From: John Runyon <redacted>
Date: Mon, 4 Sep 2023 00:20:38 +0000 (-0600)
Subject: urls: add content-length and content-type
X-Git-Url: https://jfr.im/git/erebus.git/commitdiff_plain/ecbed32887af35897bfd78e2289d6253a21c2559

urls: add content-length and content-type
---

diff --git a/modules/urls.py b/modules/urls.py
index ae80a4e..80ae1ee 100644
--- a/modules/urls.py
+++ b/modules/urls.py
@@ -81,7 +81,7 @@ def process_line(line):
 	responses = []
 	num_found = 0
 	limit = lib.parent.cfg.getint('urls', 'limit', 2)
-	for action, group, prefix in regexes:
+	for action, group in regexes:
 		for regex in group:
 			for match in regex.findall(line):
 				if match:
@@ -92,8 +92,8 @@ def process_line(line):
 						resp = action(match)
 					else:
 						resp = action(*match)
-					if resp is not None:
-						responses.append("%s: %s" % (prefix, resp))
+					if resp is not None and resp != "":
+						responses.append(resp)
 	return responses
 
 @lib.hooknum("PRIVMSG")
@@ -206,33 +206,80 @@ def gottwitch(uri):
 	except:
 		return 'Channel offline.'
 
+def _humanize_bytes(b):
+	b = int(b)
+	i = 0
+	table = " kMGTPEZYRQ"
+	while b > 1024:
+		i += 1
+		b /= 1024.0
+	if i == 0:
+		return "%dB" % (b)
+	else:
+		return "%.2f%siB" % (b, table[i])
+
 def goturl(url):
-	for _, group, _ in other_regexes:
+	output = []
+	for _, group in other_regexes:
 		for regex in group:
 			if regex.match(url):
 				return None
 	request = urllib2.Request(url, headers={'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36'})
 	opener = urllib2.build_opener(SmartRedirectHandler())
+
+	# Send request and handle errors
 	try:
-		soup = BeautifulSoup(opener.open(request, timeout=2))
-		if soup.title:
-			return unescape('%s' % (soup.title.string))
-		else:
-			return None
+		response = opener.open(request, timeout=2)
 	except urllib2.HTTPError as e:
-		return 'Error: %s %s' % (e.code, e.reason)
+		return 'Request error: %s %s' % (e.code, e.reason)
 	except urllib2.URLError as e:
-		return 'Error: %s' % (e.reason)
+		return 'Request error: %s' % (e.reason)
 	except TimeoutError as e:
-		return 'Error: request timed out'
+		return 'Request error: request timed out'
 	except Exception as e:
-		return 'Error: %s %r' % (type(e).__name__, e.args)
+		return 'Unknown error: %s %r' % (type(e).__name__, e.args)
+
+	# Try to add type and length headers to reply
+	c_type = response.getheader('Content-Type', '').split(';', 1)[0]
+	c_len = response.getheader('Content-Length')
+	if c_type != '':
+		output.append("[%s] " % (c_type))
+	else:
+		output.append("[no type] ")
+	if c_type != "text/html": # else length will be provided by HTML code below
+		if c_len is not None:
+			output.append("[%s] " % (_humanize_bytes(c_len)))
+		else:
+			output.append("[no length] ")
+
+	# Try to add title if HTML
+	if c_type == 'text/html':
+		try:
+			responsebody = response.read(1024*1024)
+			print(type(responsebody))
+		except Exception as e:
+			output.append('Error reading response body: %s %r' % (type(e).__name__, e.args))
+		else:
+			if c_len is not None and len(responsebody) != int(c_len):
+				output.append("[actual %s; Content-Length %s] " % (_humanize_bytes(len(responsebody)), _humanize_bytes(c_len)))
+			else:
+				output.append("[%s] " % (_humanize_bytes(len(responsebody))))
+			try:
+				soup = BeautifulSoup(responsebody)
+				if soup.title:
+					output.append('Title: ' + unescape('%s' % (soup.find('title').string)))
+				else:
+					output.append('No title')
+			except Exception as e:
+				output.append('Title error: %s %r ' % (type(e).__name__, e.args))
+
+	return ''.join(output)
 
 url_regex = (
-	re.compile(r'https?://[^/\s]+\.[^/\s]+(?:/\S+)?'),
+	re.compile(r'https?://(?:[^/\s.]+\.)+[^/\s.]+(?:/\S+)?'),
 )
 other_regexes = (
 )
 regexes = other_regexes + (
-	(goturl, url_regex, 'Title'),
+	(goturl, url_regex),
 )