Merge branch 'rss_1' into rss_1

Reviewed-on: #3

example/Makefile

@@ -21,7 +21,8 @@ settings:
 
 .PHONY: clean
 clean: 
-	rm ./pages/*.html 
+	- rm -f ./pages/*.html 
-	rm ./tags/*/*.html 
+	- rm -f ./tags/*/*.html 
-	rm ./webring/*.html 
+	- rm -f ./webring/*.html 
-	rmdir ./pages ./tags/* ./tags ./webring
+	- rm -f ./feeds/*.xml
+	- rmdir ./pages ./tags/* ./tags ./webring ./feeds

example/settings.toml

@@ -85,3 +85,4 @@ title="Your Microblog Title Here"
 title_tagged = "#{__tagname__} on {__title__}"
 url = "https://yourdomain.tld/microblog/"
 description = "Your description here."
+trailing_punctuation = ".,)!?>];:"

src/check-settings.py

@@ -1,4 +1,3 @@
-
 import os, argparse
 from tomlkit import loads
 from tomlkit import dump
@@ -34,7 +33,7 @@ class MicroblogConfig:
 
     def check(self, r, u): # (reference, user)
         for key in r:
-            if key == "latestpages": continue;
+            if key == "latestpages": continue
             # post and webring have subtables
             #   webring.profile
             #   webring.following

src/microblog.py

@@ -549,7 +549,11 @@ if __name__ == "__main__":
         updated = []
         updated += writepage(tpl, tl, tc, cfg["page"], 
             paginate=True if new_posts is None else False)
+        cfg["rss"]["accepted_images"] = cfg["post"]["accepted_images"]
         if cfg["rss"]["enabled"]:
+            # ensure output directory for feeds exists
+            if not os.path.exists("feeds"):
+                os.mkdir("feeds")
             feed_filename = "feeds/__index__.xml"
             rss.write_feed(
                 p[:cfg["rss"]["limit"]], \

src/rss.py

@@ -1,4 +1,3 @@
-
 from microblog import My_Html_Parser
 # from html.parser import HTMLParser
 from html import escape
@@ -7,12 +6,12 @@ try:
     import feedgenerator
 except ImportError:
     traceback.print_exc()
-    def write_feed(posts, filename, params):
+    def write_feed(posts, filename, params, tagname=None):
         pass
     print("\trss disabled - missing dependency", file=sys.stderr)
     print("\tYour microblog still may have been generated.", file=sys.stderr)
 
-def line2words(lines, limit):
+def line2words(lines, limit, trailing_punctuation):
     output = []
     char_count = 0
     break_outer_loop = False
@@ -22,7 +21,7 @@ def line2words(lines, limit):
             l = len(word)
             if limit > 0 and (l + char_count > limit):
                 break_outer_loop = True
-                break;
+                break
             output.append(word)
             char_count += l
         if break_outer_loop:
@@ -30,15 +29,37 @@ def line2words(lines, limit):
     # last char of last word
     last_char = output[-1].strip()[-1]
     # print(output[-1], file=sys.stderr)
-    punctuation = [".", ")", ",", "!", "?", ">", ']']
+    if last_char not in trailing_punctuation:
-    if last_char not in punctuation:
         output.append("...")
     return output
 
+def _is_image_token(token: str, extensions):
+    parts = token.rsplit('.', 1)
+    return len(parts) == 2 and parts[1] in extensions
+
 # this is similar to the markup function in microblog
-def enrich_msg(lines, is_atom=True):
+def enrich_msg(
+    lines,
+    is_atom=True,
+    accepted_images=[],
+    base_url="",
+    trailing_punctuation="",
+    desc_len_limit=-1,
+):
     if not is_atom:
-        return string
+        from urllib.parse import urljoin
+        words = line2words(lines, desc_len_limit, trailing_punctuation)
+        for i in range(len(words)):
+            token = words[i]
+            core = token.rstrip(trailing_punctuation)
+            suffix = token[len(core):]
+            if len(core) == 0 or "<" in core or ">" in core:
+                continue
+            if _is_image_token(core, accepted_images):
+                abs_url = urljoin(base_url, core)
+                anchor = f"<a href=\"{escape(abs_url)}\">{escape(abs_url)}</a>"
+                words[i] = anchor + suffix
+        return words
     content = []
     parser = My_Html_Parser([])
     for line in lines:
@@ -66,9 +87,9 @@ def enrich_msg(lines, is_atom=True):
             content.append(line)
     return content
 
-def write_feed(posts, filename, params, tagname=str()):
+def write_feed(posts, filename, params, tagname=None):
     feed = None
-    if tagname != str():
+    if tagname:
         t = params["title_tagged"].format(
             __tagname__ = tagname,
             __title__ = params["title"])
@@ -96,12 +117,17 @@ def write_feed(posts, filename, params, tagname=str()):
         # len of post.message is number of lines
         msg = post.message
         ti = " ".join(
-            line2words(msg,TITLE_LEN_LIMIT))
+            line2words(msg, TITLE_LEN_LIMIT, params["trailing_punctuation"]))
-        if params["use_atom"]:
-            de = " ".join(enrich_msg(msg))
-        else:
         de = " ".join(
-                line2words(msg,DESC_LEN_LIMIT))
+            enrich_msg(
+                msg,
+                is_atom=params["use_atom"],
+                base_url=params["url"],
+                accepted_images=params["accepted_images"],
+                trailing_punctuation=params["trailing_punctuation"],
+                desc_len_limit=DESC_LEN_LIMIT,
+            )
+        )
         li = base_url + ("#%i" % post.num)
         p = dateutil.parser.parse(post.timestamp)
         if params["use_atom"]: