squeeze

commit 86beb3defce8569f43265b1b10f79f413edf9dca
parent 666b1fab707b27abf35d802748c5c1445f0ddf66
Author: St John Karp <contact@stjo.hn>
Date:   Sun, 17 May 2020 13:29:04 -0500

Generate RSS from HTML output

Generate the RSS feed by parsing the HTML output, which allows
us to get the generated HTML instead of posting raw Markdown.

Diffstat:
M
generate_rss.pl
 | 
18
+++++++++---------
M
html.pl
 | 
3
++-
M
squeeze.sh
 | 
5
+++--

3 files changed, 14 insertions(+), 12 deletions(-)
diff --git a/generate_rss.pl b/generate_rss.pl
@@ -4,7 +4,7 @@
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 
 :- include('helpers.pl').
-:- include('markdown.pl').
+:- include('html.pl').
 :- include('rss.pl').
 
 % generate_rss(+BuildDate, +Filenames).
@@ -29,12 +29,12 @@ files_to_articles([], []).
 
 files_to_articles([Filename|Filenames], [article(Date, Title, Link, Description)|Articles]):-
 	open(Filename, read, Stream),
-	read_file(Stream, Markdown),
+	read_file(Stream, HTML),
 	close(Stream),
 	% Grab the link.
 	get_link(Filename, Link),
-	% Extract the title, entry, etc. from the Markdown.
-	markdown(Entry, Title, _, Date, Markdown, []),
+	% Extract the title, entry, etc. from the HTML.
+	page(Entry, Title, _, Date, HTML, []),
 	% XML escape the description.
 	replace("&", "&", Entry, EntryAmp),
 	replace("<", "&lt;", EntryAmp, EntryLT),
@@ -48,18 +48,18 @@ 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_lists(_, "index.md", FilenameCodes),
+	append_lists(_, "index.html", FilenameCodes),
 	site_url(URL, []),
-	append_lists(_, "/source", StartPath),
+	append_lists(_, "/output", StartPath),
 	append_lists(StartPath, Path, FilenameCodes),
-	append_lists(PathWithoutFile, "index.md", Path),
+	append_lists(PathWithoutFile, "index.html", Path),
 	append_lists(URL, PathWithoutFile, Link).
 
 get_link(Filename, Link):-
 	atom_codes(Filename, FilenameCodes),
 	site_url(URL, []),
-	append_lists(_, "/source", StartPath),
+	append_lists(_, "/output", StartPath),
 	append_lists(StartPath, Path, FilenameCodes),
-	append_lists(PathWithoutExtension, ".md", Path),
+	append_lists(PathWithoutExtension, ".html", Path),
 	append_lists(PathWithoutExtension, "/", PathWithSlash),
 	append_lists(URL, PathWithSlash, Link).
 \ No newline at end of file
diff --git a/html.pl b/html.pl
@@ -7,7 +7,8 @@
 page(Entry, Title, Subtitle, Date) -->
 	doctype,
 	newline,
-	html(Entry, Title, Subtitle, Date).
+	html(Entry, Title, Subtitle, Date),
+	newline.
 
 html(Entry, Title, Subtitle, Date) -->
 	html_open,
diff --git a/squeeze.sh b/squeeze.sh
@@ -58,6 +58,9 @@ ARTICLES=$(grep --recursive --include=\*.md "^Date: " "$SITE_PATH/$SOURCE_DIR" |
 	cut --fields=2 |
 	# Get the last (i.e. most recent) posts for the RSS feed.
 	tail -5 |
+	# Convert paths so we operate on the generated HTML, not the unformatted Markdown.
+	sed "s|^$SITE_PATH/$SOURCE_DIR|$SITE_PATH/$OUTPUT_DIR|" |
+	sed 's|.md$|.html|' |
 	# Glue the file names together to be passed to Prolog.
 	paste --serial --delimiters=',' - |
 	sed "s|,|','|g")
@@ -68,4 +71,3 @@ swipl --traditional --quiet -l generate_rss.pl -g "consult('$SITE_PATH/site.pl')
 	# Strip everything before the XML declaration.
 	awk "/<?xml/{i++}i" \
 	> "$SITE_PATH/$OUTPUT_DIR/feeds/rss.xml"
-	
-\ No newline at end of file