Current Status

This is a very early draft, many of the core chunks have technical documentation, but most are missing explanatory text. Each chunk also has its own status.

Change Log

If you find an error or would like additional information added, please check the issue tracker and comment on an existing issue or add a new one. Thank you in advance!

Release template

[version] - [release_date] - [summary]
- [things that were done]

Annotations: NEW, IMProved, FIX, DEPrecated, REMoved, SECurity

Releases

Upcoming - release_date -
- [NEW] add BEXT1996:bext
- [FIX] correct md5 chunk name in reference data
0.2.0 - 2024-06-10 - Add RIFF1994 chunks.
- [IMP] update fmt to document WAVEFORMATEX tags: WAVE_FORMAT_UNKNOWN, WAVE_FORMAT_ADPCM, and WAVE_FORMAT_DVI_ADPCM.
- [IMP] add language and dialect codes from RIFF1994
- [IMP] compare wFormatTags table on RIFF1994 pg20 to current list
- [NEW] add RIFF1994:smpl
- [IMP] add table of fields defined by each specifcation on its spec page.
- [IMP] refactor source spec data to reduce duplication (combining WRID mapped name & type with orginal spec name & type). Show both names and types in full WRID list.
- [IMP] documentation of repeated sub-structs: ex: plst, cue, smpl
- [NEW] add RIFF1994:inst
- [NEW] 2 new INFO chunks from RIFF1994: ISMP, IDIT
- [FIX] PAD is actually from RIFF1994
0.1.4 - 2024-03-29 - Finish RIFF1991 chunks
- show more of the WRID prefix on subchunks
- add RIFF1991:LIST-wavl
- add RIFF1991:slnt
- add MediaArea:MD5
- add RIFF1991:LIST-adtl
- add RIFF1991:JUNK (also, PAD, FLLR)
- add RIFF1991:plst
- add RIFF1991:cue
- add chunk RIFF1991:fact
- add chunk RIFF1991:CSET
- add RIFF1991: country codes
- add RIFF1991: language and dialect codes
- add /glossary --> /g redirect?
- add more WAV Formats from MS web page and mmreg.h
- move introduction.md content back to root index and create manual link to it as "Introduction" in sidebar.
- fix broken tables in Safari.
0.1.0 - 2013-01-14 - Initial release, base structure and TODOs
- setup build and deployment
- wrid_data.csv refactor: literal --> condition
- added WAVE Format Categories from {{ spec=(ref"RIFF1991") }}
- better table column wrapping
- introduction: add text and pointer to gh issue tracker for corrections/additions
- add favicon.ico
- add column display filtering for desc and wrid to wrid_table() and wrid_table_subchunks()
- add wrid_table() and wrid_table_subchunks() filtering on "spec"
- decide on data structure changes to include field details from specs. (include them as additional rows with same WRID and different spec value - a new column with spec reference ids.
- glossary: use heading links and custom CSS for links
  - ex: [chunk](/g#chunk) renders as: chunk
- fix shortcodes when no value in spec.see
- added chunk overview page and status info to each chunk
- more intro text, move intro to it's own file, redirect root to intro
- better blockquote via CSS (last quote in series treated as source)
- display release version in the HTML
- add wrid_table_subchunks shortcode, test with LIST-INFO
- sort headings manually, approximately by date spec released
- use spec taxonomy to create backlinks from specs to chunks
- create spec taxonomy: enables full generation of Specification section of chunk entries, and in the future, backlinks from specs to chunks
- create spec_detail shortcode (draft version for now)
- create spec and spec_bib shortcodes, using static spec data
- Setup Zola and prototype initial structure

Future maybe
- categorize chunks by rarity. and/or tag with usage categories?
- fix sidebar (cut off at bottom)
- "Chunk reference" page should show LIST subchunks as well.
- implement spec details widget (shortcodes/spec_detail.html)
- convert chunk reference page to a table
- write chunk reference info
  - id, size data
  - uppercase RIFF, lowercase forms (in theory, but not in practice)
  - etc
- figure out wavref types (things like fourcc)
- create a system for noting that a field overlaps or duplicates data in another field, hopefully automatically create crosslinks between them.
- WRID "sentinel" value mapping, special ints/strings for each field
- create references page, similar setup to glossary page
- add section of useful references? or just generalize specifications --> references?
- add reference link to FADI BWF guidelines: https://www.digitizationguidelines.gov/audio-visual/documents/BWF_Embed_Guideline_v3_2021.pdf
- update site colors: https://iamkate.com/data/12-bit-rainbow/ (see how those look... if not, maybe solarized?)
- wav files: both simple and complex, depending on what you want to do with them.
- wav files: better intro to riff (maybe with a visual?)
- archeology
  - try to find additional docs listed in RIFF1994 Overview
- write intro to what a chunk is on Chunk reference page
- metadata: why do people care about metadata?
- create list of WRID normalized data types and shortcode to link to them
- CSS: scroll wide tables horizontally without affecting overall page
- CSS: fix HTML layout on narrow windows - what to do about TOC?
- question... how deep to dig into fmt: WAV Format Tag inconsistencies between different docs & applications?
- go through each chunk and pull a few spec quotes
- write WAV overview - wav_general.md
- write initial glossary
- write initial doc about metadata
- chunk: add one XML based chunk, probably iXML. Needed before release to verify WRID data model.
- write WRID spec and overview
  - current WRID assumes chunks can only ever have one kind of parent... is this true?
  - how to refer to the chunk itself? Include trailing > to make ids with spaces clearer?
  - field names: use words from spec, but lowered_and_underscores
  - XML chunks? Use simplified XPath? /Path/To/Node[idx]
  - fix table scrolling under nav bar
  - ok for chunks to repeat for WRID, since the goal is to reliably map field names to which chunk they're stored in.
    - question: if this is true... do we actually need the full chunk path? Or can we assume chunk ids are unique? Would shorten WRIDs a lot if we go with that.
  - what to do about sequence numbers (cue, for example)? ignore? we're just trying to get to specific fields in the general sense, not specific data
- update spec/_index to show more info
- article / essay on common parsing errors ("header", fixed position parsing, etc)
  - references
    - https://www.appletonaudio.com/blog/2012/wave-file-format-implementation-errors/
- --- initial release once above complete ---
- article / essay on which chunks actually matter for different use cases
- add footer note to all pages: link to file an issue for any innacuracies
- research: AIFF file type... how similar to WAVE?
- make example WAV file with specific values mapping to WRIDs (or specific ints, etc) to double-check/validate field locations from other software
  - add page describing how to use this to learn what metadata different apps support and specifically where they read/write it.
- create an "SDK" of WRID and descriptive information for embedding in other applications.
- create reference/example WAV files and WRID based metadata dumps
- reference: FADGI Guidelines for Embedding Metadata in Broadcast WAVE Files
- add reference to archive.org verson of sonicspot WAVE overview because it's referenced so often: sonicspot wavefiles
- reference: a nice WAVE overview with formulas for calculating fmt chunk values: The Dr. Agon WAV File Format Page
- chunk iXML: iXML spec
- chunk XMP: XMP specification
- chunk UMID: UMID
- chunk CART: CART website "The Radio Industry's Traffic Data WAVE File Standard. Now formally ratified as AES46-2002!"
- chunk ID3: ID3.org seems to be offline. Alternate spec locations: Mutagen Spec collection, fdd useful references
- update sidebar with vertical line (similar to Zola docs)
- chunks: Audio Definition Model
  - EBU ADM Guidelines Introduction
- add live fuzzy match filtering to the long WRID table
- chunk atem: appears to contain sampler related JSON