$exports $local_exports \
-		-v truefilename="$outf" -f "$CSG_TMP_DIR/csg-filename.awk" \
-		$lib_script $on_html -f "$CSG_TMP_DIR/csg-print.awk" \
-	> "$outf"
-
-	# For hardlinks
-	cat "$outf" > "$file"
-	mv "$file" "$outf"
+		-f meta.begin.awk -f lib_stack.awk -f single.awk \
+		"$file" > "$outf"
+	rm "$file"
 }
 
-tempifs=$IFS
-IFS='
-	'
-
 while read -r file
 do
-	transform_file "$file" &
+	convert_file "$file" &
 done << EOF
 $(find "$DESTINATION" -type f -name "*.md")
 EOF
-
-IFS=$tempifs
-
-if [ -n "$on_other" ]
-then
-	transform_other() {
-		local file="$1"
-
-		$CSG_AWK \
-			$exports $local_exports \
-			$lib_script $on_other -f "$CSG_TMP_DIR/csg-print.awk" \
-			"$file" > temp
-		mv temp "$file"
-	}
-
-	while read -r file
-	do
-		if [ -f "$file" ]
-		then
-			transform_other "$file" &
-		fi
-	done << EOF
-$(find "$DESTINATION" -type f -not -name "*.md")
-EOF
-fi
-
 wait
-
-overwriteLine "[Stage 2] Generating sitemap"
-  # This cannot really be done while transforming files, since you could have static .html pages
-
-$CSG_AWK -v url="$URL" -v dest="$DESTINATION" -v date="$(date +%Y-%m-%d)" '
-	BEGIN {
-		print "<?xml version=\"1.0\" encoding=\"UTF-8\"?>" ORS \
-		      "<urlset" ORS \
-		      "xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\"" ORS \
-		      "xsi:schemaLocation=\"http://www.sitemaps.org/schemas/sitemap/0.9" ORS \
-		      "http://www.sitemaps.org/schemas/sitemap/0.9/sitemap.xsd\"" ORS \
-		      "xmlns=\"http://www.sitemaps.org/schemas/sitemap/0.9\">"
-		lendest = length(dest)
-	}
-	{
-		$0 = substr($0, index($0, dest) + lendest)
-		print "<url><loc>" url $0 "</loc><lastmod>" date "</lastmod><priority>1.0</priority></url>"
-	}
-	END { print "</urlset>" }
-'<< EOF > "$DESTINATION/sitemap.xml"
-$(find "$DESTINATION" -type f -name "*.html")
-EOF
-
-endTimer "[Stage 2] Done"
diff --git a/meta.awk b/meta.awk
new file mode 100644
index 0000000..98651fd
--- /dev/null
+++ b/meta.awk
@@ -0,0 +1,36 @@
+function emit(ind, val) {
+	print "meta[\"" FILENAME "\", \"" ind "\"]=\"" val "\""
+}
+
+title != "" && stopdesc == 0 && /^[[:alpha:]]/ {
+	startedGathering = 1
+	# Take only anchor text
+	gsub(/\]\([^)]*\)/, "")
+	gsub(/\[/, "")
+	# Append line
+	if (desc == "") desc = $0
+	else desc = desc " " $0
+	# Trim line to 200 characters
+	if (length(desc) > 200) {
+		desc = substr(desc, 1, 197) "..."
+		stopdesc = 1
+	}
+}
+stopdesc == 0 && startedGathering == 1 && /^[[:blank:]]*$/ { stopdesc = 1 }
+
+{
+	words += NF
+}
+
+title == "" && /^# / {
+	sub(/^#[[:blank:]]*/, "")
+	sub(/[[:blank:]#]*$/, "")
+	# sub(/.*/, titlef)
+	title = $0
+}
+
+END {
+	emit("title", title)
+	emit("description", desc)
+	emit("words", words)
+}