Fix time formatting and other small format changes

sgu/data/templates/dumbest.j2x

@@ -10,4 +10,5 @@
 |article_title = ((( article_title )))
 |publication = ((( publication )))
 }}
+
 ((( transcript )))

sgu/data/templates/email.j2x

@@ -7,4 +7,5 @@
 text: ((( item )))
 ((* endfor *))
 -->
+
 ((( transcript )))

sgu/data/templates/interview.j2x

@@ -1,4 +1,4 @@
 {{anchor|interview}}
 == Interview with ((( name ))) <small>( ((( start_time ))) )</small> ==
-((( source )))
+
 ((( transcript )))

sgu/data/templates/logical_fallacy.j2x

@@ -1,4 +1,4 @@
 {{anchor|ntlf}}
 == Name That Logical Fallacy <small>( ((( start_time ))) )</small> ==
-((( source )))
+
 ((( transcript )))

sgu/data/templates/news.j2x

@@ -1,9 +1,9 @@
 {{anchor|news((( item_number )))}}
-=== News Item #((( item_number ))) - ((( topic ))) <small>( ((( start_time ))) )</small> ===
+== News Item #((( item_number ))) - ((( topic ))) <small>( ((( start_time ))) )</small> ==
 {{shownotes
 |weblink = ((( url )))
 |article_title = ((( topic ))) ((# TODO: Replace with title #))
 |publication = ((( publication )))
 }}
-((( source )))
+
 ((( transcript )))

sgu/data/templates/noisy.j2x

@@ -1,5 +1,5 @@
 {{anchor|wtn}}
 {{anchor|futureWTN}}
 == Who's That Noisy? + Announcements <small>( ((( start_time ))) )</small> ==
-((( source )))
+
 ((( transcript )))

sgu/data/templates/quickie.j2x

@@ -1,4 +1,4 @@
 {{anchor|quickie}}
 == ((( title ))): ((( subject ))) <small>( ((( start_time ))) )</small> ==
-((( source )))
+
 ((( transcript )))

sgu/data/templates/quote.j2x

@@ -1,10 +1,10 @@
 {{anchor|qow}}
 == Skeptical Quote of the Week <small>( ((( start_time ))) )</small> ==
-((( source )))
 {{qow
 |text = ((( quote )))
 |author = ((( attribution )))
 |lived =
 |desc =
 }}
+
 ((( transcript )))

sgu/data/templates/science_or_fiction.j2x

@@ -1,7 +1,6 @@
 {{anchor|sof}}
 {{anchor|theme}}
 == Science or Fiction <small>( ((( start_time ))) )</small> ==
-((( source )))
 {{SOFinfo
 |theme = ((( theme )))
 |hiddentheme =
@@ -31,4 +30,5 @@
 |win = <!-- at least one Rogue guessed wrong, but not them all -->
 |swept = <!-- all the Rogues guessed right -->
 }}
+
 ((( transcript )))

sgu/data/templates/tiktok.j2x

@@ -1,5 +1,5 @@
 {{anchor|tiktok}}
 == From TikTok: ((( title ))) <small>( ((( start_time ))) )</small> ==
-((( source )))
 ((( url )))
+
 ((( transcript )))

sgu/data/templates/unknown.j2x

@@ -1,3 +1,3 @@
 == ((( title ))) <small>( ((( start_time ))) )</small> ==
-((( source )))
+
 ((( transcript )))

sgu/data/templates/whats_the_word.j2x

@@ -1,5 +1,5 @@
 {{anchor|wtw}}
 == What's the Word? <small>( ((( start_time ))) )</small> ==
-((( source )))
 * ((( word )))
+
 ((( transcript )))

sgu/episode_segments.py

@@ -10,7 +10,7 @@
 from sgu.exceptions import StartTimeNotFoundError
 from sgu.helpers import are_strings_in_string, find_single_element, string_is_url
 from sgu.template_environment import template_env
-from sgu.transcript_formatting import format_transcript_for_wiki
+from sgu.transcript_formatting import format_time, format_transcript_for_wiki
 
 if TYPE_CHECKING:
     from sgu.transcription import DiarizedTranscript
@@ -62,7 +62,7 @@ def to_wiki(self) -> str:
         template = template_env.get_template(f"{self.template_name}.j2x")
         template_values = self.get_template_values()
         return template.render(
-            start_time=self.start_time,
+            start_time=format_time(self.start_time),
             transcript=format_transcript_for_wiki(self.transcript),
             source=f"<!-- data source: {self.source.value} -->",
             **template_values,

sgu/transcript_formatting.py

@@ -9,19 +9,28 @@ def format_transcript_for_wiki(transcript: "DiarizedTranscript") -> str:
     transcript = _join_speaker_segments(transcript)
     _abbreviate_speakers(transcript)
 
-    text_segments: list[str] = []
-    for transcript_chunk in transcript:
-        start_time = _format_time(transcript_chunk["start"])
-        end_time = _format_time(transcript_chunk["end"])
+    text_segments = [
+        f"'''{transcript_chunk['speaker']}''':{transcript_chunk['text']}" for transcript_chunk in transcript
+    ]
 
-        text_segments.append(f"<!-- {start_time} - {end_time} -->")
-        text_segments.append(f"'''{transcript_chunk['speaker']}''':{transcript_chunk['text']}<br />")
+    return "\n\n".join(text_segments)
 
-    return "\n".join(text_segments)
 
+def format_time(time: float | None) -> str:
+    """Format a float time to h:mm:ss or mm:ss if < 1 hour."""
+    if not time:
+        return "???"
 
-def _format_time(time: float) -> str:
-    return f"{int(time) // 3600:02d}:{int(time) // 60 % 60:02d}:{int(time) % 60:02d}"
+    hour_count = int(time) // 3600
+
+    hour = ""
+    if hour_count:
+        hour = f"{hour_count}:"
+
+    minutes = f"{int(time) // 60 % 60:02d}:"
+    seconds = f"{int(time) % 60:02d}"
+
+    return f"{hour}{minutes}{seconds}"
 
 
 def _join_speaker_segments(transcript: "DiarizedTranscript") -> "DiarizedTranscript":

sgu/wiki.py

@@ -47,7 +47,8 @@ async def create_podcast_wiki_page(client: "Session", podcast: "PodcastEpisode")
 
     # Above: Generic actions
     # Below: Wiki-specific actions
-    # convert segments to wiki
+    # we must grab speaker data before we convert transcript to wiki
+    speakers = {s["speaker"].lower() for s in episode_data.transcript}
     wiki_segments = "\n".join(s.to_wiki() for s in episode_segments)
     qotw_segment = _extract_quote_of_the_week_for_wiki(episode_segments)
 
@@ -58,8 +59,6 @@ async def create_podcast_wiki_page(client: "Session", podcast: "PodcastEpisode")
         episode_image_url = get_episode_image_url(episode_data.show_notes)
         episode_icon_name = _upload_image_to_wiki(client, episode_image_url, episode_data.podcast.episode_number)
 
-    speakers = {s["speaker"].lower() for s in episode_data.transcript}
-
     logger.debug("Creating wiki page...")
     wiki_page = _construct_wiki_page(episode_data, episode_icon_name, wiki_segments, qotw_segment, speakers)
     _edit_page(client, page_text=wiki_page)  # TODO: Change for "Create page"