Add docstrings for all functions except main().

gemfeed.py

@@ -9,10 +9,32 @@ import urllib.parse
 from feedgen.feed import FeedGenerator
 
 def is_world_readable(filename):
+    """
+    Return True if the named file is world readable, otherwise return False.
+    """
     st = os.stat(filename)
     return st.st_mode & stat.S_IROTH
 
+def extract_first_heading(filename, default=""):
+    """
+    Open a file which is presumed to contain text/gemini content and return
+    the contents of the first heading line (regardless of heading level).
+    If no heading lines are found, return the specified default.
+    """
+    with open(filename) as fp:
+        for line in fp:
+            if line.startswith("#"):
+                while line[0] == "#":
+                    line = line[1:]
+                return line.strip()
+    return default
+
 def get_feed_title():
+    """
+    If an index.gmi or index.gemini file exists and is worldreadable, return
+    the content of the first heading line in the file, otherwise return a
+    default feed title.
+    """
     default = "Just another Gemini feed"
     for index_file in ("index.gmi", "index.gemini"):
         if os.path.exists(index_file) and is_world_readable(index_file):
@@ -20,6 +42,10 @@ def get_feed_title():
     return default
 
 def find_files(n=10):
+    """
+    Return the n most recently created world readable files with extensions of
+    .gmi or .gemini, as a list sorted from most to least recent.
+    """
     files = []
     for extension in ("gmi", "gemini"):
         files.extend(glob.glob("*.{}".format(extension)))
@@ -30,16 +56,31 @@ def find_files(n=10):
     files.sort(key=os.path.getctime, reverse=True)
     return files[0:n]
 
-def extract_first_heading(filename, default=""):
+def urljoin(base, url):
-    with open(filename) as fp:
+    """
-        for line in fp:
+    Return an absolute URL formed by combining the provided base and relative
-            if line.startswith("#"):
+    URLs.
-                while line[0] == "#":
+
-                    line = line[1:]
+    This is necessary because the various functions in Python's urllib to do
-                return line.strip()
+    this do not function as expected if the URL scheme is not recognised,
-    return default
+    which of course gemini:// is not.  Thus, we need to do a little dance
+    where we transform gemini URLs to https URLs, join them, and then undo
+    the transformation.
+    """
+    base = urllib.parse.urlsplit(base)
+    base = base._replace(scheme="https")
+    base = urllib.parse.urlunsplit(base)
+    joined = urllib.parse.urljoin(base, url)
+    joined = urllib.parse.urlsplit(joined)
+    joined = joined._replace(scheme="gemini")
+    return urllib.parse.urlunsplit(joined)
 
 def populate_entry_from_file(filename, base_url, entry):
+    """
+    Set the id, title, updated and link attributes of the provided
+    FeedGenerator entry object according the contents of the named
+    Gemini file and the base URL.
+    """
     url = urljoin(base_url, filename)
     entry.guid(url)
     entry.link(href=url, rel="alternate")
@@ -49,15 +90,6 @@ def populate_entry_from_file(filename, base_url, entry):
     title = extract_first_heading(filename, filename)
     entry.title(title)
 
-def urljoin(base, url):
-    base = urllib.parse.urlsplit(base)
-    base = base._replace(scheme="https")
-    base = urllib.parse.urlunsplit(base)
-    joined = urllib.parse.urljoin(base, url)
-    joined = urllib.parse.urlsplit(joined)
-    joined = joined._replace(scheme="gemini")
-    return urllib.parse.urlunsplit(joined)
-
 def main():
 
     # Get default title from index page, if there is one