squeeze

commit 6b21fede3a2b3fc43782c91f9ef90c6d0cfd0a91
parent ba0061ce782c9ac4e7754f9ddf66a9d739beb29d
Author: St John Karp <contact@stjo.hn>
Date:   Fri, 12 Jul 2019 20:17:04 -0500

Fixes after doing a bigger site

Added code and fixes to handle more complex scenarios after testing
this with a much bigger data source.

Diffstat:
M
html.pl
 | 
15
++++++++++++++-
M
markdown.pl
 | 
18
+++++++++++++++---
M
rss.pl
 | 
27
+++++++++++++++++++++------
M
tastic.sh
 | 
70
+++++++++++++++++++++++++++++++++++++---------------------------------

4 files changed, 87 insertions(+), 43 deletions(-)
diff --git a/html.pl b/html.pl
@@ -61,13 +61,26 @@ article(Entry, Title, Subtitle, Date) -->
 	article_close,
 	{ [First|_] = Entry,  char_code('<', First) }.
 
+% An article without a title, subtitle, or metadata.
 article_header(null, null, null) --> [].
 
+% An article without a subtitle or metadata.
+article_header(Title, null, null) -->
+	article_title(Title).
+
+% An article without a subtitle.
 article_header(Title, null, Date) -->
 	article_title(Title),
 	whitespace,
 	article_meta(Date).
 
+% An article without metadata.
+article_header(Title, Subtitle, null) -->
+	article_title(Title),
+	whitespace,
+	article_subtitle(Subtitle).
+
+% An article with all header components.
 article_header(Title, Subtitle, Date) -->
 	article_title(Title),
 	whitespace,
@@ -107,7 +120,7 @@ title(null) -->
 
 title(Title) -->
 	"<title>",
-	Title,
+	anything(Title),
 	"</title>".
 
 title(_) -->
diff --git a/markdown.pl b/markdown.pl
@@ -1,5 +1,14 @@
 % Markdown definition
 
+markdown(Entry, Title, Subtitle, Date) -->
+	metadata("Title", Title),
+	"\n",
+	metadata("Subtitle", Subtitle),
+	"\n",
+	metadata("Date", Date),
+	"\n\n",
+	anything(Entry).
+
 markdown(Entry, Title, null, Date) -->
 	metadata("Title", Title),
 	"\n",
@@ -7,12 +16,15 @@ markdown(Entry, Title, null, Date) -->
 	"\n\n",
 	anything(Entry).
 
-markdown(Entry, Title, Subtitle, Date) -->
+markdown(Entry, Title, Subtitle, null) -->
 	metadata("Title", Title),
 	"\n",
 	metadata("Subtitle", Subtitle),
-	"\n",
-	metadata("Date", Date),
+	"\n\n",
+	anything(Entry).
+
+markdown(Entry, Title, null, null) -->
+	metadata("Title", Title),
 	"\n\n",
 	anything(Entry).
 
diff --git a/rss.pl b/rss.pl
@@ -17,12 +17,7 @@ files_to_articles([Filename|Filenames], [article(Date, Title, Link, Description)
 	read_file(Stream, Markdown),
 	close(Stream),
 	% Grab the link.
-	atom_codes(Filename, FilenameCodes),
-	site_url(URL, []),
-	append(_, "/source", StartPath),
-	append(StartPath, Path, FilenameCodes),
-	append(PathWithoutFile, "index.md", Path),
-	append(URL, PathWithoutFile, Link),
+	get_link(Filename, Link),
 	% Extract the title, entry, etc. from the Markdown.
 	markdown(Entry, Title, _, Date, Markdown, []),
 	% XML escape the description.
@@ -31,6 +26,26 @@ files_to_articles([Filename|Filenames], [article(Date, Title, Link, Description)
 	replace(">", "&gt;", EntryLT, Description),
 	files_to_articles(Filenames, Articles).
 
+get_link(Filename, Link):-
+	atom_codes(Filename, FilenameCodes),
+	% Just assert that this is an index file before we go further.
+	% Backtracking after this point will take us down a rabbit hole.
+	append(_, "index.md", FilenameCodes),
+	site_url(URL, []),
+	append(_, "/source", StartPath),
+	append(StartPath, Path, FilenameCodes),
+	append(PathWithoutFile, "index.md", Path),
+	append(URL, PathWithoutFile, Link).
+
+get_link(Filename, Link):-
+	atom_codes(Filename, FilenameCodes),
+	site_url(URL, []),
+	append(_, "/source", StartPath),
+	append(StartPath, Path, FilenameCodes),
+	append(PathWithoutExtension, ".md", Path),
+	append(PathWithoutExtension, "/", PathWithSlash),
+	append(URL, PathWithSlash, Link).
+
 rss(BuildDate, Articles) -->
 	rss_open,
 	"\n",
diff --git a/tastic.sh b/tastic.sh
@@ -9,58 +9,62 @@ if [ "$1" == "ungenerate" ]
 then
 	# Create the directory structure.
 	rm -rf $SITE_PATH/$SOURCE_DIR/*
-	find $SITE_PATH/$OUTPUT_DIR -type d | \
-		sed "s|^$SITE_PATH/$OUTPUT_DIR|$SITE_PATH/$SOURCE_DIR|" | \
+	find $SITE_PATH/$OUTPUT_DIR -type d |
+		sed "s|^$SITE_PATH/$OUTPUT_DIR|$SITE_PATH/$SOURCE_DIR|" |
 		xargs -0 -d '\n' mkdir -p --
 
