squeeze

commit 36e075c970751fc7efa2b0676a4548ac9328530d
parent bd3d10834ac8498123a5c49c56b2a12c5bc05dc3
Author: St John Karp <contact@stjo.hn>
Date:   Tue, 31 Mar 2020 12:41:24 -0500

Delete files for which the source was removed

Because we're no longer flushing and recreating everything, we need
to check to see if there are any output files that no longer exist
in the source. rsync takes care of this for assets, but we need
a manual implementation for Markdown/HTML. This works both ways,
both for squeezing and unsqueezing.

Diffstat:
M
squeeze.sh
 | 
15
++++++++++++++-
M
unsqueeze.sh
 | 
11
+++++++++++

2 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/squeeze.sh b/squeeze.sh
@@ -9,11 +9,24 @@ SITE_PATH=$1
 # This will also create the folder structure for the destination Markdown files.
 rsync --archive --delete --verbose --exclude "*.md" --exclude "*.html" --exclude "feeds" "$SITE_PATH/$SOURCE_DIR/" "$SITE_PATH/$OUTPUT_DIR/"
 
+# Delete any HTML files for which the source was removed.
+find "$SITE_PATH/$OUTPUT_DIR" -type f -name "*.html" -print0 |
+	while IFS= read -r -d '' file; do
+		OLD_PATH=`echo "$file" |
+			sed "s|^$SITE_PATH/$OUTPUT_DIR|$SITE_PATH/$SOURCE_DIR|" |
+			sed 's|.html$|.md|'`
+		if [ ! -f $OLD_PATH ]; then
+			rm $file
+		fi
+	done
+
 # Parse and create all the HTML files.
 find "$SITE_PATH/$SOURCE_DIR" -type f -name "*.md" -print0 |
 	while IFS= read -r -d '' file; do
 		echo $file
-		NEW_PATH=`echo "$file" | sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|" | sed 's|.md$|.html|'`
+		NEW_PATH=`echo "$file" |
+			sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|" |
+			sed 's|.md$|.html|'`
 		# Only process files whose destination doesn't exist, or which has been recently changed.
 		if [ ! -f $NEW_PATH ] || [[ $(find $file -mtime -7) ]]; then
 			# Get everything after the metadata and feed it through Pandoc.
diff --git a/unsqueeze.sh b/unsqueeze.sh
@@ -10,6 +10,17 @@ SITE_PATH=$1
 # This will also create the folder structure for the destination Markdown files.
 rsync --archive --delete --verbose --exclude "*.html" --exclude "*.md" --exclude "feeds" "$SITE_PATH/$OUTPUT_DIR/" "$SITE_PATH/$SOURCE_DIR/"
 
+# Delete any Markdown files for which the output was removed.
+find "$SITE_PATH/$SOURCE_DIR" -type f -name "*.md" -print0 |
+	while IFS= read -r -d '' file; do
+		OLD_PATH=`echo "$file" |
+			sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|" |
+			sed 's|.md$|.html|'`
+		if [ ! -f $OLD_PATH ]; then
+			rm $file
+		fi
+	done
+
 # Parse and create all the markdown files.
 find "$SITE_PATH/$OUTPUT_DIR" -type f -name "*.html" -print0 |
 	while IFS= read -r -d '' file; do