Comparison

mod_pubsub_summary/mod_pubsub_summary.lua @ 4426:3fe2c264aac4

mod_pubsub_summary: Mangle HTML payloads in Atom/RSS feeds This module is meant for use with mod_pubsub_feeds and tries to improve on mod_pubsub's built-in Atom summary generator.
author Kim Alvefur <zash@zash.se>
date Thu, 04 Feb 2021 01:12:41 +0100
child 4435:a620bf249e63
comparison
equal deleted inserted replaced
4425:b3e0295e14a3 4426:3fe2c264aac4
1 -- No, not trying to parse HTML here. It's an illusion. Just trying to read RSS feeds.
2 --
3 -- Compose a textual representation of Atom payloads
4 module:hook("pubsub-summary/http://www.w3.org/2005/Atom", function (event)
5 local payload = event.payload;
6 local title = payload:get_child_text("title");
7 local content_tag = payload:get_child("content") or payload:get_child("summary");
8 local content = content_tag:get_text();
9 if content_tag.attr.type == "html" then
10 content = content:gsub("\n*<p[^>]*>\n*(.-)\n*</p>\n*", "%1\n\n");
11 content = content:gsub("<li>(.-)</li>\n", "* %1\n");
12 content = content:gsub("<a[^>]*href=[\"'](.-)[\"'][^>]*>(.-)</a>", "%2 <%1>");
13 content = content:gsub("<b>(.-)</b>", "*%1*");
14 content = content:gsub("<strong>(.-)</strong>", "*%1*");
15 content = content:gsub("<em>(.-)</em>", "*%1*");
16 content = content:gsub("<i>(.-)</i>", "*%1*");
17 content = content:gsub("<img[^>]*src=[\"'](.-)[\"'][^>]*>", " %1 "); -- TODO alt= would have been nice to grab
18 content = content:gsub("<br[^>]*>", "\n");
19 content = content:gsub("<[^>]+>", "");
20 content = content:gsub("^%s*", ""):gsub("%s*$", "");
21 content = content:gsub("\n\n\n+", "\n\n");
22 content = content:gsub("&(%w+);", {
23 apos = "'";
24 quot = '"';
25 lt = "<";
26 gt = ">";
27 amp = "&";
28 nbsp = utf8 and utf8.char(0xa0) or " ";
29 });
30 end
31 local link = payload:get_child("link");
32 local summary;
33 if title and content then
34 summary = title .. "\n\n" .. content;
35 elseif title or content then
36 summary = content or title;
37 end
38 if link and link.attr.href and link.attr.href ~= content then
39 summary = (summary and summary .. "\n" or "") .. link.attr.href;
40 end
41 return summary;
42 end, 1);