 	# Parse and create all the markdown files.
-	for file in `find $SITE_PATH/$OUTPUT_DIR -type f -name "*.html"`; do
-		NEW_PATH=`echo $file | sed "s|^$SITE_PATH/$OUTPUT_DIR|$SITE_PATH/$SOURCE_DIR|" | sed 's|.html$|.md|'`
-		cat $file | \
-			swipl --traditional -q -l entries.pl -g "parse_entry." | \
-			# Unsmarten the punctuation.
-			sed "s|&nbsp;| |g" | \
-			sed "s|&#8216;|'|g" | \
-			sed "s|&#8217;|'|g" | \
-			sed "s|&#8220;|\"|g" | \
-			sed "s|&#8221;|\"|g" \
-			> $NEW_PATH
-	done
+	find $SITE_PATH/$OUTPUT_DIR -type f -name "*.html" -print0 |
+		while IFS= read -r -d '' file; do
+			NEW_PATH=`echo "$file" | sed "s|^$SITE_PATH/$OUTPUT_DIR|$SITE_PATH/$SOURCE_DIR|" | sed 's|.html$|.md|'`
+			cat "$file" |
+				swipl --traditional -q -l entries.pl -g "consult('$SITE_PATH/site.pl'), parse_entry." |
+				# Unsmarten the punctuation.
+				sed "s|&nbsp;| |g" |
+				sed "s|&#8216;|'|g" |
+				sed "s|&#8217;|'|g" |
+				sed "s|&#8220;|\"|g" |
+				sed "s|&#8221;|\"|g" \
+				> "$NEW_PATH"
+		done
 
 	# Copy anything else directly.
-	for file in `find $SITE_PATH/$OUTPUT_DIR -type f -not -name "*.html"`; do
-		NEW_PATH=`echo $file | sed "s|^$SITE_PATH/$OUTPUT_DIR|$SITE_PATH/$SOURCE_DIR|"`
-		cp $file $NEW_PATH
-	done
+	find $SITE_PATH/$OUTPUT_DIR -type f -not -name "*.html" -print0 |
+		while IFS= read -r -d '' file; do
+			NEW_PATH=`echo "$file" | sed "s|^$SITE_PATH/$OUTPUT_DIR|$SITE_PATH/$SOURCE_DIR|"`
+			cp "$file" "$NEW_PATH"
+		done
 elif [ "$1" == "generate" ]
 then
 	# Create the directory structure.
 	rm -rf $SITE_PATH/$OUTPUT_DIR/*
-	find $SITE_PATH/$SOURCE_DIR -type d | \
-		sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|" | \
+	find $SITE_PATH/$SOURCE_DIR -type d |
+		sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|" |
 		xargs -0 -d '\n' mkdir -p --
 
 	# Parse and create all the HTML files.
-	for file in `find $SITE_PATH/$SOURCE_DIR -type f -name "*.md"`; do
-		NEW_PATH=`echo $file | sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|" | sed 's|.md$|.html|'`
-		cat $file | \
-			swipl --traditional -q -l entries.pl -g "consult('$SITE_PATH/site.pl'), generate_entry." | \
-			tidy -quiet --indent auto --indent-with-tabs yes --wrap 0 -xml --tidy-mark no | \
-			~/.local/bin/smartypants \
-			> $NEW_PATH
-	done
+	find $SITE_PATH/$SOURCE_DIR -type f -name "*.md" -print0 |
+		while IFS= read -r -d '' file; do
+			NEW_PATH=`echo "$file" | sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|" | sed 's|.md$|.html|'`
+			cat "$file" |
+				swipl --traditional -q -l entries.pl -g "consult('$SITE_PATH/site.pl'), generate_entry." |
+				tidy -quiet --indent auto --indent-with-tabs yes --wrap 0 -xml --tidy-mark no |
+				~/.local/bin/smartypants \
+				> "$NEW_PATH"
+		done
 
 	# Copy anything else directly.
-	for file in `find $SITE_PATH/$SOURCE_DIR -type f -not -name "*.md"`; do
-		NEW_PATH=`echo $file | sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|"`
-		cp $file $NEW_PATH
-	done
+	find $SITE_PATH/$SOURCE_DIR -type f -not -name "*.md" -print0 |
+		while IFS= read -r -d '' file; do
+			NEW_PATH=`echo "$file" | sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|"`
+			cp "$file" "$NEW_PATH"
+		done
 
 	# Generate the RSS feed.
 	mkdir -p $SITE_PATH/$OUTPUT_DIR/feeds
 	ARTICLES=`grep -Rl --include=\*.md "^Date: " $SITE_PATH/$SOURCE_DIR | paste -sd ',' - | sed "s|,|','|g"`
 	BUILD_DATE=`date +"%Y-%m-%d %T"`
-	swipl --traditional -q -l rss.pl -g "consult('$SITE_PATH/site.pl'), generate_rss(\"$BUILD_DATE\", ['$ARTICLES'])." | \
+	swipl --traditional -q -l rss.pl -g "consult('$SITE_PATH/site.pl'), generate_rss(\"$BUILD_DATE\", ['$ARTICLES'])." |
 		tidy -quiet --indent auto --indent-with-tabs yes --wrap 0 -xml --tidy-mark no \
 		> $SITE_PATH/$OUTPUT_DIR/feeds/rss.xml
 else