Skip to content

archive.md

Latest commit

 

History

History
67 lines (50 loc) · 4.72 KB

archive.md

File metadata and controls

67 lines (50 loc) · 4.72 KB
Raw

Archive

Every prior edition lives under /archive/{year}/. The structure mirrors the current edition and stays stable across re-imports, so external links to old talks and speakers never break.

URL layout

/archive/                                            # edition index
/archive/{year}/                                     # one edition (Event JSON-LD, EventCompleted)
/archive/{year}/talks/                               # talk list for that edition
/archive/{year}/talks/{slug}/                        # one talk (canonical)
/archive/{year}/talks/{CODE}/                        # 301 redirect → slug URL (HTML meta-refresh + canonical)
/archive/{year}/speakers/                            # speaker list
/archive/{year}/speakers/{CODE}/                     # one speaker (Person JSON-LD)
/archive/{year}/tracks/                              # track index for the edition
/archive/{year}/tracks/{slug}/                       # talks in one track (CollectionPage JSON-LD)

Slugs are derived from titles at migration time and frozen in each talk's contents.lr (slug: field). Pretalx codes remain the stable internal identifier — speakers still reference talks by code, and the redirect from /talks/{CODE}/ to /talks/{slug}/ keeps any externally-shared code URL working.

Editions currently archived: 2018, 2019, 2022, 2023, 2024, 2025. (Gap years had no PyCon DE.)

Anatomy of an archived talk

A talk's content/archive/{year}/talks/{slug}/contents.lr carries:

Field Source Notes
code Pretalx Immutable Pretalx submission code, used for cross-references
slug derived Slugified title; frozen after first migration
title, abstract, description, track, room, day, start_time, duration Pretalx Re-imports overwrite
youtube_id, video_link, video_thumbnail, video_duration_iso, recording_available YouTube sync See docs/recordings.md
transcript, transcript_status, transcript_language transcript import See docs/transcripts.md
do_not_record Pretalx Suppresses both video embed and transcript display

Speakers list talks by slug (post-migration). Talks list speakers via the speakers index (lookup by Pretalx code stored on the speaker record).

Common tasks

Task Doc
Re-import speakers or talks from Pretalx for an archived year docs/pretalx.md (use --year YYYY on the importer)
Link more YouTube recordings to an archived edition docs/recordings.md
Import transcripts for a previously-skipped edition docs/transcripts.md
Regenerate track pages after the talks list changed uv run python utils/generate_tracks.py
Add a Pretalx-code redirect after renaming a slug uv run python utils/migrate_pretalx_slugs.py (idempotent)
Add a manual /old/ → /new/ redirect docs/redirects.md
Verify cross-edition speaker continuity speaker.html iterates each year's /speakers/{CODE}/; same Pretalx code in multiple years renders the "Also at PyCon DE & PyData" card

Cross-edition integrity

A few invariants the build relies on:

  • Pretalx codes are unique within a single edition but not stable across editions — Pretalx mints fresh codes per event. The "Also at" card on a speaker page therefore only catches direct code matches (which Pretalx does sometimes preserve when a speaker re-applies through the same account). A databags/speaker_aliases.yaml to bridge non-matching codes is an open item — see docs/seo-crosslinking-concept.md §7.
  • Track names drift between editions ("DevOps" → "DevOps & MLOps" → "MLOps & DevOps"). Each edition's track index is its own page; tracks are not merged across years.
  • Slug URLs are immutable. If a talk title changes after the slug is committed, the slug stays — preventing 404s. The displayed title updates normally.

Adding a previously-skipped edition

The 2020 and 2021 editions don't exist (cancelled / online-only). If a prior edition needs to be added retroactively:

  1. Provide a Pretalx event slug; record it in databags/pretalx.yaml under that year alongside the year-specific question and submission-type IDs.
  2. Run uv run python utils/talks.py --year YYYY and uv run python utils/speakers.py --year YYYY.
  3. Run the slug migration to produce slug URLs + Pretalx-code redirects: uv run python utils/migrate_pretalx_slugs.py.
  4. Run track generation: uv run python utils/generate_tracks.py.
  5. Optional: import transcripts and link recordings — same workflow as a post-conference cutover.
  6. Add the edition to the footer year row in templates/layout.html and to databags/past_editions.yaml (homepage card).